« e 1r. w "v varie- hrav'avcvvg‘ LIB-Rx” r " Michigan State: University WHESES 3,4 This is to certify that the l thesis entitled Computer Simulation to Predict Inside Temperature and Heat Production in Swine Housing for Tropical Conditions presented by Irenilza de Alencar N333 has been accepted towards fulfillment of the requirements for Ph.D. . A ric. En . degree in g g WM \. lumnpnmuun Date October 25, 1980 0-7639 M IIIIIIIIIIIIIIIIIIIIIIIIIIII willgxfllllxigfllinstill OVERDUE FINES: 25¢ per day per item RETURNIMS LIBRARY MATERIALS ___________.__———————-— Place in book return to real charge from circulation rec: COMPUTER SIMULATION TO PREDICT INSIDE TEMPERATURE AND HEAT PRODUCTION IN SWINE HOUSING FOR TROPICAL CONDITIONS BY NH Irenilza de Alencar Naas A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Agricultural Engineering 1980 ABSTRACT COMPUTER SIMULATION TO PREDICT INSIDE TEMPERATURE AND HEAT PRODUCTION IN SWINE HOUSING FOR TROPICAL CONDITIONS By Irenilza de Alencar Nags Brazil has a swine herd of 35.5 million head. The southern producing areas of the country has a herd of 18 million head mainly located in the states of Sao Paulo and Santa Catarina. About 11.2 million animals are slaughtered per year to supply the growing pork demand. Hog production is usually on contract with cooperatives. Present swine production facilities in Brazil are based upon animal housing concepts transferred from temperate climate countries. The environmental conditions during the year in those countries differ from the Brazilian conditions. There was a need to develop a mathematical model to predict inside environmental parameters based on natural ventilation adequate for tropical conditions. Combining ventilation, geometry of the buildings and livestock opera- tions for various climate conditions requires a look at the entire system. A computer model was developed to mathematically simulate the inside temperature and total heat production within swine facilities in Brazil. Irenilza de Alencar N333 The model was validated by comparing simulated results to three major swine production centers in the southern states. General results showed the model to perform well for simulation the environment of the production centers. Approved Major Professor Approved Major Professor This thesis is dedicated to my husband Bengt and my son Olof. ACKNOWLEDGEMENTS The author wishes to express gratitude to Dr. Merle L. Esmay for his advice, encouragement and patience in guiding as major professor. Appreciation is extended to CAPES (Coordenacao de Aperfeicoamento de Pessoal do Ensino Superior) for the financial support. Thankful acknowledgement is also due to the understand— ing and assistance of my father and mother, brother and sisters and family in law. Special appreciation is also extended to EMBRAPA (Empresa Brasileira de Pesquisas Agropecuarias) and the Extension Service of Santa Catarina for the research assistantship. iv TABLE OF CONTENTS Chapter LIST OF TABLES .......................... LIST OF FIGURES ......................... INTRODUCTION ............................ I OBJECTIVES .............................. II LITERATURE REVIEW ....................... Environmental Needs of Farm Animals... Air Temperature .................... Humidity ........................... Body Sensible Heat ................. Moisture and Gas Balance ........... Ventilation ........................ Radiation .......................... Effects of Departure from Optimum Condition ........................... Swine Housing Mathematical Model ...... III THEORY of COMPUTER MODELS ............... Weather Simulation Model ............. Temperature Generation and Heat Balance Model ........................ Page IV VI \OCDUWU'IU'l-L‘H l7 l9 IV VI VII VIII GENERAL DESCRIPTION of the SIMULATION MODEL ..................... Input Data Description ............. Output Data Description ............ COLLECTION OF DATA Weather Simulation Model ............. Temperature Generation and Heat Balance Model ........................ Experimental Procedure ............. Methods of Temperature Measurements. EXPERIMENTAL VERIFICATION of the COMPUTER Concordia (SC) ....................... Ribeirao Preto (SP) .................. Aracatuba (SP) ....................... Indaiatuba ........................... Sumare (SP) .......................... Itatiba .............................. DISCUSSION of the RESULTS ................ CONCLUSIONS .............................. RECOMMENDATIONS FOR FUTURE WORK ......... BIBLIOGRAPHY ............................ 55 55 59 61 62 62 69 123 126 vi APPENDICES Appendix A Hourly total heat balance in a swine house (600 animals) in Concordia (SC) ................... Appendix B Temperature generation and heat balance model Subroutine CALOR ................ Appendix C Example of the yearly weather simulation output ................ Appendix D Example of the temperature generation and heat transfer model output for Ribeirao Preto (SP) on January 6 ................ Appendix E Views of the Studied Swine Houses at Concordia, Ribeirao Rao Preto, Aracatuba, Indiaiatuba, Sumare and Itatiba ........................... 135 138 147 151 154 Table 10 11 vii LIST OF TABLES Page Experimental parameters relating inside temperature with the radiant heat load in buildings of different shapes and different roof materials ................. 57 Climatic data for the production areas of Concordia, (Santa Catarina) Ribeirao Preto, Aracatuba, Indaiatuba, Sumare and Itatiba (Sao Paulo) ...................... 66 Structural data on growing-finishing swine production on the selected production areas .................................... 67 More data about swine production on the selected production areas ............... 67 Summary of weather simulated parameters for Concordia (SC) ...................... 79 Wall areas and properties for calculating inside temperature and heat transfer coefficients for Concordia (SC) .......... 80 Building input parameters used in the verification study for Concordia (8C).... 81 Summary of weather simulated parameters for Ribeirao Preto (SP) ................ 89 Wall areas and properties for calculating inside temperature and heat trarsfer coefficients for Ribeirao Preto (SP) ...... 90 Building input parameters used in the verification study for Ribeirao Preto (SP).9l Summary of weather simulated parameters for Aracatuba (SP) ....................... 96 12 13 14 15 16 17 18 19 20 21 22 viii Wall areas and properties for calculating inside temperature and heat transfer coefficients for Aracatuba (SP) ........... 97 Building input parameters used in the verification study for Aracatuba (SP) ..... 98 Summary of weather simulated parameters for Indaiatuba (SP) ...................... 103 Wall areas and properties for calculating inside temperature and heat transfer coefficients for Indaiatuba (SP) ......... 104 Building input parameters used in the verification study for Indaiatuba (SP).... 105 Summary of weather simulated parameters for Sumare (SP) .......................... 110 Wall areas and properties for calculating inside temperature and heat transfer coefficients for Sumare (SP) ............. 111 Building input parameters used in the verification study for Sumare (SP) ........ 112 Summary of weather simulated parameters for Itatiba (SP) ......................... 117 Wall areas and properties for calculating inside temperature and heat transfer coefficients for Itatiba (SP) ............ 118 Building input parameters used in the verification study for Itatiba (SP) ...... 119 Figure 10 11 12 13 ix LIST OF FIGURES Page Values for the total thermoneutral heat production of pigs at different body weights .................................. 12 Causal diagram for swine production ...... 22 Model Identification: determination of resulting environmental range in swine housing in tropical climate ............... 24 Diagram of information flow in the computer program .......................... 25 Flow diagram of weather simulation model . 38 Heat transfer sources considered in modeling typical swine house .............. 43 Flow chart of the swine housing heat transfer computer model .................. 54 Computer model input data ................ 58 Comparison between measured and simulated temperatures for Concordia (SC) .......... 64 Major swine production centers in Brazil.. 65 Hourly outside and inside temperatures at the swine house on January in Concordia (SC) ......................... ‘... 82 Hourly outside and inside temperature at the swine house on June in Concordia (SC) ............................ 83 Monthly measured and simulated temperatures inside the swine house at Concordia (SC)... 85 14 16 17 18 19 20 21 22 23 X Monthly variation of calculated and simulated animal heat production in a swine house in Concordia (SC) ........... Calculated and simulated animal heat production for a swine house in Concordia (SC) ........................... Monthly measured and simulated inside temperature in the swine house at Ribeirao Preto (SP) ....................... Monthly variation of the calculated and simulated animal heat production in a swine house in Ribeirao Preto (SP) ...... Comparison between calculated and simulated total animal heat production in a swine house at Ribeirao Preto (SP) ...................................... Monthly measured and simulated inside temperature in the swine house at Aracatuba (SP) ............................ Monthly calculated and simulated animal heat production in a swine house in Aracatuba (SP) ............................ Calculated and simulated animal heat production in a swine house at Aracatuba (SP) ............................ Monthly measured and simulated temperature inside in the swine house at Indaiatuba (SP) ..................................... Monthly calculated and simulated animal heat production in a swine house in Indaiatuba (SP) ........................... 86 88 92 93 99 100 101 106 107 24 25 26 27 28 29 30 31 32 33 34 35 xi Calculated and simulated animal heat production for a swine house in Indaiatuba (SP) ........................... Monthly measured and simulated temperatures inside in the swine house at Sumare (SP) ............................ Monthly variations of the calculated and simulated animal heat production in a swine house in Sumare (SP) .............. Calculated and simulated animal heat production for a swine house at Sumare (SP) ............................... Monthly measured and simulated temperature inside in a swine house at Itatiba (SP) .............................. Monthly calculated and simulated animal heat production in a swine house in Itatiba (SP) .............................. Real and simulated animal heat production for the swine house in Itatiba (SP) ....... Hourly total heat balance for January in a swine house (600 animals) in Concordia (SC) ............................ Hourly total heat balance for June in a swine house (600 animals) in Concordia (SC) ............................ Side View of the studied house at Concordia (SC) ............................ Interior view of the studied house at Concordia (SC) ............................ End view of the studied swine house at Concordia ................................ 109 113 114 115 120 121 122 136 137 155 155 156 36 37 38 39 40 41 42 43 44 45 46 47 xii Side view of the studied swine house at Ribeirao Preto (SC) .................... End View of the studied swine house at Ribeirao Preto (SP) .................... Side view of the studied house at Aracatuba (SP) ............................ Interior view of the studied swine house at Aracatuba (SP) ........................ Black globe thermometer view in the interior of the studied swine house at Aracatuba (SP) ......................... Side View of the studied house at Indaiatuba (SP) ........................... End view of the studied swine house at Indaiatuba (SP) ........................ Side view of the studied swine house at Sumare (SP) ............................ End view of the studied house at Sumare (SP) .............................. Totally open building studied at Itatiba (SP) ............................. Piglets behavior in the pen interior at Itatiba (SP) .......................... Black globe thermometer view at the open building at Itatiba (SP) ............ 157 157 158 159~ 160 160 161 161 162 162 163 CHAPTER I INTRODUCTION Brazil has a swine herd of 35.5 million head. Approximately 70% is concentrated in the southern states with 30% in the state of Sao Paulo and 40% in Santa Catarina. About 11.2 million animals are slaughtered per year to supply a growing pork demand. In the last few years processed pork meat has become an important export product as well as a popular source of animal protein in the domestic market. Many producers, specially in the southern states, are affiliated with cooperatives that have slaughter houses. The hog production is usually on contract with the cooperative. There is an increasing demand for improved technology in order to increase the swine production through better genetics, nutrition and environmental control. Any environmental control is usually a product of the individual producer since no particular technology or production system has been developed for the Brazilian climatic conditions. Present swine production facilities in Brazil are mainly based upon animal housing concepts transferred from the temperate countries Germany, Denmark, Belgium and the United States. The temperature and humidity ranges during the year in those countries are different from the Brazilian swine producing regions. In the southern states of Brazil the temperature ranges from 20 to 40 Celsius and the humidity from 70 to 90 percent during the winter. The adoption of partially controlled environmental housing for swine production in Brazil started around three years ago. Presently the operational character- istics of the environmentally controlled housing is based on forced ventilation. Knowledge of housing character- istics such as floor area per animal, ventilation rate and type of structural components (roof and walls) can provide better ventilation design. The forced system employs electrically driven fans controlled by thermostats or manual switches. Natural ventilation takes place by thermal convective currents in the air or by wind effect. Forced ventilation has been adopted where it should be possible to achieve good results by natural ventilation. Natural ventilation is achieved by openings, windows and doors according to requirements. One disadvantage of relying on natural ventilation is that it required continous adjustment, with changing weather. The most efficient ventilation system and structural design for a specific climate is necessary. Prototype testing of many livestock housing types for determination of the best combination of ventilation, geometry and construction materials is not only costly but virtually impossible. An alternative is to math- matically model the animal housing parameters for different degrees of environmental control and then simulate various conditions. Heat and moisture transfer through standard livestock structural materials can be calculated from published handbook data, ASHRAE (1978). Climatic data and environmental conditions can then be mathematically modeled to simulate energy balance for livestock buildings. In hot conditions large openings and a natural air movement are required to dissipate heat by convection. Building design should be based upon optimization of structural and environmental characteristics to adequately provide good conditions for swine production. Computer modeling is adaptable to this type of study. A basic swine housing model can be developed to simulate the effect of ventilation rate and construction materials on the environmental conditions within different types of swine housing. Furthermore, by using productivity models already developed the animal performance can be predicted. By programming the costs of various parts of the system, the financial feasibility can be studied. OBJECTIVES The overall objective of this research is to predict the inside environmental conditions inside different types of swine housings under tropical conditions. Within the major objective, the following objectives were pursued: 1. To appropriately adapt weather model simulation to generate weather data for the southern hemisphere. 2. To develop a systems model that will predict the environmental parameters based upon various combinations of climate, structural design criteria and management factors. 3. To validate the model by comparing simulated data to actual data taken at six different swine houses at specific climates in the states of Sao Paulo and Santa Catarina. The computer model input variables are averages and standard deviations of minimum and maximum daily tempe- ratures, relative humidities, and solar coefficients; and the thermal, mass transfer and specific heat properties of structural materials. Model parameters include livestock sensible and latent heat production, livestock number, infiltration rate, and possible ventilation rate, inside relative humidity and temperature . 2. 1. CHAPTER II LITERATURE REVIEW Environmental Needs of Farm Animals Animal houses are designed both to modify adverse weather conditions and to improve labor efficiency. The environment is the total of all external conditions that affect the animals health, per- formance and growth. The environmental factors that may effect animal performance are of three types: physical, social and thermal. The physical factors are those that the animal physicaly contacts, such as; light, sound and space. The social factors are more behavior in nature such as the "pecking" order within the pens. The ambient factors relate to air temperature, relative humidity, air movement and radiation. This particular review pertains primarily to the study of the termal factor within the animal structure. 2.1.1. Air temperature Air temperature is the most important environmental factor that influences an animal's comfort. Temperature differences effect the heat exchanges between the animal and its surroundings as illustrated by McLean (1969), shows that in the ther- moneutral zone the sensible heat loss decreases while the evaporation increases and the total heat production is constant. Over a limited range (the thermoneutral zone) the metabolic heat production of an animal is independent of environmental temperature. Within this neutral zone the metabolic rate is primarily determined by normal body functions and requirements. At environmental temperatures above the thermoneutral zone, heat dissipation is reduced which reduce appetite and the reduced food intake results in lowered heat production. The animal compensates for this reduction by increasing evaporation through the respiratory track. As air temperatures approach body temperature, which is near to 390C for domestic animals reduced sensible heat loss is possible. It is at higher temperatures that high humidity can also become an important factor by limiting the evaporative potential. At environmental temperatures below the thermoneutral zone, sensible heat losses increases beyond the thermoneutral level of heat production. Animals can compensate some for this through reflex actions to increase thermal insulation, such as re- duction of blood flow in the outer body layers and erection of the hair coat. As cold conditions prevail the animal must also increase heat production. To support the greater metabolic activity the animal con- sumes more feed. If additional feed is not available it metabolizes body reserves and loses weight. Both processes are un- desirable. During cold conditions the evaporative losses are minimal. The tem- perature limits of these zones can only be approximately defined. A number of studies have been reported comparing per- formances of farm animals in different environments and widely divergent results have been obtained. Mangold SE a1 studied the effects of air temperature on feed efficiency of growing-finishing swine and found that the feed efficiency of an animal housed at temperature levels of 200C to 250C tend to be higher than at 140C. Pigs raised at temperatures below 90C were found to be less efficient than at 140C. The decrease in efficiency for the light and medium weight pigs is presented by the author to approach to be significant. There is some disagreement in the literature as to the optimum environmental temperature 8 to finish swine. Hanzen gt El (1960) suggest a temperature of about 17°C and Morrison 35 a1 (1968) suggest a tempera- ture of about 20°C. Mangold SE El (1967) attempted to thermo- dynamically model the response of swine to air temperature. Morrison 25 El (1968) developed a semi-theoretical relationship for predicting the rate of gain of swine given temperature and humidity. The main difficulty with the thermodynamic model appears to be in adequately describing the internal processes of the animal that determine the heat energy transferred from the animal to the environment. Nelson 32 El (1970) presented an analysis of swine per- formance under warm environments discussing the cited predictors of swine response and stated that performance predicted curves differ significant by of the field experi- mental results. 2.1.2. Humidity Environmental humidity influences the comfort of animals at high temperature, although it has little effect at low temperatures when evaporation from the animals is 2. l. 3. minimal. Air humidity, by influencing evapor— ative heat loss, particularly from the lungs, affect animal gains when temperatures are above those that favor maximum gains. The effect of this variable on swine at three different temperatures was studied by Morrison SE El (1969) in tests by the California Experiment Station and the United States Department of Agriculture. Results showed that the rate of weight gain and feed con- sumption decreased with increasing humidity up to 80% at temperatures from 220C to 33°C. This rate of gain data were consistent with rates predicted from analysis of the effect of humidity on lung evaporative heat losses presented by Bond £5 £1 (1966). Body Sensible Heat Mammals must maintain their body temperature within the thermoneutral zone limits in order to be productive. This type of heat balance requires the adjustment of their rate of heat loss to their surrounding 2. l. 4. 10 environment. It is shown by Kelly g3 El (1954) that mammals dissipate about one third of the total digestible nutrients of their feed as heat. If they are not able to do so their body temperature rises and they attempt to reach a comfortable heat balance by cutting down on the feed intake resulting on reduction of their rate of weight gain. As shown in Kelly's studies at temper- atures ranging from about 21 to 270C the heat lost by the animal was almost equally divided among the three main modes - convection, radiation and evaporation. Not much heat was lost by conduction. Approximate values for the total thermoneutral heat production in a temperature range of 20 to 25°C is given by McLean (1969) and summarized in Figure 1. Moisture and Gas Balance The maximum rate of air movement near the animal in a house is recommended by McLean (1969) and Carpenter (1974) not to exceed 11 0.2 m/s, and even this level is considered by the authors to be high. For air velocities above that value lung diseases may occur. When the ventilation air exchange is not enough to balance the mositure a buildup of ammonia and odors normally occurs. Bianca 3E 31 (1961) studied the influence of gas concentrations on animal health under housing conditions and estimated that the maximum desirable gas concentrations in the air are of 0.01, 0.05 and 0.25 percent of hydrogen sulphide, ammonia and carbon dioxide respectively. Accumulated excreta and urine on the floor in open drainage channels or pits can also be a major source of gas. In addition to the mositure output from the animal, evaporation from the floor and bedding can make an important contribution to the ventilation air latent heat load. ASHRAE Handbook of Fundamentals (1978) gives data on room latent 12 Pcwgmwmfie pm mafia mo coaposcopm pew: n3: 2: mmmz >pom cma .Acmoqoz Eonmv .muanmS xcon HeapsmCOEmep fleece opp pom mmdam> ”H opsmflm OOH _ o.H o.m (utm/Ieox) uotionpoad ieaH Ieiog 2. l. 5. 13 heat production in a hog house at different temperatures. However it is difficult to give generally applicable figures for the quantity of moisture so produced as it depends on the management system, such as washing the pens twice a day in some regions of Brazil. Ventilation Ventilation systems determine the air temperature and velocity around animals housed in closed structures and thus is an important part in controlling animal heat loss. The role of a ventilation air exchange is to remove excess water vapor or heat when necessary. The performance of the ventilation system can be measured in two ways: one, by physically measuring the environmental factors, or two by monitoring various output criteria such as; pollution, operator welfare and livestock health and productivity. Experiments to determine the effects of the various ventilation systems on temperature and air velocity at 14 stock level were carried out by Carpenter (1974), at the National Institute of Argicultural Engineering, England. The results were shown in curves that help estimate parameters for ventilation system design for open buildings. Most farm buildings in Brazil rely on natural ventilation. A direct measurement of the ventilation air exchange in open buildings is difficult. A useful method of determining the adequacy of a ventilation system is by the con- centration of C02 production divided by the difference of C02 con- centration over that of fresh air. This method was developed by Findlay (1948). The essential requirement of the ventilation system is to remove the water vapor from the building as it is produced by the animals. This must be done while maintaining a suitable temperature. Unfortunately an amount of heat is lost with the 2. l. 6. 15 ventilation air. Pattie (1973) studied the relationships between porosity of structural materials and the ventilation of animal housing. The experiments showed that conduction heat losses are greatly reduced when the ventilation air is filtered through the porous surfaces of the building. A reduction of conduction heat loss was shown when the filtration of air is outward as well as inward. It was observed that perforated surface buildings could be warmer and drier, or cooler and drier, than other building which were other- wise well constructed and insulated, but vapor tight. Radiation Each part of the environment radiates at an intensity depending upon its temperature and emissivity. The net exchange ot thermal radiation between the animal and its sur- rounding depends upon the difference between the two rates of emission as well as upon the shape and 16 relative position of the radiating surfaces. As radiant energy is absorbed it converts to a thermal heat. In studies of heat tolerance of farm animals the importance of the energy reflectance of an animal's hair coat is presented by Kelly g5 El (1954). It is also mentioned that the quality of incident radiation, together with reflectance characteristics of an animal surface, will determine what percentage of the total incident energy will affect the animal. Bond 33 al (1969) determined the radiant heat loads near walls and found the influence of building walls on radiant heat load of nearby animals. For Brazilian conditions where the animal's buildings are not insulated or poorly insulated, the incident solar energy penetrates into the building environment as thermal heat. The concept of combining the radiation aspects with convective 2. l. 7. 17 heat exchange such that the environ- mental heating load reduces to a convection problem, is reviewed by Timmons (1976). The author in- vestigated the directional dependence of sol-air temperatures and quanti- fied the results for the calculation of heat loads. Effects of Departure from Optimum Conditions The optimum environmental conditions varies for each animal production. Beckett (1964) presented a summary of suggestions for the optimum swine environment according to age and body activity. The author's guidance is only with respect to air temper- ature and relative humidity. He states that swine effective temper- atures change when the air temper- ature increases towards the 30°C along with relative humidity over 80%. ASHRAE Handbook of Fundamentals (1978) suggests the optimum environ- mental conditions for hog production is air temperature in a range of 12 to 20°C with relative humidity 18 of 70%. If environmental conditions depart from the optimum, economic production is adversely affected in many ways. As the environmental temperature falls growth of young pigs becomes slow, and with further cold the efficiency of food conversion (meat output per unit of food input) is reduced, as presented by Mangold (1967). High air temperatures also result in reduced rate of weight gain and reduced carcass quality of pigs. Reports of trials at the Hannah Research Institute, Scotland, cite the onset of such adverse effects in the temperature range 25-3000. High temperature may also cause low reproductive performance, as cited by McLean (1969). In- adequate ventilation, leading to high concentration levels of gases, dust and moisture vapor has a direct adverse effect on animal performance and, in addition, promotes conditions favoring the onset and spread of disease. 2. 2. 19 Swine Housing Mathematical Model Past modeling work on heat flow provided a basis for modeling the animal housing environment. The method for calculating heat gain and losses through building sections is presented in detail in the ASHRAE Handbook of Fundamentals, (1978). Designing buildings to create necessary indoor thermal environment generates an extensive array of interacting variables defining building components, materials, orientation, geometry, occupancy and animal comfort requirements. The large number of variables and the complexity of their interaction have induced many simplifying assumptions in the thermal design process, as presented by Buffington (1975). The application of the transmission matrix method with a simplified procedure for its use was developed by Albright g5 31 (1974). The trans- mission matrix method relates the periodic temper- ature and heat flux on one side of a homogenous layer to the periodic temperature and heat flux on the other side of a layer by means of a transmission matrix where the temperatures are expressed in the form of a Fourier series. The convection and radiation heat transfer constants and the steady state conduction equation derived 20 from Fourier's theoretical equation can be used to determine steady state conduction-convection and radiation heat transfer from the structure surfaces. The differential equation developed by Fourier combined with material properties data, predicts conduction heat transfer and heat storage based on temperature differences (Kreith 1966). A periodic Fourier series livestock environment model, including conduction, thermal storage, sensible heat production, solar energy and venti- lation heat transfer effects was programmed and tested by Albright g5 g1 (1974 a, b). Heat transfer mechanisms were required to be linearly related to inside temperature for implementing the Fourier solution. This model uses the sol-air temperature as input requiring the reading of the sol-air thermometers; the work was not based on the outside temperature. However this method of approach using a controlled air volume was the starting point in this research. The effect of solar radiation on heat gains and losses can be modeled by expressing sol-air temperatures for each wall and roof as related to outside temperature, solar radiation and absorption and convection parameters as shown in the ASHRAE Handbook of Fundamentals (1978). 21 CHAPTER III THEORY OF COMPUTER MODELS A major emphasis of this work was to develop an analytical method for evaluating the environment of various tropical swine housing systems, specifically for the main production area of Brazil. The procedure was not complex; however, the information generated from the heat balance involved a considerable amount of calculation which was best performed with the use of a computer. A computer model was formulated to combine two sub- computer models. These sub models were: one, the weather simulation and two the simulation of the temper- ature variation within a building by calculating the total heat balance. The sub models were called on by the main program to perform their particular function, and data created by each model were collected and coordinated by the main program. The weather simulation program was developed by Degelman (1973) previously for conditions in the northern hemisphere. It was appropriately altered to fit the needs of the study for the southern hemisphere. The model for the temperature variations in a swine structure was based on two basic concepts essential to its formulation: simulation of the inside temperature by one, steady periodic analysis and two, total heat balance. Each model is described in some detail to illustrate the 22 equipment management factors swine production (environment) processing industry input of product markets Figure 2 : Causal diagram for swine production. 23 the theory envolved in their development and the way in which they were used. The causal diagram for swine production is illustra- ted in Figure 2. There is interaction between the animal production and its surroundings that may influence the market. In order to describe each model it is neces- sary to identify the system and define the inputs and output of the model. This system identification is illustrated in figure 3. The overall flow of information in the computer program is illustrated in Figure 4. The first step was to call the weather simulation model. It calculated a set of weather conditions for a certain location at a particular hour. The temperature generation and heat balance model was then called to calculate a final environmental condition under the set of weather conditions. These steps were performed each hour for a full year. Hourly outputs were totaled to provide monthly and yearly results. 3.1 Weather Simulation Model The model used to simulate weather data was developed by Degelman (1973). The weather program is capable of simulating hourly weather conditions for any region of the world. Input data required by the program are the location and average monthly weather information. The basic approach to the simulation model development 24 CLIMATIC DATA INPUTS . , OUTPUTS -construction HEAT BALANCE -resultant environ- material ment inside the build~ ——————c~ ~———.» in -livestock number MODEL -this environment meets/or does not meet the standard requirements DESIGN PARAMETERS -livestock heat production -infiltration rate -ventilation rate -environmental needs Figure 3. Model Identification: Determination of resulting environmental range in swine housing in tropical climate. 25 INPUT DATA - START WEATHER SIMULATION MODEL HOURLY INFORMATIONS TEMPERATURE GENERATION AND HEAT BALANCE MODEL ll OUTPUT Resultant environment inside the building Figure 4. Diagram of Information Flow in the Computer Program. 26 was first to decide which variables were deterministic and which were probabilistic. This separates the model into two parts with very distinct purposes and techniques. The two parts are further delineated as follows. The deterministic portion of the model is the part that is directly computable and has a form which is independent of the location on earth. By employing natural laws (through use of equations) where they are appropriate the model is kept unbiased toward any specific area. Therefore, it was determined that following guide- lines would be employed in a deterministic sense: 1. The general shape of a dry bulb temperature for any day follows a consistent trend. Minimum temperature occurs at sunrise and maximum temper- ature occurs at 3:00 p.m. local standard time. It should be emphasized however, that the model will in some cases permit the 3:00 p.m. tempera- ture to be near or even less than the sunrise temperature. Thus, the algorithm which establishes hourly temperatures between these two times merely has to follow a characteristic curve which is predetermined in shape but not in absolute values. The "shape” of the daily temperature curve was stored in the computer program along the curve. The ordinate values were in decimal form specifying the fraction 27 of the distance between the sunrise temperature and the 3:00 p.m. temperature. 2. Another deterministic model is the earth-sun- relationship geometry. This includes the sun's hourly altitude and azimuth angles, time of sun- rise and sunset, equation of time and other earth-sun time dependent variables. The equation for direct normal solar radiation is IDN = A * exp -B/sin8 (W-h/mz) (l) where A Apparent solar radiation at an air mass of zero (W-h/mz) (I) ll Atmospheric coefficient (dimension- less ratio) 8 = Solar altitude angle. The value of 8 is directly computable by use of the earth-sun equations. The value of A in the equation varies from around 10 W-h/m2 in January to 9 W-h/m2 in July, corresponding to respective monthly changes in the solar constant. Its average value is 9.5W—h/m2. This agrees with values published in the ASHRAE Guide (1978). Unlike the ASHRAE Guide, which give specific values for B, values for B are determined stochastically for the weather simulation model. The largest portion of the simulation model is 28 probabilistic in nature. It is based on the premise that weather sequences can occur in a variety of ways under certain limiting constraints. The constraints in this model are basically the items which are to be provided by the user so as not to bias the model toward any specific location. 1. The solar radiation calculation is actually a result of modeling the atmospheric conditions, i.e. determining the opaqueness of the sky - whether they be clouds, haze, smoke, fog or pollution. This determination is made on a daily basis and held constant throughout the day except for intermittent random periods of slight cloud changes. The method of determining the sky's opaqueness is to set a value for B, the atmospheric extinction coefficient, on a statis- tical basis. The process of simulation is basically one of selection of random numbers (one for each day of the month), entering a part of the statistical distribution curve and obtaining a "type" of sky condition for each day of the month. Once the type of sky is established a value of B is computed and then the hour-by-hour computations take place, thus establishing solar radiation values for each hour of the day. The diffuse 29 fraction of solar radiation is also based on the sky's opaqueness and is computed via methods outlined in the references of the study. The the development of this model it was dis- covered that the process of random number selections cannot be permitted to be totally random. It if were, there would potentially be an uncontrolled vacillatin of clear-cloudy- clear-cloudy-clear-cloudy----days which is not representative of trends in weather patterns. Ten years of data for all variables were analyzed to see what a ”normal” period of weather per- sistence is. It was discovered that the mean persistence period was 3.22 days. This is the average period during which solar radiation values high or low. Another way of stating this is that this is the average persistence period for clear weather and also of cloudy weather. This is just the average period, however - random number selections will permit this period to vary from 1 to 6 days in the model. Amazingly enough, the same persistence period turned up for temper- atures, wind, and pressures. It was concluded that this indicates the average endurance period of weather movements over a geographic location. Establishing dry bulb temperatures was mainly a 30 job of determining the right mix of maximum temperatures for all the days throughout a month. If the mix is correct then the month comes out looking like actual weather data. The problems here were to achieve some degree of dependency on the solar pattern and still permit some degree of randomness. While it is simple to statistically determine a distribution of daily maximum temperatures for 30 days in a month - and daily average temperatures for 30 days in a month - it is not so simple to determine which maximum temperature should go with each average temperature. Sometimes where the maximum is high the average is low. Sometimes the reverse is true. Compounding this selection problem is the influence the solar radiation might have on the dry bulb temperatures. In order to find the extent of the interrelationship between the weather variables, correlation and multiple regression analyses were made. Though there are various influences on the way the maximum and average dry bulb temperatures are determined, the process can basically be described as follows. The maximum (or 3:00 p.m.) temperature is determined first (partially random but influenced by the total solar radiation on 31 that day). Second, the difference between the maximum temperature is determined as influenced by the solar radiation. Third, the minimum (sunrise) temperature is determined. The average temperature is then computed and checked against its required statistical distribution. If there is a conflict, the minimum temperature is deter- mined again until the average temperature fits its appropriate distribution statistics. After the maximum and minimum temperatures are fixed the model then places hourly values according to the deterministic model described earlier. Dew point temperatures are determined by first setting the average daily dew point depression. This is influenced by random statistics, the minimum dry bulb temperature and the solar radiation values. Once the depression is deter- mined values are simply subtracted from the dry bulb temperatures for each hour of the day. From this information wet bulb temperature and relative humidity are determined. The periods of oscillation of the barometric pressure curves are much longer than one day and may even be as long as 9 or 10 days. The mean period, however, is about the same as the 32 other weather cycles (around 3 days). The pressures are computed along a sine wave with variations as a result of randomness and solar intensity in the daylight hours. This curve was adjusted experimentally to match existing pressure curves and their means and standard deviations. Wind speeds are erratic but do tend to be opposite to solar radiation values, i.e., when radiation is high wind speed is low. A certain degree of randomness is permitted to influence this, however. Basing the model on a week interval or a month interval seemed to be the most practical. Possibly the weekly statistics would be more accurate, but the monthly statistics were chosen because of two reasons: 1. From the user's viewpoint, entering monthly values is more practical because it limits entry to 12 values (rather than 52 for the weekly case). Climatological summaries are issued on a monthly 'basis and are thus readily available to users. There is one drawback in simulating weather using monthly statistics - that is, that there is the same chance of a certain type of day occurring on the first day of the month as on the last day of the month. The 33 model is designed to help estimate building energy usage, and it does enable prediction of daily and weekly uses of energy. It does not discern, however, whether the day or week is the first of the month or the last of the month. This amount of indeterminancy was considered to be an acceptable compromise in this situation. Therefore, the simulation base of one month was chosen. The simulation of a typical year of weather begins by using statistics for January. The values required for this simulation are as follows: 1. Average Daily Insolation on a horizontal surface. 2. Average dry bulb temperature. 3. Standard deviation of the daily average temperatures. 4. Average of daily maximum dry bulb temperatures. 5. Standard deviation of the daily maximum temperatures. 6. Average dew point temperature. 7. Standard deviation of the daily average dew point temperatures. 8. Average wind speed. Four other factors are required, but not on a monthly basis-these are: latitude, longitude, standard time meridian, and elevation above sea level. To minimize the input, barometric pressure is omitted as well as standard deviations for wind speeds. The 34 average barometric pressure is computed from the elevation above sea level and then varied in a prescribed manner which follows the solar energy trends throughout the year. Standard deviations for wind velocities are considered by Degelman (1973) to be approximately 1/3 of their average. This approximation is supposed to be sufficient when precision in this variable is not crucial to the thermal energy computation in closed buildings. The simulation process begins on January 1 by setting all variables to their average value for January. The hour-by-hour computations proceed through the first day using these average values. The day begins at sunrise and ends at sunrise on the following day. Conditions are actually set one day in advance of the hour-by-hour computations, so that the model knows what points to direct itself toward for the following morning. The first change the model makes (to diverge from the average conditions) is to establish the persistence period for sky conditions. This period (normally from 1 to 6 days) is selected randomly from a normal distri- bution curve. If the sky conditions are on the ”clear side” the upper half to the solar statistics are used to determine the sky condition for each day during this clear period. After the sky condition is determined for one day the "atmospheric extinction coefficient" is derived. There are 31 possible sky conditions where 35 number 1 is the cloudiest, number 16 is the mean value, and 31 is the clearest value. The model is designed to use as many of the 31 values as possible each month without repeating any condition. This assures a full coverage of solar conditions each month and usually forces the monthly average to match the known value. After the solar radiation magnitude for a day has been determined, the maximum dry bulb temperature is established. The maximum dry bulb temperature is established from its own statistics, i.e. mean and standard deviation, but the random number selection is biased in the same direction as the solar radiation. This was incorporated into the model after it was dis- covered that the daily maximum dry bulb is somewhat correlated to intensity of solar radiation. The daily maximum dry bulb temperatures are normally distributed about their mean value, so the selection is based on a random number entered into a normal distribution curve. The normal cumulative distribution curve is stored in the computer program as a list of 31 values. The curve extends from around -2 standard deviations to +2 standard deviations. The 16th value on the curve is the mean value, i.e. zero. A value from 1 to 31 is chosen randomly to determine a position on the cululative distribution curve. Once the value (called the F-value in statistics) is selected it is multiplied by the standard deviation. This 36 determines the number of degrees the temperature is above or below its mean. When this is added to the mean value the maximum temperature for an individual day is determined. The minimum and average dry bulb temperatures are established next, followed by the average, minimum and maximum dew point temperatures. These are all based on individual means and standard deviations similar to the process used to determine maximum dry bulb temperature. In each, however, there are different restraints imposed based on influences of previously determined values - such as, solar radiation and maximum temperature. The next step is to determine barometric pressure conditions and wind speeds. These values are determined by random number selections, again restrained by the solar radiation value. After the daily extremes and averages are set the model performs the hour-by-hour computations of all variables. As this occurs summations are made which ultimately produce means and standard deviations for the "simulated" data. These summaries are printed out at the end of each month along with the known inputs. A brief summary of the simulation sequence will indicate the hierarchy of influence that the weather variables have on each other, these steps are as follows: 1. Set the weather persistence period (in days). 2. Set the sky condition for one day. 37 3. Establish resultant solar accumulation for one day. 4. Establish maximum dry bulb temperature for one day. 5. Establish minimum and average dry bulb temperature for one day. 6. Establish average, minimum, and maximum dew point temperature for one day. 7. Establish average wind speed for one day. 8. Establish average barometric pressure for one day. (can also follow immediately after step 3). 9. Compute hourly values for 24 hours. 10. Accumulate daily summations for all variables. 11. Repeat steps 2 through 10 until persistence period is finished, then repeat step 1. This process goes on continuously hour-by-hour, day- by-day and month-by-month with no break points. The only thing which changes are the statistics by which the daily variables are derived; these change at the beginning of each month. The computer program was modified into a subroutine to generate the weather variables desired as input for the heat balance in the next model. A flow diagram of this program is presented in Figure 5. v 3. {k INITIAL CALCULATIONS] l- CALCULATION OF DAILY CONDITIONS SOLAR 4 WIND b TEMPERATURE I CLOUDINESS l CALCULATION OF HOURLY CONDITIONS PRESSURE l TEMPERATURE I PSYCHROMETRICS 1 SOLAR HOURLY DATA OUTPUT [j TOTAL OF HOURLY CONDITIONS DAILY OUTPUT OUTPUT [j I_TOTAL or DAILY CONDITIONS] MONTHLY ’3‘ OUTPUT O UTPUT TOTAL OF MONTHLY CONDITIONS ] I} OUTPUT Figure 5. FLOW DIAGRAM OF WEATHER SIMULATION MODEL 3. 2. 39 Temperature Generation and Heat Balance Model Optimization of the environment modification system for specific degrees of environmental control with the corresponding effects on livestock pro- duction is an important step towards livestock housing design. Determination of the most efficient combination of ventilation and structural geometry for specific climate and livestock operation is necessary for optimizing environmental system design. To determine the best combination of venti- lation, geometry and livestock operations for various climate conditions is difficult. One method is to mathematically simulate the heat require- ments for different degrees of environmental control in livestock housing. Relationships with various degrees of sophistication that adequately describe heat and moisture transfer exist. Using these relationships with climate data, environmental requirement, construction characteristics and material properties a computer technique was developed to mathematically analyse the existing swine facilities in the southern states of Brazil. Models for predicting livestock environment based on the integration of heat transfer relation- ships have been developed. Prior to the advent of 40 the digital computer, transient or periodic methods of analyzing livestock environments were not employed due to the complexity and number of equations. More recently computer techniques combining the physical laws of heat and mass trans- fer, material and construction types, and the previously cited heat transfer equations have re- sulted in the development of transient or periodic livestock environmental simulations. The simulation provides a more realistic representation of the factors affecting environmental characteristics and their effects on animals. Fourier developed the differential equation which, combined with material properties data predicts conduction heat transfer and heat storage based on temperature differences. These relation- ships are presented by Kreith (1966). An analog computer solution to the problem of steady periodic temperature changes in ventilated buildings was developed by Jordan gE 31 (1968). Albright gg 31 (1974) analysed the steady-periodic building temperature variations in warm weather and tested the final model for livestock buildings in poultry housing at the San Joaquim Valley of California. A number of factors may influence the 41 temperature response of an open structure. The following were considered: outside temperature; mass of air flow through the building; steady state component of heat transfered by conduction through the walls and roof; heat generated within the building; solar energy absorbed by the walls and roof and conducted to the interior; heat radiated from the building surfaces back to the outside environment; steady-periodic component of heat transferred by conduction through the walls and roof; heat storage capacity of the air within the building; and heat lost or gained through the floor. Other phenomena occuring within the control volume such as evaporation or condensation of moisture, artificial heating or cooling and fermentation of litter were neglected. Such effects could be added to the solution; however, only thermal energy was considered in the energy balance. Pressure work and viscous dissipation were neglected. When a structure is considered as a control volume the energy exchanges are shown schematically in Figure 6. The control volume is bounded by the inner surfaces 42 of the walls and roof and the upper surface of the floor. For a control volume we may write energy gained - energy lost = change of energy stored ........................ (1) In terms of the heat flows the energy balance is q0 + qp + qt - qi - qs - qf = MC 3—1 ..(2), T where c = specific heat of air (W/kgOC) M = mass of air inside the building (kg) qf = heat transfered with floor (W) qi = heat taken in with ventilation (W) q0 = heat taken out with ventilation (W) q = sensible heat production within the facilities (W) qS = steady state heat conducted through walls and ceiling (W) qt = steady - periodic heat conducted through walls and ceiling (W) t. = temperature inside (CC) I = time (hr) The objective was to solve equation (2) for ti. To do so required knowledge of the functional dependence of the various terms on time. The terms were considered individually, starting first with models for the tempera- ture variations. 43 ///§‘~///\‘\‘ ~7//\Q///\§///S///S//E~\H/S7/’§/// ll tf heat transferred through floor (W/hr) heat taken out through ventilation (W/hr) heat taken in through ventilation (W/hr) sensible heat production within the building (W/hr) steady state heat conduced through walls and ceiling (W/hr) steady periodic heat conduced through walls and ceiling (W/hr) floor temperature (0C) outside temperature (0C) sol-air temperature (0C) 6. Heat transfer sources considered in modeling typical swine house. 44 The model begins with a set of outside temperatures. Since continuous functions defined over an interval of finite length can be approximated by a Fourier series, it was assumed an outside air temperature approximated mathematically as a Fourier series. Since the thermal energy exchanges were described by linear equations, the system response to an individual term in the series could be tested and by superposition of responses to all terms obtain the total system behavior. However, instead of the more familiar real form of the Fourier series used by Jordan SE 31 (1968), the Euler formula was used to express the Fourier series in the exponential form (Hsu 1970, p. 52). to = To + j = - m Tojexp i j wdT, j#0...(3) where T0 = outside temperature (0C) TOj = outside temperature at a certain harmonic in Fourier series (0C) wd = frequency of daily variations i = imaginary constant j = index indicating harmonics in Fourier series. But the coefficient, TOj was also complex Toj = TOj exp 1W3 ........................ (4) For convenience it was denoted the modulus in equation (4) by Toj’ dropping the brackets. Now writing the expression for the Fourier series describing one day's variation of ambient air temperature, it becomes 45 t = T + o LJ.M8 Tojexp i(jwd + wj) ......... (5) where t0 = temperature outside (0C) Since tO is a real function, the Fourier series describing it is symmetric and may be started at j = 1. To preserve linearity in the equations the solar heating of a building and reradiation to the surroundings was considered in terms of the sol-air temperature. It, too, was expressed by the exponential form of the Fourier series. tS = TS + §= - m Tsjexp l (jwdrj + tj), j = 0 ................................ (6), where TS = sol-air temperature (0C) 5j = phase angle at a certain harmonic (radians). A phase angle, Cj’ is introduced to include the possibility that the sol-air temperature and the air temperature variations were not in phase. The series is expressed in this form for generality but may be computed so that tj = 0. The difference would be asborbed into the co- efficient, T .. Since Tsj is complex, it may be written SJ as TSj = Tsj exp IOj ............................... (7) As was done with the outside air temperature, the brackets are dropped and Tsj will denote the modulus. The sol-air temperature is Tsn = Tsn + g T . exp 1(deT + Q. 46 where Tsn = sol—air temperature for a certain building surface (0C) an = phase angle at a certain building surface (radians). It should be noted that the Fourier series for the air and the sol-air temperatures contain the same fundamental frequency, wd' Also, since the sol-air temperature variation was different for each wall, the individual walls are indexed by n. The inside air temperature may also be expressed as a Fourier series. The terms of the series will be of the same mathematical form as the outside temperature variations, however with a different mean, a different amplitude, and possibly a phase lag, cj. That is, in a linear system a circular forcing function will produce a circular function solution. ti = Ti + g = - ooTijexp 1(deT - cj) , m Tij is complex and may be written Lij exp(i0j). Dropping the brackets around the modulus, ti becomes — 00 t. = T. + Z T 1 i . =1 ij exp ij - cj) + Zsin(cpj - cj). The system responses of interest are the average inside temperatures, the instantaneous deviation of the inside temperature from the average and the phase lag with which the inside temperature follows the outside variations. The flow chart describing the entire process of computing the equations is shown in Figure 7. 54 START READ INPUT DATA COMPUTE SOL-AIR TEMPERATURE AND OUTSIDE AIR TEMPERATURE COMPUTE PARAMETERS FOR HEAT TRANSFER IN THE FLOOR CALCULATE THE HEAT TRANSFER THROUGH WALLS AND ROOF COMPUTE INSIDE TEMPERATURE HOURLY DATA eAOUTPUT) COMPUTE TOTAL HEAT BALANCE Figure 7. Flow chart of the swine housing heat transfer computer model. 3. 3. 55 General Description of the Simulation Model A computer simulation system was developed to analyse the environmental behavior within the swine housing under different types of structures. The model uses the climatic data to simulate daily inside temperature on an hourly basis. Six harmonics of the Fourier series wer used as suggested by Albright (1974). This number of harmonics was required for an adequate representation of the boundary conditions on the building. Since a large number of equations represented the heat transfer balance a computerized solution was most convenient. 3.3.1. Input Data Description Climatic and structural data, environmental parameters and management choices affect the optimal ventilation design of confined swine environment. This computer model required hourly climatic data consisting of average dry bulb temperature and average daily insolation. The thermal and con- struction characteristics including the thermal resistance and heat capacity, area, inside and outside convection-radiation coefficients, depth, thickness, conductivity, specific heat, absorptivity and 56 transmissivity data was supplied for each surface. These values were partially found in the ASHRAE, Fundamentals (1978). A relationship of inside temperature versus the shape of the building was determined and included in the model as a parameter. A relationship between inside temperature and amount of radiant heat in three types of roof was determined and included in the model. These two parameters are Shown in Table 1. Environmental parameters including livestock sensible and latent heat production are found in the ASAE Hand book (1978). The summary of all input data is presented in Figure 8. 57 novmsmnma prcmefinomxm one wcwpopwmcoo enzymnmmEop mummcw .madwnopme moon ucopommwp can mommnm pconommap mo mwcwvawsn cw.pmoa pro: panacea on“ saws onsvmhmafimv opflmcw mewumaon whopuEmnmm Hmucofiwnomxm .H manna u b Hobos one ma popmasoamo Unapmnmanp nuance u ofifi a all m .L I .m m . o o o owmmmaoéuap oweimoéuflt camaooéuai amigéuam oflSOJuam amaooéuam ammaoéuaa m m n .a a .v a r. n56 my Amoon Amoon Amooh “M_w. AEE may movmmnmm ASE Hv Umfimvmcoa maammvnwufi manmwvwcoa AHOOL 90.7.0 or. m... mawv hmHo povmwsnnoo EswcwESHm amaswcmvon hmmswcmvmn amasmcmpop cacao .t _ Hmfipovme moon mo oa>p wcwpawsn mo mamnm 58 @JPUT DATA) AREA OF ALL WALLS (m2) AREA OF FLOOR SURFACE (m2) CONVECTION COEFFICIENT AT FLOOR SURFACE (W/hr - m2 - OC) CONSTANTS ASSOCIATED WITH SENSIBLE HEAT PRODUCTION (W/hr‘- °C - head) DENSITY OF CONSTRUCTION MATERIALS (kg/m HOURLY VALUES OF OUTSIDE TEMPERATURE (0C) MASS or VENTILATED AIR (kg/hr) MASS or CONTROL VOLUME AIR (kg) NUMBER or ANIMALS THERMAL CHARACTERISTICS OF FLOOR AND WALLS (W/hr - m2 - 0C/ m) THICKNESS or WALLS AND FLOOR ( m) 3) Figure 8. Computer model input data 3. 3. 2. 59 Output Data Description The output consist of hourly heat production inside the building according to the given outside climatic conditions. Conservation of mass and energy are the scientific laws providing the basis for the livestock simulation. Utilizing the laws of conservation of mass, mositure removal ventilation requirements were determined. The conservation of energy law combined with environmental, construction materials, climatic simulation and the moisture removal ventilation requirements enabled calculation of the inside temperature, heat transfer sources and quantities, and ventilation. The law of energy conservation indicates: Heat in - Heat out = Heat stored. The literature indicates the sources of heat transfer and they are as follows: (a) livestock and equipment sensible heat pro- duction, (b) solar heat gain, (c) conduction heat transfer, (d) ventilation and infil- tration heat transfer, and (e) thermal storage heat transfer effect. Since there is one independent equation for each unknown, there can be only one solution for hourly 60 ventilation requirements and the inside temperature for a particular livestock building and management system. This solution can be found by finding the inside and outside temperatures and the corres- ponding heat transfer sources which satisfy the energy balance equation. Most of the heat transfer coefficients are known. Also known are the area of the building, number of animals and type of construction material. It was desired to record the inside and outside temperatures hourly and consequently determine the heat transfer values. Volume of the building was determined as by the product of the dimensions in the drawing. The output data consisted in hourly values of inside temperature and hourly animal heat production within the building. 61 CHAPTER IV COLLECTION OF DATA In order to verify the results obtained from the computer modeling procedure, each of the models was validated with actual data. Simulated results were compared to the actual environment of the real situation to form the validation. When the available data was not enough to verify the system model, a check "in loco" was made to identify the possibilities of considering the data representative. 4.1. Weather Simulation Model The weather simulation model has been validated by Degelman (1973) for the northern hemisphere. He validated his model by comparing simulated monthly, daily and hourly data to known weather data. On a monthly basis simulated and known climatic data means and standard deviations of monthly data were compared. This comparison was also available with the output from the program every time it was used. Degelman (1973) validated daily and hourly results as well. Daily extremes and frequency of occurence of simulated conditions were checked against known conditions. For a check on hourly results, a matrix of dry bulb and wet bulb coincidence of Simulated data was compared to that of actual data. Also included was a visual comparison of a plot of hourly 4.2. 62 simulated data to a plot of real data for an entire year. This model was used by Rotz (1977) for generating weather variables to heat transfer models. The validation of the weather simulation model for the southern hemisphere followed the procedure adopted by Degelman. When the program was modified for the southern hemisphere it was compared simulated data with the actual weather condition data of five years in two different locations of the swine production region in southern Brazil. The validation however, was done on daily and monthly basis because of availability of data. The comparison between measured and simulated temperatures is presented in Figure 9. Temperature Generation and Heat Balance Model The validity of the temperature generation and heat balance model was checked by comparing simulated and actual environmental data in swine housing at Brazil. This was done on a daily and monthly basis for a period of a year. 4.2.1. Experimental Procedure The regions where the swine herd is con- centrated in the southern states of Brazil studied. Three major areas were defined as important for pork production (both export and domestic market) from data 63 provided by the Brazilian Ministry of Agriculture. These areas are shown in Figure 10. The selected area 1 included the cities of Concordia, Chapeco, Xanxere and Joacaba. Area 2 included Itatiba, Indaiatuba, Jundiai, Campinas, Braganca Paulista, Sumare and Vinhedo. The cities in area 3 were Ribeirao Preto, Araraquara, Lins, Aracatuba and Novo Herizonte. Six producers were chosen from each of these areas; each one with its own characteristics of swine housing and management. The variations included type of structure, dimensions of buildings, construction materials and degree of confinement. The means and standard deviations of some environmental parameters for the producer locations are summarized in table 2. Table 3 and table 4 provides structural and supplementary data on the performance of the selected swine production centers. Climatic data were collected from each area and from inside the various swine houses in order to better evaluate the performance of the mathematical model. 64 Tempera- _xmlr-___l_- .__1-1_ 1. -1 .1 ture 0C 30 it“ .1 ._- I ._- ___w-_i.e.___-m-4_mi_i 3 8 —’ fiP—r 0" -' d— ‘ «~— ‘ —-—Jv - -— 1r—-- -- -< ‘0‘. -——1A~----—- -< ---» — - - — - T —— 76 SET-AT: ~ .h—L—w -.-- ~--o~» — 4 _——J ...... iv ———->—-o~~—-d> ‘4 5“ 24 --—-——~—-<——-.———.- ._.—._. .— S...___._T_ -. ... — . ._. «._.. -. . .- y .- 1.. -—- -—fluwaou oo oo mnsumuogfimu epauflumq Hencc< aEmu cw: mama xmz Hmncn< I .Aoasmm OMmV wnwumuH m meaam .onnu mwmch .mnsumomu< .ououm omnfiopwm .Amawuumu mucmmv mapuoocoo mo mmoum coauoapoua Ono you name owumawao "N «Heme 67 E moummnmm mumnocoo monoum pmumwsnnoo ow w.o mo mxooHn Ammv mnwumuH moummnmm ouwnocoo peumwnunoo no N.H Oumnocoo pmuuoam mam ocwu mo mxooan Ammv unmanm Oumnocoo eeeeoae mane Reno on o.H exeaee Aemv meeeeameeH moumonmm i mumuocoo pmumwnuuoo om o.H W mnemonoo ” _ m mo mxooan Ammv manumomu< _ 1 w Oumuocoo moummnmm _ n w pouuoam pmumwswuoo om M w.o i mxowwn ououm omuaonwm eumuocoo “ m eeeeoHe ease seae mm m w.o I escape Aomv eweeoueoo a l Ammwm Ham3 m suitcases 38 to .c $5 I Hooam AHmaumumEv cowumawucm> Hmfificm Hmwnmuma mo mama moon mo OQ%H 1 mo mon< Anna meu< composnumcoo coauospona mzfism .moum cowuospoua pmuomaom mzu no maH£MHcHMIwaw3ouw no «use Hmunuoauum "m edema 68 NEW, econ . . HH .xaaeeaueewe Ieeeeemae emcee ma NH Ammo eeaueeH N.OH Omvuooouas mommmmww mwada OH HH Ammv Onwabm m.OH OH nOmQOMHO mwcsq . . . mmunmwa NH «H Ammv ensumHMOcH A.oH e memeeeae emcee m newsman «H ON Ammv manumomn< 144' mi m.oH i N maeoameaeoeaz ON ma Aomv cu m use newsman ououm owuaonwm m.OH M m _ memmmmflm wand OH ON Aomv aflwwoocoo w _ a m. umucfiz Hmaadm 3ounmm Hem mpmuuwa who: ecu mo NV memmmmwp Azounmm ecu mo NV mo nmnadc owmum>< maeanoum moom Hanna 1 numow meum>< .ucmEOQHmGOOTHBOm cw maamwuumm OH cowuoavoua O£H ARV .muounoo cowuoapoua mouoeamm mnu CH coauospoum oaw3m usonm «new who: "e «Heme 4.2. 2. 69 Methods of Temperature Measurements Temperature and heat flow are closely related. Beghart (1971) describes temperature as the potential controlling heat flow. Instruments for measuring one are generally useful for measuring the other. The rate of heat loss by convection must be known for estimating the dry-bulb temperature and the rate of air flow. The cal- culation of a rate of heat loff by radiation involves the temperatures of surfaces or the direct measure- ment of emission by means of suitable instruments. The estimation of the total heat loss and moisture removal through a ventilation system brings in the variable of wet bulb tempera- ture of the air. Small heat-flow meters are available for directly measuring the heat flow through surfaces. Radiometers are available for measuring total solar and sky radiation and the intensivity of radiation from a surface. 70 Instruments such as the globe thermometer measure the combined effects of temperature, radiation, and air flow. Three problems may arise measuring the inside air temperature (dry-blub temperature): (1) location of the instrument, (2) measurement of the dry-bulb without effect of radiation from surrounding surfaces, (3) measurement of rapidly changing temperatures. Bond (1969) recommended that the location of the thermometer be near the animal. However, it is difficult to meet this recommendation without resulting equipment damage. Often a compromise is made by locating the instrument above the animals. The effect of radiation on the dry-bulb temperature as indicated by mercury thermometers is covered by Gebhart (1971) but not quantified. Sudden climatic temperature changes of several degrees in two or three minutes are not usually encountered; thus 71 usually a recording instrument with a bimetallic or Bourdon tube sensitive element will respond fast enough. At an ambient temperature of 240 C a 120 kg hog has near a 300 C surface temperature. The potential controlling convective heat loss is the difference of both temperatures and it varies with the air temperature as the surface temperature depends on the air temperature. Kelly 33 gl (1949) tested various instruments for measuring temper- ature in animal Shelters and the studies of the inside surface temperatures of hog shelters, checked with a radiometer using handbook data for emissivities, have indicated that the thermocouple will give consistent and fairly accurate results. The psychrometric chart provides the means for determining the air characteristics in a heat loss process. The air wet-bulb 72 temperature is important for ventilation calculations at a given barometric pressure. The wet-bulb temperature is important for ventilation calculations at a given barometric pressure. The wet-bulb can be ordinarily measured with a sling or aspirated psychrometer. This is a Simple reliable method but has the disadvantage of requiring an observer at the instrument location. Hygrometers with hair sensing elements may be used for conditions of constant or slow changes in the air humidity. Dust from feed, floor, etc., make the hair elements dirty and somewhat un- reliable. The wet and dry-bulb temperatures describe the influence of relative humidity and heat load on the environment. However, these measurements do not describe the influence of wind and radiant load on the environment inside housing facilities. The black globe 73 thermometer provides a combined measurement on the effects of radiant energy, air temperature and air velocity. These are three impOrtant factors that affect animal comfort. The radiant energy component of the environment is of particular importance. Bond 32 El (1969) showed that, over a temperature range of 60 - 350 C, the body heat-loss by radiation from hogs was 35% of the total heat loss when the temperature of the surrounding surfaces was the same as the ambient air and the air motion was low. According to Bond 35 31 (1969) experiments, at higher environmental temperatures, more of the heat-load transfer was shifted to evaporation, evidenced by a rapid increase in the animal's respiration rate. This apparently is the reason why LeRoy gE El (1976), Morrison 33 El (1968) and Nelson 3; El (1970) who presented performance models of swine in warm environments 74 utilized just the total sensible and latent heat to measure the animal heat-loss. Beckett (1964) cites that clouds tend to reduce the amount of solar radiation and in- crease the long wave radiation. For this reason in warm humid climates where there are generally some clouds during the day, the influence of the total radiant load on building may be different. The black globe thermometer has the advantage of being practical and inexpensive as its particular advantage is its simple construction. Negative radiation can be determined as readily as positive radiation. According to Bond gt_§l (1955), the limitations of the globe thermometer are in the response time which is relatively slow. Thus it must be carefully used where the environmental factors are rapidly changing. It is not accurate where the convection heat is large and it does not account for evaporative 75 heat loss, which is a consideration in animal comfort under warm environments. The range of interest for animal housing is broadly from 00 C to 300 C. Except for research purposes the random and rapid fluctuations in air temperature which occur naturally are of little interest. The globe thermometer's thermal lag, which is a function of its mass and specific heat is not sufficient but it can be ignored for most uses. The time constant of a thermometer determines its response to rapid temperature changes, and it is recommended that this should not be less than 30 seconds for normal meteorological use. An accuracy Gfmeasurement to the nearest 0.50 C is usually sufficient, and this is a common scale division on thermo- meters. The above concept can be applied for the measurement of animal housing temperature with the same order of 76 response and accuracy. Long term trends in air temperature are commonly recorded on thermographs which reduce the available accuracy a little, but are useful instruments. The bi—metallic thermometer can be adjusted to read correctly at, or near, a scale datum mark at the time of checking. All sensing elements must be kept clean. There are various existing types of instruments for measuring temperature and humidity. They range from simple to highly sophisticated. Two limiting factors are the 3 financial resources and the technical help available. Hydrothermographs and thermometers were available for this study as well as globe thermo- meters for reading the radiation and final determination of the total heat balance. 77 CHAPTER V EXPERIMENTAL VERIFICATION OF THE COMPUTER MODEL Verification studies were conducted on commercial swine production structures in the previously cited locations. Inside air temperature were recorded. The inside air temperature was compared with that which the analytical model predicted from the outside ambient conditions and the structural characteristics. The outside input temperature was generated by the weather Inodel. Six harmonics were considered in the Fourier Series for the sol—air temperature variations. All the buildings studies were oriented east-west. Tune buildings studied were for growing and finishing (Df hogs for slaughter. The selected units presented (iifferent degrees of natural ventilation based upon tflne percentage of wall openings. Animal heat production was calculated by the equation Y = 2.477 + 0.034 w - 0.577 T + 0.148 w2 + 0.710 T2 - 0.313 WT (Bond, 1959) Where Y = log heat production (BUT/hr-pig) W = log body mass (lb) temperature/100 (OF) Adequate modifications were made in order to use tihis equation in SI units. With a programmable calculator tZine average body weight was entered in kilogram and 78 converted to pounds. The same procedure was adapted for the temperature. It was given the recorded inside temperature in degrees Celsius and then converted to degrees Fahrenheit. Finally the Y was calculated as the logarithm of the heat production in BTU/hr/pig. The inverse function heat production was determined, and when multiplied by the constant of 6.048 it was given in kilogram calories per day per animal. The summary of the simulated weather parameters were determined for each location studied. The building dimensions and heat transfer parameters are also given. 5.1. Concordia (SC) Table 5 summarizes the weather simulated para- meters which are the input for the temperature generation and heat transfer in the analytical model. Tables 6 and 7 give the building structural and heat transfer data used in the verification study. For the heat transfer data the ASHRAE handbook (1978) was utilized. For the conditions listed in Tables 5, 6 and 7, the hourly outside temperature predicted by the weather model, the calculated outside temperature and the inside temperature determined by the model is plotted for January and June respectively in Figures 11 and 12. The graphs show an agree- ment between the two predicted outside temperature. 79 a; 2:3 «4. H H4: mg .+. mew I o; ..+. 92 own 1 ON ‘ 22. TN“. 3.3 HS.“ «.3. méumau >oz a; .83 Heneéa Toneem oAHe.- poo NJ :3 mi... ..... 0.2 e4 H Emu e; .+. 4...; Ram I e4. 3% méume SAMSON NAM ex: .us< :4. $2 mg .+. :8 m.~ H TS e.~ ..... mg: 33. H4. 2.: Tau we 04“.. TS DAMON 0:3. TN 3% TN H we «A H «.3 TN .+. H: me: «4 32 m.“ .+. RS m; H 0.2. :4 H 93 3994. o.~ 2?. EN H 53 A; H 0.3. «4 H new page: e; 3? ms .+. mu: {ANTS «.mnaem gen 7 o.~ 33 mg .+. 93 «A H «.3 . 04 .+. new :2. 3:5 use. Anaheofi e>< Soc e>< Soc 5...: Sec 25. fieowl. comma acme .eeeeH emaom peace 3mm . nasnusea ease I sue .nomv mwpnmocoo pom mnovoemnmm waywasswm nocuwoz mo hauEESm " n manna 80 (mzlhr) Table 6 : Wall areas and properties for calculating inside temperature and heat transfer coef- ficients for Conc5rdia (SC). Wall N E S £4 Nroof Sroof Area (m2) 61 105 61 105 235 236 Thermal conductance 0.35 0.35 0.35 0.35 u.1 u.1 (KCal/hr.m2.°C) ThICkness 0.20 0.2a 0.20 0.20 0.015 0.015 (m) Thermal ‘ conductivi y 0.u1 0.01 0.u1 0.01 0.29 0.29 (KCal/hr.m .OC) Thermal diffusivity 0.014 0.014 0.0” 0.0“ 0.0146 0.0'45 81 Table 7 : Building input parameters used in the verification study for Concordia (SC). Air mass in the building (kg) 3.96H8 x 103 Specific heat of (KCal/Kg.OC) air and floor 0.2U and 0.30 “Beta (KCal/hr) 151.2 x 103 Gamma (KCal/hr.OC) 315.0 x 102 Outside and inside 0 contact resistance ( C/KCal.hr.m 2) 0.20 and 0.0 Floor density (Kg/m3) 2.2u2 x 103 Floor thermal 2 o 0 9 conductivity (KCal/hr.m . C) ' Floor thermal 3 diffusivity (m2/hr) 0'23 x 10 2 Floor area (m ) 551.7 Floor film '3 coefficient (KCal/hr.m2.oC) .0.1u x 10 Ventilation 17.395 x 103 mass rate (Kg/hr) .Aomv manpmocoo cw >9m:cmw co mmson mcfizm ocp pm monopmpmaEmw mpmeM can opwmpso hahsom "Ha mmzmam boom 2 ma 2 m 2 h. H . . am am 1.. a \X? . o ./o . . G mN mmwhom 9.73:5» \ /_./o %n opopmpmaEmp . 9/ I: VN wcflwyso UoPmHsono o .ImHoI No 2 ./ O/ N \ mm 8 opsumpmm /. ac \\ uEmp mcwmcw cmpmHsEHm x“ /; o Hobos pmcpmmB J414 .\H\. om EOLM mpspmpmaEmu mpfimuzo . .Im/ \ ,\&r 5N dOHO/ 0‘ O \ X/ x .w .o. \_\ x I. wN III A W10 \ x/ A o Okawuo x i 0'0!‘ 1\ 0N / x xii/K a\ .N j \v on f!‘x'r\fi HM F 1,. (30) sanieaadmal .Aomv mwpamocoo cw wash :0 mmso: mewzm opp pm wohsvm9waEmu moflmcw can moflmyso hapaom "MWfi madman hsom 2 ma 2 m 2 ._. _ mmwpom m.poflcsom kn QusumpmaEmu 3 RV wcflmvso pmpmazoamo o mpovmnmm IEmp mnflmcH UmVMHSEwm x HwUOE pmnpmoz Eonm mpswmngEmp mtwmuso o ~1i Ir . o 4..., m /./0/ \\.W\ .0. \ .\ .wflflnx \rlifiwl x f\# xix le x/ x\ x» K x/x/ F\fii .\ XIX/8, ‘7x\£i _ J _ L ma 0H 5H ma ma ow Hm mm mm em mm sanieaadmal (00> 84 For a better understanding of the behavior of the of the inside temperature and animal heat production on a yearly basis the inside temperature and heat production were plotted for a year. Figures 13 and 14 show the comparison between the measured and predicted inside air temperature and animal heat production respectively. The scale was adopted large enough to show eventual disagreement with the curves. With this set of data there are some points from the analytical model curve that do not follow exactly the actual data graph. There may be two explanations for this fact first, the measurement of the inside temperature was made by a hygrothermograph often subjected to dust and second, for the simulated model the number of animal and their weight was assumed to be an average and constant through the year, even though this actually did not happen. During the month of April a new population of piglets was added to the building which reduced the average weight of the animals. Although there were auxiliar thermometers to measure the inside temperature, because of the natural ventilation the heat lost by convection is relatively high. For Figure 14 as the heat production is calculated with the data coming from the weather model, and the weather 85 .numv mappoucou um omso: ocflzm one owwmcfi moeoumhomEou woumflsefim cam venomous xfizucoz " nose: mono oopmfisaam o mumo othmmoE ' O z o m < b 5 Z < 2 ma magmaa m b .m_ X‘ ._.! 1% \ ma P ma ON Hm NN / / , / / z em 7L. mm om 5N H fl IlllmN mm (30) canieaedma; aptsul .flumv machoocou cfl omso; onwzm m ca :ofiuusoopm umo: HmEacm woumaoeflm can wopmasoamo we ceaumfihm> haeucoz gum muswfim choz nH z o m < b b 2 < z m h 86 H I 7 \ meaaaasa ago mo zocmo3ooo Havoc how memo oopmfioofimu x a mumo woumHDEHm o mumfi UQHQHDUHNU . ,/ (801 x temtue/Iegx) uotionpoad ieeq Temtuv 5. 2. 87 model utilizes a long term year compared to a specific year, some disagreement was possible to occur. Again, the variation of the number and size of animals during the experiment was unfortunate. Figure 15 shows the animal heat production versus the inside temperature and there is a certain agreement on the shape curves despite the problems cited before. Ribeirao Preto (SP). Table 8 summarizes the simulated weather parameters which were the input for the temperature calculation and heat transfer data generation from the analytical model. Tables 9 and 10 give the building structural and heat transfer data verification study. For all heat transfer data the ASHRAE Handbook (1978) was utilized. For the conditions dited in tables, the outside temperature predicted by the weather model generated the inside temperature and the animal heat production in the analytical model. The simulated inside temperature and the measured temperature are compared in a yearly basis in Figure 16. The actual simulated animal heat production are compared in Figure 17. Both graphs show a certain degree of similarity between the predicted and the measured parameters. This experiment was slightly affected by the production .numv «appoucou cw om30: ocfizm a how :ofiuuooopo one: Hmswcm poumfloefim one woumHSUHmu Aoov mQSthomEmu wowmcH om mm om ”ma ohsmfim J /./ _/ . woumazufimo pmpmasefim /m7/ .0. / / we ma mo 1/ uemfioz owmho>m pow woumasonu Fulfill—FH— _ 71—w— LO :1” uotionpoad ieeq Iewtuv (SOT x KPP'IPwIUP/IPSX) 89 +| +1 0.0 _ 0000 0.a 0.00 M 0.0 0.00 5.H H 0.05 can a.a W 00H: 0.0 H 5.00 w 5.0 H 0.0a 0.0 H 0.0a >02 ~.~ w H500 5.0 H 0.0a l 0.0 H_0.00 0.0 H 0.0a poo 0.~ w 00am a.a H_0.HH 0.a H 0.00 N.a H 0.55 pamm .1 0.a w 0000 0.a H 0.0a 0.a H 0.0a 0.H H 0.a~ .ms< 5.H . 5500 0.0 H 5.0 0.a H 0.00 5.0 H 0.0a aaae m.a 0550 0.0 H 0.0 0.H H ~J0~ H.H H 0.50 0:30 0.0 . 0000 0.0 H 0.0 .5.0 H 0.55 0.0 H 0.00 >02 0.H 000: a.a H m.aa 0.0 H «.00 0.a H a.a~ aaaa< ~.a W 0000 ~.a H_N.ma 0.0 H_~.00 0.0 H 0.- e000: 0.H 5a5m 0.0 H 0.0H 0.0 H H.0m 5.0 H 0.0m 00a » . a.a 000: 0.0 H 0.0a 0.a H.:.0m 0.H H_:.0~ awe Am\eo m>< .mm< aoov 0>< Aoov xmz Aooo 0>< spec: comma pawz _.omanH amaom fi pcwoo 30a l oasouhpo i case 1 ago .aamv ouwpa omaaoaam Rom whopoempwa omvmasswm panama: mo kHMEESm " m canoe 9O / Table 9 : Wall areas and properties for calculating inside temperature and heat transfer coef- ficients for Ribeirao Preto (SP). Wall N E S ”o W Nroof U Sroof 2 Area (m > 00.0 144.0 00.0 140.0 408.0 408.0 Thermal _ conductance 0.36 0.36 0.36 0.36 2.06 2.06 (KCal/hr.m2.OC) Thickness 0.24 0.24 0.24 0.24 0.018 0.018 (m) i Thermal _ conductivi y 0.41 0.41 0.41 0.41 1.77 1.77 (KCal/hr.m .OC) Thermal diffusivity 0.04 0.04 0.04 0.04 0.046 0.046 (m2/hr) 91 Table 10 : Building input parameters used in the verification study for Ribeirao Preto (SP). Air mass in the building (kg) Specific heat of (KCal/Kg.OC) air and floor .—_ —-_—~ ‘Hfi—H c—--_.——.—o— Beta (KCal/hr) Gamma (KCal/hr.OC) Outside and inside 0 2 contact resistance ( C/KCal.hr.m ) .ww >a- - 5..”- 3.0088 x 103 0.24 and 0.30 151.3 x 103 316.0 x 102 ._nf 0.40 and 0.13 v—vv ”-_.—‘— 3 Floor density (Kg/m3) 2.242 x 10 Floor thermal conductivity (KCal/hr.m2.OC) 0-90 Floor thermal -3 o o o 0 . diffu51v1ty (m /hr) 18 x 10 2 Floor area (m ) 659.8 Floor film -3 ° ° 0. 4 0 coeff1c1ent.(Kcal/hr,m2.°c) - l x l Ventilation 17.341 x 103 mass rate (Kg/hr) .mmmv ouohm ompfionfim pm omso: ocfizm may CH ocsumpmmEou mwflmCW wouwfisefim paw readmmoe xagucoz " 0H madman O 2 Au m < h h z < 2 m b h H coco: 92 mean webmaseam memo oocsmmoe O PO om HN mm mm «N mm om mm mm mm on Q BJUJPJBOWBJ BPISUI (0°) .flamo 0000a omufionfim ca omso: ocfizm m :fi :Oauoopouo umoc Hosacm dogmasefim cam woumHSUHmo ogu mo coMumfiam> kanucoz "ma opzwfim a z o m < b h 2 _< 2 m d ,W J 1 \ Loco: 0.: a.: 93 00000030 000 we zucmmouuo Hmuou how «our webmasuamo x o .l. 3.: mumo voyage? o / \ x3 mumo wmumasufimu . (EDI x KPP'IPWFUP/IPQX) UOIlDBPOJd 1984 IPWIUV 5. 3. 94 system "all in - all out." When the animals leave the building for desinfection it alterates the readings of inside temperature. Figure 18 illustrates the calculated and simulated heat production lines versus the inside temperature. The difference between the curves might be because of the fact cited before that, to simulate the heat production from the inside temperature was a long term year, and the calculated heat production is based on data of a specific year. Aracatuba (SP) Table 11 summarizes the weather simulated parameters which are the input for the inside temperature generation and heat transfer calculation in the analytical model. Tables 12 and 13 give the build- ing input parameters used in the verification study. For these cited building conditions, the outside temperature predicted by the weather model generated the inside temperature and the heat production in the analytical model. Figure 19 shows the measured and simulated inside temperature on a yearly basis. The actual and simulated animal heat production are illustrated in Figure 20. Both graphs demonstrate a similarity between the predicted and the measured parameters. Figure 21 shows the calculated and simulated animal .flamo oocca owpflonflm pm omson ocflzm m CM :oflpusoOha one; Hmsflcm Hmuou omumHSEHm one woumasuamu cmmzuop acmflpmoEou ”ma oeomflm 95 w B m N ow a \T. uh— 71/1 1/ 7 I ompMHSE/W/ ll //,//r [ woumHouHmul 7 // /.V, / 7/ we 00 do 0e0a02 2/ ommco>m how woumfisufimo. ' ./ J N.v v.v m.: (SOT X APP ‘ IPmtue/IEQX) uotionpoad ieeq 19m; 96 :.H +| +| m.H +| m.H Bin omzm Qua 0.0 0500 0.0 H 0.00 0.0 H 5.00 0.0 H 0.00 >02 0.0 050.. 0.0 H 0.00 0.0 H 0.00 0.0 H 0.00 0.00 0.0 5000 ...0H 0.00 0.0 H000 0.0H5.00 0.000 0.0 000.. 0.0 H 0.00 0.0 H 0.00 0.0 H 0.00 .000. 0.0 0000 5.0 H 0.0 0.0 H 0.00 0.0 H 5.00 0000 0.0 5000 0.0 H 0.0 0.0 H 0.00 0.0 H 0.00 050. 0.0 0.3.. 0.0 H 0.00 0.0 H 0.50 0.0 H 0.00. 00: 5.0 0000 0.0H 0.00 0.0H 0.00 0.0H0.00 00.000. 0.0 055.. 0.0 H 0.00 0.0 H 0.00 0.0 H 5.00 090...: 0.0 0000 0.0 H 0.00 0.0 H 0.00 0.0 H 0.00 00.0 0.. o No." II o . 0 II o 0 II o 0 0 0000 + 0 00 5 0 + 0 00 0 0 + 0 :0 000 0005 0% 00.00005 9,... 800 020 800 x0: Soc «>0. €282 00000 0003 .00000 00000 1 00000 300 0000-000 1 0000.: 000 .ammg unovuomn< 00m mpouoswhmm vovmasfiwu honvmos «0 00.0—0.00.5.6 00; manna 97 Table 12: Wall areas and properties for calculating inside temperature and heat transfer coef- ficients for Aragatuba (SP). Wall N E S w Nroof Sroof Area (m2) no.0 luu.o no.0 luu.0 “08.0 #08.0 Thermal (KCal/hr.m2.°C) Th1°kness 0.2u 0.2u 0.2u 0.2u 0.018 0.018 (m) Thermal conductivigy 0.u1 0.u1 o.u1 0.u1 1.77 1.77‘ (KCallhr.m .OC) Thermal diffusivity 0.0u 0.0u 0.0% 0.0% 0.0u6 0.0u6 (02/110) ' 98 Table 13: Building input parameters used in the verification Study for Aracatuba (SP). Air mass in the building (kg) 3.0088 x 103 Specific heat of (KCal/Kg.OC) air and floor 0.20 and 0.30 Beta (KCal/hr) 151.3 x 103 Gamma (KCal/hr.OC) 316.0 x 102 Outside and inside 0 2 contact resistance ( C/KCal.hr.m ) 0.00 and 0.13 Floor density (Kg/m3) 2.202 x 103 Floor thermal 2 o conductivity (KCal/hr.m . C) 0-90 Floor thermal 2 _3 diffusivity (m /hr) 0-13 X 10 Floor area (m2) 659.8 Floor film -3 coefficient (KCal/hr.m2.°C) '0'1“ x 10 . . 3 Ventilation 17.3u1 x 10 mass rate (Kg/hr) 99 , f mpmw wo0003E0m o 0000 00000005 . .flmmv 00500000< 00 00:0: @0030 0:0 :0 00:0000c500 oUHmC0 000003500 0:0 00050005 0000:02 "ma oQDMHm a z o m < b H. z < z m b 00.00: ._.- w 0. 0 \ 1% lflw ’/ o \._\\ 00. ON HN mm mm 0N mN om “N aanieaadmai aptsul (so) 100 0pcoz 00000000 000 mo zucmasuoo 00000 How 0000 @0000SU0mo 0000 000003500 0000 00000300mo .mmmv annumomp< 00 00:00 oc030 m :0 5000006009 00.6; 10500000 0000300000000 03200002 "ON okswfim Q 2 o m € 5 b 2 < Z w b u . ,0 0 01 ._ ._.... AHHUfiWVV 0 .“wmflfluanIllo o \\xw/ \P /< . \ _\ o \o—\ 0 V 50 .//.\0..\ o 0 r (EDI x fiep {emtue/Iegx) uotionpoad ieaq {emtuv 101 .nmmv 0oau0u0p< 00 mmso: ocfizm 0 :0 cofiuus0OHm ~00: H0Efic0 0mu0HSEHm 0:0 0mu0azuH0u Hm mhsmflm noov wpsumpmaEmp wnwmcH om mm ON #7 0TH // all; flW/Zmuflasuifi 7 UmpMHseflm/I . _ _ , 00 mm 00 000003 my mm0po>0 pom 0mu0H2uH0u. (801 x App-Ivmyue/Iegx) uoylonpoad Jeaq 19mguv 5. 4. 102 heat production versus predicted temperature. Here also the long term simulated data were used against data of a specific year. Indaiatuba (SP). Table 14 summarizes the weather simulated parameters which are the input for the inside temperature generation and heat transfer calculations in the analytical model. Table 15 and Table 16 give the building structural and heat transfer input para- meters. For these building conditions, the outside temperature predicted by the weather simulation model generates the inside temperature and the animal heat production in the analytical model. Figure 22 illustrates the real and simulated inside temperature through a year. There is a peak in the registered temperature for the month of April. According to the producer's information the piglet's population increased around 20% during that month in a certain part of the building. This might be the reason for the peak in the inside temperature. Figure 23 shows the simulated and calculated animal heat production during a year period. The dis- agreement between the graphs might be because of the number of animals that actually varies while in the simulation it was considered constant. The variation of the calculated and simulated animal 103 +| +| +| 3.00 .Ammv onsumwmccH pom whoposmuum 0ov0asawm magnum: no anMEESm ”a; manna 0.0 0003 0.0 0.00 0.0 0.00 0.0 000 0.0 0000 3.0.3.022 0.0.3. 0.00 3.0.H040 >oz 0.0 .0000 0.00 0.00 3.00 0.00 3.0.00.3 pom 0.0 3000 3.00.0.3 0.0.“ 0.00 0.0.0300 “2000 l 3.0 3030 0.0..000 0.0.0... 0.00 0000.00 .000 0.0 3000 0.0..” 0.0 0000.30 0.000.: 300 3.0 0000 0.0 H 0.0 0.0 .+. 3.30 0.0 .+. 0.00 0000 0.0 0000 0.0.03.0 0.0.3. 0.00 01000.00 00: 0.0 0000 3.0 .+. 0.3 0.0 0.0.2 3.0 H 0.00 3.000 0.0 0333 0.0 H 0.2 0.0 H 0.00 0.0 H 0.00 00002 0.0 023 0.0.00.2 0.000.000 0.000.000 00m 0.0 0003 0.0 H 0.30. 0.0 H 0.00 0.0 0 0.00 :00 B05 0>0 0,5005 m>0 800 «>0 8% x0: 800 o>0 500: 00000 0003 .0000H 00000 90000 300 0000-000 0000 n 0:0 104 Table L5: Wall areas and properties for calculating inside temperature and heat transfer coef- ficients for Indaiatuba (SP). Wall N E S ' l4 Nroof Sroof Area (03) 32.0 30.0 32.0 30.0 86.16 86.16 Thermal . conductancg o 0.36 0.36 0.36 0.36 6.5 6.5 (KCal/hr.m . C) Th1°kness 0.2u 0.2u 0.2u 0.2u 0.015 0.015 (m) Thermal ‘ . conductivi y 0.u1 o.u1 O.Ml 0.u1 1.18 1.18 (KCal/hr.m .°c> Thermal diffusivity 0.0u 0.0u 0.0u 0.0u 0.0u6 0.0u6. (m2/hr) ' Table 16: Building input parameters used in the verification study for Indaiatuba (SP). Air mass in the building (kg) ".0.6795'x 10 3 Specific heat of (KCal/Kg.OC) air and floor 0.2” and 0.30 IBeta (KCal/hr) 151.2 x 103 Gamma (KCal/hr.OC) 316.0 x 102 Outside and inside 0 contact resistance ( C/KCal.hr.m 2) O.H0 and 0.13 Floor density (Kg/m3) 2.2u2 x 103 :::SECEESEQ:I(KCa1/hr.m2.OC) 0'90 Sigffisi3::3a%m2/hr) 0'18 x 10-3 4 Floor area (m2) 159.9 Eiggfiiiigt (KCal/hr.m2.°C) 'D°1u x 10-3 Ventilation “.185 x 103 mass rate (Kg/hr? 106 .mmwv 0030000030 00 0030; 0:030 0:0 000000 0000: 0... 000300000500 0000H3500 0:0 0003m00E >£cpcoz "Nu 003wwm a z o m < 0 0 2 < z m 0 0. . . ./ \ 0 7 \././ \. /._/ o O P N- o 0000 0000H3E00 o 0000 00030005 . mm O 03 m N sanieaaomal aptsuI H M N I") mm 0m (30) 107 00000000 0:0 00 000003000 00000 000 0000 0000030000 0000030000 000; 005000 000003500 0:0 0000030000 0000002 "mm 003m0m Q 2 o m .< b b 2 0% 2 h .5 0 0. filllklull. .\.I..\ Illa. " .\\\ Willi X / ;llltlllx .11]. W _\A 1||IH\\\ o .I ///w ,//:\\ \\\ .0000 000003500 0000 0000030000 .0000 0000000000 :0 0000; 00030 0 c0 ( OI X 59p ' IPWFUP/IPOX) uquanpoad lpaq {pmguv 8 5. 5 108 heat production versus the inside temperature is presented in Figure 24. Sumare (SP) Table 17 summarizes the simulated weather parameters which were the input for the inside temperature generation and the heat transfer calculations in the analytical model. Table 18 and 19 give the building structural and heat transfer input para- meters for the heat transfer computer program. For these building parameters, the outside temperature predicted by the weather model generates the inside temperature and the animal heat production in the analytical model. Figure 25 illustrates the measured and simulated inside monthly temperature during.ayear period. The animal heat production per month in a year period is graphed in Figure 26. The disagreement between the two graphs might have two reasons first, as the hourly simulated curve depends on the number of animals and this number was not actually maintained constant through the year, and second, the average size of animals in this building was relatively small (around 50 kg) and not all the building was full while in the model was considered with the maximum possible number of animal. The inside temperature maintained relatively constant with the simulation results probably (KCal/animal.day x 103) Animal heat production 109 4; wanna.“ 0,5 ‘3calculated data with 4 average weight of 85 kg V K N. \ 4.4 x‘ \ \ \ \ \ \ . \ S 1 t d 4.3 \\ \\\1mu a e \\\w \\‘ calculated \ § 4.2 \\~H 4.1 “.0 J» It IF Figure 24: 30 35 Inside temperature (0C) Calculated and simulated animal heat production for a swine house in Indaiatuba (SP). 110 .0000 @0530 00.0 0000050000 000003500 00.00003 no 500.5030 3: 0000.0. 0.0 083 0000.30 0.0.00.3. 0.00.0.2 0.00 1 3.0 $03 0.0 0 0.00 0.0400000 0.0 0. 0.00 >02 0.0 82 0.0 0. 0.00 0.0 H 0.00 0.0 0. 0.00 000 3.0 200 90 0. 0.2 3.0 H 0.2. 3.0 0. 0.2 0000 I 0.0 $00 0.0 0. 0.3 0.0 0. 0.00 0.0 0. 0.3 .03. 0.0 $30 0.0 0. 0.0 0.0 0. 0.00 0.0 0. 0.00 003. 0.0 0000 0.00.0.0 0.00. 0.30 0.000.: 003. 0.0 3000 0.0 0. 0.... 0.0 0. 0.00 0.0 0. 0.00 0.0: 0.0 800 0.0 0. 0.00 0.0 0. 0.2 0.0 0. 0.2 00000. 0.0 33 0.0 0. 0.00 0.0 0. 0.00 0.0 0. 0.2 09000 0.0 303 0.00. 0.30 05.0.0.2 0500.2. 000 0.0 023 0.0 0 0.30 0.0 0 0.00 3.0 0. 0.2 000. 0025 0>< Lafihmoxv 0>< Gov 0>< -800 x0: 800 0>< 3.000021 00000 0:03 .00000 00000 00000 300 0030:000 0030 I 000 111 Table 18 : Wall areas and properties for calculating inside temperature and heat transfer coef- ficients for Sumare (SP). Wall N E S W ' Nroof Sroof A < 2) rea m 10 2H 10 2a 183 183 Thermal conductance 0.36 0.36 0.36 0.35 9.9 9.9 (KCal/hr.m2.°C) Thickness 0.2a 0.2” 0.2” 0.2M 0.0a 0.0M (m) Thermal conductivigyo o.u1 o.u1 o.u1 o.u1 2.36 2.35 (KCal/hr.m . C) Thermal diffusivity 0.0” 0.0M 0.0M 0.0% 0.0H6 0.0H6 (m2 Ihr) 112 Table I9 : Building input parameters used in the verification study for Sumare (SP). Air mass in the building (kg) 1.5u08 x 103 specific heat of (KCal/Kg.OC) air and floor 0.28 and 0.30 .Beta (KCal/hr) 151.3 x 103 Gamma (KCal/hr.OC) 316.0 x 102 Outside and inside 0 contact resistance ( C/KCal. hr. m 2) v 0.H0 and 0.13 Floor density (Kg/m3) 2 2u2 x 103 Floor thermal 2 o 0.9 conductivity (KCal/hr. m . C) Floor thermal 2 0.18 x 10-3 diffusivity (m /hr) Floor area'cmz) 359.9 Floor film 2 o ‘0.1u x 10'3 coefficient (KCal/hr. m . C) Ventilation 9 uses x 103 mass rate (Kg/hr? 113 .mmmv 0me3m 00 0m3o: 0cwzm 0:0 0pfimcw m0h3umhooE0u 0000H3§fim 0:0 @033mmma manucoz "mm 093mflm LPCOZ Q” 2 o m < b h z < z u b H 3 ma 1... om o 000 . \\ I, 0 mm o 0000H3Eww o o\. Allw\\\ mump 00333100,: . \r \./ _ V NN NV /._\.// a a .. Z .4 7 K .\ /No// \\... MM mN 0N (30) aanieaaomai aptsul .flmmv mpmesm um 0m3o: mcflZm 0 ca 3030030033 000: Hmeficm omumasewm 0:0 0000H30Hmu 0gp mo :oflumwhm> xflzucoz ”om 003mfim o z o m < a a z < 31 0 .0 30:0: _ .3 _ _ .\/ .//:\\\x\V .V‘ m.m 114 33.03.33 330 mo 30:03:000 H0303 pom mumo o0umfi3ufimo x G N <7 0000H3ewm o 0000 o0umfi3ofimu . 4/l, ( at X KPP'IPWIUP/IPOX) notionpoad lean IPmtuv '1“ - an Mr ". OT 9 O “on V v E 5. 6. 115 because of the radiant load. Figure 27 illustrates the calculated and simulated animal heat production graph versus simulated inside temperature. Itatiba (SP). Table 20 summarizes the weather simulated parameters which were the input data for the inside temperature generation and the heat transfer calculations in the analytical model. Table 21 and 22 show the building structural and heat transfer parameters for the heat transfer model. With these building parameters and the outside temperature predicted by the weather simulation model it can be generated the inside temperature and the animal heat production by the analytical model. Figure 28 illustrates the measured and simulated inside temperatures each month during a year period. Figure shows the animal heat production per month in a year period. The discrepancy between the measured and simulated graphs can be explained as first, not total occupancy of the building and second, as this is a building totally open the radiant load is very low and inside temperature did not change significantly because of the reduced pig population in the studied year. Probably for these reasons both the temperature and animal heat production followed about the same (KCal/anima1.day x 103) Animal heat production 116 Figure 27: ; \ u.s \\ \ \ \ \ 4 4 calculated data\\i ' for average ' weight of 80 kg \\ . simulated ‘ K 4.3 \ k 4 2 4‘ o .\ \\ 4.1 J\ ”.0 \x, calculated 3.9 \\“ 3.8 J L L I 25 30 Inside temperature (0C) Calculated and simulated animal heat production for the swine house at Sumare (SP). 117 3.3 .3003 3.3 + 0.33 3.3 + 0.03 0.3 + 0.33 own 3.3 .3030 3.3 H 3.33 0.3 H 3.03 3.3 H 3.33 >oz 3.3 .3003 0.3 H 3.03 0.3 H 3.33 3.3 H 0.03 poo 0.3 .3303 3.3 H 3.0 0.3 H 3.03 0.3 H 3.03 3000 3.3 .3333 0.3 H 3.0 0.3 H 3.33 3.3 H 3.33 .050 3.3 .3003 3.3 H 3.3 0.3 H 0.33 3.3 H 3.03 0330 3.3 .3033 3.3 H 3.0 3.3 H 0.33 0.3 H 0.03 0:30 0.3 .3003 3.3 H 0.0 0.3 H 3.03 0.3 H 3.03 00: 0.3 .0303 0.3 H 3.0 3.3 H 0.03 3.3 H 0.03 3330< 0.3 .0303 3.3 H 3.33 0.3 H 0.03 3.3 H 3.33 0030: 3.3 .0003 0.3 H 0.33 0.3 H 3.03 3.3 H 0.33 003 7 0.3 0033 0.3 H 0.33 0.3 H 3.03 0.3 H 3.33 :00 Am\ev u>< .mwwamuxv o>< noov o>< 3000 x02 noov o>< :pco: 00030 0:33 .00333 30300 3:3oa 303 3333-030 3333 u 030 how. mnoumsmnwn voumHSEHm nonpmuz mo humessm 3 c«.oanma .Aomv mnwpmvH 118 Table 21: wall areas and properties for calculating inside temperature and heat transfer coef- ficients for Itatiba (SP). Wall N E S w Nroof Sroof Area (m2) 7 27 7 27 12a 12a Thermal conductance 0.36 0.36 0.36 0.36 9.9 9.9 (KCal/hr.m2.°C) Thickness 0.2“ 0.2u 0.2u 0.2u 0.015 0.015 (m) Thermal conductivigy 0.91 0.u1 0.Hl 0.u1 2.36 2.36 (KCal/hr.m .°c> - Thermal diffusivity 0.0a 0.0a 'o.ou 0.0a 0.0u6 0.0u6 (m2/hr) ‘ 119 Table 22: Building input parameters used in the verification study for Itatiba (SP). Air mass in the building (kg) 71.1529 x 10 3 Specific heat of (KCal/Kg.°C) air and floor 0.2” and 0.30 Beta (KCal/hr) Gamma (KCal/hr.OC) 151.2 x 103 315.0 x 102 Outside and inside 0 2 contact resistance ( C/KCal.hr.m ) 0.“ and 0.13 Floor density (Kg/m3) 2.292,x 103 Floor thermal 2 o 0 conductivity (KCal/hr.m . C) '9 Floor thermal -3 diffusivity (mzlhr) 1.86 x 10 Floor area (m2) 239.9 Floor film -3 coefficient (KCal/hr.m2.°C) ;'u9 x 10 Ventilation 6.305 x 103 mass rate (Kg/hr) (0C) Inside temperature 120 33 \ -3..-_ ‘qp~. .. ""1"’_"‘—"‘?“-" ——4p—- 32 o 30 29 \ /‘L"$\o/or 28 o' L— 27 ._. _.3_,_ . ._.____ - measured data 26 9: ~o simulated data ‘\\r/// x measured data with 25 50% occupancy of the building 24 ____ __ ‘_ I I 23 _.___. <1"—‘“” __ _...__ __./ 22 \ 313-33-- 1 2. \\ / 20 . ,// 19 \. // \ -/ ./ 18 *1\ .’/7 17 ._. ./l J 1 0......3..- ..L.__J...._........_...J Month J F M A M J J A S O N D Figure 28: Monthly measured and simulated inside temperature in the swine house at Itatiba (SP). 121 03:02 00303330 030 mo >oc0asooo 30303 303 030v voH0HSU30o x 030p 003033530 0 0300 po303:U30o . .Ammv 0333033 :3 0030: 0:330 0 c3 :033u3003a 30o; 3053:0 003033530 0:0 003033o30o 3323002 ”mm musm3m - o z o m < d 5 z < z m h 3 ,3. fl ./. /. . \.\. 3 .3 / 3 ’0 III V u To m e I w. w.m w .d m m.m D. n a up 0.: o u ) v3 . m 3 v 31 m. u 0 I. N v w TL n.v w. .A x 0.0 .L 0 Cc m.: 122 Auov mpsumpmasmp m0wmcH .Aamv 0033033 30 0030; 0:330 0:3 pom :033u3003a 30m; H053C0 0030H3530 0:0 003033u30u ”om ohsmwu 11.! om mm ON . Lrlw / 003033uawu1/l ./ / /' I A / / 0030.3530/ / » _ x x I, “£0302 >000 0x om pow . 0300 0030HSUH0U If n.n m.m m.m uoyxonpoad 199q {emguv /IPOX) emxue (cor x KPP'I 122 disagreement. The same thought could be applied for Figure 30 that shows the animal heat production versus the simulated inside temperature. 123 CHAPTER VI DISCUSSION OF THE RESULTS Field measurements were made from June 1979 to June 1980 to evaluate the performance of the computer model at six different locations. The heat transfer analytical equations model were based on a set number of animals occupying the building. For any particular case the number of animals in a building was assumed to be a function of the floor area and each producer adopted certain average area per animal. However, through the year the producers allowed the swine population in each building to vary in number and size. These changes affected the thermometer readings. The model predicting inside temperature and animal heat production were based on the total occupancy of the building and on an average body weight of the total number of animals, on a yearly basis for each production center studied. The measured inside temperatures were the readings of a black globe thermometer located at the geometric center of a pen at 1.20 m distant from the floor. The animal heat production was based on Bond's equation as a function of the body weight and average inside temperature. The model predicted and measured temperatures were at variance in some of the studied buildings pro— portional to the producer's variation of animal density. 124 For the studied buildings at Ribeirao Preto and Aracatuba the veterinarian doctors responsible for the health of the swine herd were very careful about recording temperature readings and providing other information about number and size of animals housed. For these two locations the model predicted temperature and calculated animal heat production curves were in close agreement to those measured on-site. In order to compare the computer model predicted data with the on—site observed data for the animal heat production an additional curve was drawn. It was based on measured inside air temperature data and an assumed total occupancy of the building, with a given average animal size used in the analytical model. The inside temperature curves could not be changed as they were observed data. However, the measurements taken at Itatiba were compared to three months measurements taken from May to July in 1978 with 50% occupancy of the building. In this particular situation the low number of animals in the open building caused lower inside temperature readings. There is an agreement between the curves when the predicted heat production graph is compared with the calculated heat production graph assuming total occupancy of the building at the observed inside air temperature. From this similarity it can be concluded that the analytical model performed well for the various degrees 125 of natural ventilation in terms of percentage of wall openings. Air movement through an open building is caused by wind or thermal forces, acting alone or together. The flow due to thermal force is considered when there is no significant building interval resistance, and indoor and outdoor temperatures are close to 25°C. The equation given by the ASHRAE Handbook of Fundamentals (1978) shows that the flow due to stack effect is proportional to the area of openings, the height from inlets to outlets, and the difference between the inside and outside temperatures. In the model it was considered only the wind effect associated to a constant. However, at times when the wind flow may be close to zero values the stack effect could be considered. The value of the model is that it will predict the inside environment of swine houses based on the weather parameters and structural and management characteristics. These input parameters can be changed in order to optimize the environment for the swine houses and to better accomplish housing design for naturally ventilated buildings. 126 CHAPTER VII CONCLUSIONS Simulations using the mathematical model can be used in the design of agricultural structures for animal confinement. They can also be used to evaluate the efficiencies of existing facilities, specially the pre- diction of the inside temperature and the efficiency of the ventilation system in order to maintain or modify this temperature. 1. The weather model simulation was appropriately adapted by modifying the earth-sun angles, to generate weather data for the southern hemisphere. The data obtained with this model was utilized as input variables for the temperature generation and heat transfer model. A model was developed to predict the inside temperature. The animal heat production and heat losses based upon building structural characteristics and number of animals can be changed in order to optimize the production. The computer program calculates the sol-air temperatures for each wall and roof of the building and calculates the steady state and periodic heat losses per wall per hour. The model was validated for six locations in the southern states of Brazil. Each production 127 choosen has its own management characteristics and building design. The simulated data was checked with measured data by plotting the predicted values of temperature and heat production, and the measured temperature data and calculated heat production data. 128 CHAPTER VIII RECOMMENDATIONS FOR FUTURE WORK The results of this research suggest additional work as follows: 1. Measurement of air temperatures at different heights in the building in order to better describe the inside environment. Measurement of surface temperatures of roof and wall materials to further verify the model predicted sol-air temperatures. The output generated by the heat transfer model could be considered as input of a swine performance model to optimize the animal's growth and feed efficiency. The daily environmental conditions for the months of extreme temperatures should be studied further for design of swine housing. Determine the producer's cost and return benefits from modifications on the environment through changes in the housing structural design and management. BIBLIOGRAPHY 129 BIBLIOGRAPHY Albright, L.D. and N.R. Scott. 1974. An analysis of steady state periodic building temperature variations in warm weather - Part I: A mathematical model. Transactions of the ASAE l7 (1): 88-92, 98. Albright, L.D. and N.R. Scott. 1974, An analysis of steady state periodic building temperature variations in warm weather - Part II: Experimental verification and simulation. Transactions of the ASAE 17 (1): 93-98 Albright, L.D. and N.R. Scott. 1977. Diurnal Tempera- ture fluctuactions in multi-airspaced buildings. Transactions of the ASAE 20 (2): 319-326. ASAE, St. Joseph, MI. Anuario Estatistico. Boletim Estatistico. 1978. Vol. 36 (141). IBGE. Instituto Brasileiro de Geografia e Estatistica. Castelo. Rio de Janeiro, Brazil. ASAE data D270. 4. 1975. Design of ventilation systems for poultry and livestock shelters. Agricultural ‘ Engineers Yearbook. ASAE, St. Joseph, MI. 49085 ASHRAE. 1978. Handbook of Fundamentals. American Society of Heating and Refrigeration and Air Con- ditioning Engineers. Inc. New York. Beckett, F.E. 1964. Effective temperature for evaluating or designing hog environments. ASAE paper No. 64-911. ASAE, St. Joseph, MI. Beckett, F.E. and F.R. Barron. 1971. Heat transfer from a pig to the floor. Transactions of the ASAE l4 (4): 700-703. Bianca, W. and K.L. Blaxter. 1961. The influence of the environment on animal production and health under housing conditions. Festschrift zum VIII. Intern. Tierzucht - Kongress in Hamburg. pp 113-147. 130 Bond, T. E., C.F. Kelly and H. Heitman. 1952. Heat and moisture loss from swine. Agricultural Engineering 33 (3): 148-152, 154. Journal of ASAE, St. Joseph, MI. Bond, T.E., C.F. Kelly and N.R. Ittner. 1954. Radiation Studies of painted shade materials. Agricultural Engineering 35: 389-392. Bond, T.E. and C.F. Kelly. 1955. The globe thermometer in Agricultural Research. Agricultural Engineering. April, 1955: 251-260 Bond, T.E. 1959. Environmental studies with swine. Agricultural Engineering 40 (9): 544-549. Bond, T.E., C.F. Kelly and H. Heitman. 1959. Hog house air conditioning and ventilation data. Transactions of the ASAE 2 (1): 1-4. Bond, T.E., C.F. Kelly, W.N. Garret and L. Hahn. 1961. Evaluation of materials for livestock shades applicable to other open-type structures. California Agriculture Journal, July, 1961. Bond, T.E., H. Heitman and C.F. Kelly. 1965. Effects of increased air velocities on heat and moisture loss and growth of swine. Transaction of the ASAE 8 (2): 167-169, 174. Bond, T.E., S.R. Morrison and H. Heitman. 1966. Skin and lung moisture loss from swine. Transactions of the ASAE 10 (5): 691-692 and 696. Bond, T.E., S.R. Morrison and R.L. Givens. 1969. Influence of surroundings on radiant heat load of animals. Transactions of the ASAE 12 (2): 246-248. ASAE, St. Joseph, MI. Buffington, D.E. 1975. Simulation models of transient energy requirements for heating and cooling buildings. ASAE paper no. 75-4522. ASAE, St. Joseph, MI. Carpenter, G.A. 1974. Ventilation of buildings for intensively housed livestock. Chapter 19 of "Heat loss from animal and man". Edited by J.L. Monteith. Butterworths. Carslaw, H.S. and J.C. Jaeger. 1959. Conduction of heat in solids. Oxford University Press. 131 Christianson, L.L. and M.A. Hellickson. 1977. Optimization of energy requirement for animal housing. Transactions of the ASAE 20 (2): 327-335, ASAE, St. Joseph, MI 49085 Collins, N. E. and E. W. Walpole. 1976. Computer simulation of commercial breeder stock operation. ASAE paper No. 76-4040. ASAE. Degelman, L.O. 1973, A weather simulation model for building energy analysis. Department of Architectural Engineering, The Pennsylvania State University, University Park, PA. Esmay, M. 1969. Principles of animal environment. The AVI Publication Company. FAO Yearbook. 1978. Food and Agricultural Organization of the United Nations. Rome, Italy. ndlay, J.D. 1948. Experiments on ventilation of cattle byres. Journal of Agricultural Science. Vol. 38:411-424. Gebhart, B. 1971. Heat Transfer. McGraw Hill Book Co., Inc. Grabeau, K.W. 1978. Precision temperature analysis. E.G. .G. - Environmental Equipment Division Temperature Measurement Handbook. Omega Eng. Inc. Hanzen, T.E. and D.W. Mangold. 1960. Functional and basic requirement of swine housing. Agricultural Engineering 41 (9): 585-590. ASAE, St. Joseph, MI. Harman, D.J., A.C. Dale and H.W. Jones. 1966. Effect of floor type on required moisture-vapor removal rate from swine finishing houses. ASAE paper No. 66-442. ASAE, St. Joseph, MI. Heitman, H., T.E. Bond and C.F. Kelly. 1963. The zone of thermoneutrality in swine on full feed. Proceedings of the Third International Biometeorological Congress. Pergamon Press. Oxford. Hellickson, M.A., H.G. Young and W.B. Witmer. 1974. Ventilation design for closed beef buildings. Proceedings of the International Environment Symposium. ASAE, St. Joseph, MI. 132 Hsu, Hwei P. 1970. Fourier analysis. Simon and Schuster, New York. Jedele, D.G. 1966. Exposure factor curves applied to swine building ventilation design. ASAE paper No. 66-443. ASAE, St. Joseph, MI. Jordan, K.A., A.J. Barwick and J.A. DeShazer. 1968. Frequency analysis of weather for estimation of temperature in animal shelters. Transactions of the ASAE 11 (l):98-103. Kelly, C.F. 1949. Instrumentation for animal shelter research. Agricultural Engineering. Vol. 30, No. 6: pp. 297-304. Kelly, C F., T.E. Bond and N.R. Ittner. 1950. Thermal design of livestock shades. Agricultural Engineering ‘ 31 (12):601-606. ;ASAE, St. Joseph, MI. Kelly, C.F., T.E. Bond and H. Heitman. 1954. The role of radiation in animal ecology. Ecology Journal, Vol. 35, No. 4, October 1954. Kreith, F. 1966. Principles of heat transfer. Inter- national Textbook Co., Scranton, PA. LeRoy, G. H. and J.A. Nienaber. 1976. Summer weather variability and livestock production. ASAE paper No. 76-1033. ASAE, St. Joseph, MI 49085. Mackey, C.O. and L.T. Wright. 1946. The sol-air thermo- meter. ASHVE Transactions, Vol. 52, p. 271. Mangold, D.W., T.E. Hazen and V.W. Hays. 1967. Effect of air temperature on performance of growing-finishing swine. Transactions of the ASAE 10 (3): 370-375 and 377. McLean, J.A. 1969. The environmental needs of farm animals and their output. Journal of J.I.H.V.E. Vol. 37, December 1969. Hannah Dairy Institute. Ayr. England. Midwest Plan Service. 1977. Structures and Environment Handbook. MPWS-l, Iowa State University, Ames, IA. Morrison, S.R., T.E. Bond and H. Heitman Jr. 1968. Effect of humidity on swine at high temperature. Transactions of the ASAE, 11 (4):526-528. ASAE, St. Joseph, MI. 133 Morrison, S.T., L. Hahn and T.E. Bond. 1970. Predicting summer production losses for swine. Production Research Report No. 118. U.S. Department of Argiculture. Agricultural Research Service. Muehling, A.J. 1976. Automatic control of natural ventilation for hog confinement. ASAE No. 76-4538. ASAE, St. Joseph, MI 49085. Nelson, D P., C.H. Read, B.J. Barfield, J.N. Walker, V. Hayes and G. Cromwell. 1970. The performance of swine under warm environments. ASAE paper No. 70- 442. ASAE, St. Joseph, MI. Neubauer, L.D. and R.D. Cramer. 1965. Effect of shape of the building on interior air temperature. Transactions of the ASAE ll (1) 537-539. Pattie, D.R. 1973. Ventilation by diffusion and filtration. Proceedings of the fourth Canadian Congress of Applied Mechanics. University of Guelph. Guelph, Canada. Randall, J.M. 1975. The prediction of airflow patterns in livestock buildings. Journal of Agricultural Engineering Research, Vol. (20):199-215. Randall, J.M. 1975. Where the air moves in your buildings. Farm Bldg. Dept. NIAE. Bedford. England. Rotz, A.C. 1977. Ph.D. Thesis research. Computer simulation to predict energy use and system cost for greenhouse environmental control. Department of ,Agricultural Engineering. The Pennsylvania State University,PA. Schuler, R.T. and D. Broten. 1976. Computer simulation of animal ventilation. ASAE paper No. 76-4023. ASAE, St. Joseph, MI. Schulte, D.D. et a1. 1968. Effect of slotted floor on air-flow characteristics in a model swine confinement building. ASAE paper No. 68-945. ASAE, St. Joseph, MI. Systems Engineering in Agriculture. 1965. Symposium proceedings. ASAE publication PROC-567. ASAE, St. Joseph, MI 49085. 134 Teter, N.C., J.A. DeShazer and T.L. Thompson. 1973. Operational characteristics of meat animals. Part I - Swine. Transactions of the ASAE 16 (1):157-159. Teter, N.C., J.A. DeShazer and T.L. Thompson. 1976. Animal performance models. ASAE paper No. 76-5013 ASAE, St. Joseph, MI. Timmons, M.B., L.D. Albright and R.B. Furry. 1976. Distortion effects in using building models to predict time dependent thermal behavior. ASAE paper No. 76-4024. ASAE, St. Joseph, MI. Timmons, M.B. and L.D. Albright. 1976. Wind directional dependence of sol-air temperatures. ASAE paper No. 76-4533. ASAE, St. Joseph, MI. Wilson, J.D. 1970. Determining transient heat transfer effects in structures with the use of a digital computer. Transactions of the ASAE 15 (4): 726-728, 731. APPENDICES 135 Appendix A Hourly total heat balance in a swine house (600 animals) in Concordia (SC) for January and June 136 .Aomv mflopmocoo :3 “0305300 0003 0030: 0:330 0 ca mhmscmb pom moc0H0n 300: H0303 >Hpsom " Hm whswam 950m 2 ma 2 m 2 JIJIJJj o L ,: c; OH (EDI x IRON) aouvtea lPBH I910; 137 ocwzm Loom .Aomc mfinpmocoo c3 30305320 cows 00:0: 0 CM mash pom mundamo 300: H0303 mapsom ” 2 ma 2 mm 032030 z > < < /x O H (801 x 193x) BOUPIPQ lean {910$ 138 Appendix B Temperature Generation and Heat Balance Model Subroutine CALOR C II. 1112 c 0.0 102 c on. 331 103 C no. 603 1:39 UURUUTN‘ AU!“ 1‘! 5 HI x NM DIMENSION STO(21195T5(9.20),STI(24)aTO(9)oTS(909)pTI(9)a570(9)! ~9TU(9)0ATS(90910375(999105ATI(9)0571(9)0l516)0UH(9)9LW(610K"(6)O .ALPHANlb),0WEGA(9,9)9P51(9),FREO(9),PHIEP(9).AVBT5(°)OATI(9)' in!”9),Ah(9),KUUHTC?1),SUH(21),SSTO(24)¢THETA(9.24).XH(9,9), 50P(291001(24).00(211005(24)¢0F(29)00T(9p2‘):DUM(24)oXJ(9p9), 550b1(6324) REAL KF,NF.Ld,KH cunPLLx-A,u.c.n,xu.xa,Anc,sF.stN,ucos,csxp THOPI=6.?831653 pxrouuao.7asa9az N0. PTS. IN SERIES'NO. HARMONICS. hO. BLDG HALLS DATA KpLoNH DATA A“ DATA LN DATA ALPHAH DATA KN DO 1112 IL!=1.6 ALPHAA(IL1)=O;S SOLAR INCIDENCY 0N HALLS 1 AND 3 DO 102 181020 SOL1(1.I)=SOL 5001(3oI)¥SOL SOLAR INCIDENC! ON HALL AT SUNSET 1430.5 00 331 181.16 5001(4.1)=X4¢$OL X48X490.1 x481.1 no 10) 1:17.24 50b1(4¢1)=x4550L 14814-0.2 SOLAR INCIDENC! on "ALL AT SUNRISE x2:o.5 00 603 13106 5001(2ol)=XZOSOL x2=X290.1 604 605 636 C 222) 3333 0444 5555 C 00! 1:2 113 114 115 111 C or. D)) 140 x2=1.1 03 004 1:7,!) 5901(2o1)=x2-300 x2=x2ou.3 12=1.9 ‘ DO 605 1:11.17 5001(2o1)=x20500 x2=x2-o.n 50L1(2.18)=30L X2=3.9 DU 606 1:19.24 SOL1(2'1)=X2'SJL 12:12-0.1 SOLAR xx=0.s DO 2722 131.6 5001(5o1)=AX¢SJL SULIIO.I)=XX¢SJL XA=XKOU.1 XIV-'81. DU 3))! 187.12 5001(5.1)=AX¢50L SULI(6.1)=XK‘5JL xxaxx+u.2 XA=2. DO 1414 1:13.10 sunl(5.1)=xxcsab SUL1(b.I)=XA‘5JL xxsxx-o.2 xx=u.9 DU 5555 1:19.21 SUL1(SoI)=Xx¢53b SULI(6.1)=XX*SOL Xklxx-U.1 I'dCIDE-‘lCY O’J ROOF SOL-AIR TEHPS FOR EACH HALL IN TURN DO 111 181.u4 no 112 J=l.lnn 5T5(l.J)=STU(J)+0.lSO(SUL1(IoJ)/(AN(1)‘KW([J.L4(I))) DU 113 J=IHH91.12 sr5(l,J)=STU(J)+(0.15¢(J-IHN)no.025)r(5nL1(1.J)/(AN(I)'Kfi(l)*LN(1 ) DO 114 J823.18 STStl.J)=bTU(J)+(0.3-(J-12)-o.025)utsutl(I.J)/(Ax(1).Kw(I)uLw(I))) DU 115 J=18o2+ ' 515(I.J)=STU(J)+0.150(SUL1(10J)/(A~(I)iKN(I)iLfi(I))) CONTIJUE OTHERS VARIABLES AINHAS: CPAIR: BETA: 79 C 009 97 C on. 90 C on. 95 c 6.. 991 62 141 Git"."\= Ru: #1: CPF: ROMP: KF: HF: AF: ALPHAF= uhu=lhr(h.09-AP) VLNTM:bO.6NAd- CJJPUTH NEAL CJEFFICIUHTS 0F FOUHIEH SEVIES 'FUH UUI'SIIHJ AU TEMPEPATUHE VARIATIUU Chbb IUURIE(STO.K.L.AVETD.ATO.BTU) COMPUTE DELL COEFFICIENTS UP FOURIEN SERIES FUR SUD-AIR TEWPERATURE VARIATION DU 79 J=1.Nu DJ 7! 1:1.K STI(1)=5TS(J.I) DU 7" 1:191; AT5(J.I)=TI(I) bTS(J.l):TU(I) CQHPUTE CUWPLKX COEFFICIEJTS FOR FUURIRR SERIES OF UUTSIHK RIR TEMPERATURE VARIATIOH DO 97 J=1oh TU(J)=((ATutJ)IATU(J)+(UTU(J)'DTO(J)))0'0.5)/2. P51(J)=AT§N2(-0T0(J).hTO(J)) CUHPUTB CU“PLEA COEFFICIFJTS FOR FOURIEH SERIES FUN SOL-AIR TEMPERATURE VARIATION DO 9b J:1.L DU 90 1:1.nu 15(1.J)=((A75(x,a).nT3(1.J)+(HT5(I.J)-RTS(I.J)))'*0.5)/2- OMEGA(I.J):ATAJZ(-BTS(I.J).ATS(1.0)) COMPUTE FREQUEJCY FOR HARHUNICS no 95 J=1.L x=J Y8K FREU(J)=(TWOPI‘X)IY COMPUTE PARAMETERS FOR MEAT TRANSFER IN THE FLOOR DO 9‘ Jgigb IF(HF.GT.1.0E'10) GO TO 62 89(J)8P1FUUR AA(J)80.0 Cu 10 9% BS'SORT(((((KFIFREQ(J))0BH0F)DCPP)/(2.*(HFIHF)))) 68(J):ATANZ(EE.EE+1.0) 94 lli2 AA(J)=SURT(1./(1.9(2.*EC)+(2.fi(EEuEt)))J Continua C IQOCOHPUTS PARAHETERS FOR TRANSIENT HEAT 1RANSFER C 990 93 C CO. 91 97 To . THROUGH THE uALLs Do 9) J81.L DU 9) ISIOHd . ANG=(50HT(FREU(J)I(2.lALPHAN(I)))uLn(I))fi(1.0.1.0) £€:cbxp(nnc) “511:0.55(EP-1./EF) "60580.50(EF+1./Er) ABtICUS DauCUS EF8(Kd(I)oARG)ILw(I) bl(-H$lu)/EF Ct(-£F)OHSIN EFIS-Abko-uouloco(fluonl) XH(IoJ)=(C00-ADD)/EF XJ(1.J)=(D-C¢RU)IEF COMPUTE NAGNITUDE AND PHASE ANGLE of THE RESPONSE no 92 J:1.L 1:0. 1:0. 0:0, 2:0. ALG'(SUPT(PREU(J)/(1.0ALPHAF)))l(((AFDKF)§AA(J))§1.)O(1.0.0.0) no 91 1:1.nd X=1RW(1)lTS(I,J))'((RCAL(XU(13J))‘CUS(O“ECA(I.J)))'(AI“AC(XH(19J)) u'Sln(OMEGA(IoJ))))9X Y=(A~(I)ITS(IoJ))'((REAL(XH(IoJ))'SIN(O”ECA(IoJ)))9(AIMAG(XH(10J)) .oc05(UNEGA(I.J))J)OI Had-rhu(1)9n£Ah(XJ(1.J))) zs2o(nu(l)°AIW«G(XJ(1.J))) 030+(REAL(ARG)'COStPlFUUH-Mstd)))+(VEHPW‘CPAIR)+GAHMA xsxoc((veurntcphxn)*TJ(J))*C05(Psr(o))) YSX¢(((VEHTM-COAIR)'T0(J))*51H(PSI(a))) Z:Z-(REHL(ARG)051N(PIPUUR-HD(J)))- -(#1RMAS.(CPAIR!FREQ(J))) anLP(J)-\T.\u2((I'M¢XtZ). (x.I-¥§Z)) rltJ)ax/((unc03(PuIBP(J)))+(zrslu(PhIEP(J)))) A~U=0. Aa0t5=0. DO 70 1:1.Nu Ud(l)=Ku(I)/Lu(l) DUM(1)8AN(I)OUd(1) AMJBAMHUUM(1) Adur53Auurso(DUH(1)~AVBT5(I)) AdUCAUUQGhfinhotVEfl710CPAIR) AdUTSBAHUT59((VENTRICPAIR).AVETO)98kTA AVETISAMUTS/Adfl DJ 75 J31 '11 first!(J).CHXP(PHIEP(J)¢(0.0.1.0)) ATICJ132.'NEAL(SF) brI(J)8-2.OAIHAG(BP) CCC=K Do 10 [21.x KOUNT(I)-l er(I)=AVETI ssrotx)-Averu 74 .0. CI. .G. .0. Of} (1 (1 555 606 607 333 c on. c 000 C .00 C .0. 015 556 660 778 089 999 c on. 73 72 71 143 UOH31 DU 19 J31,“ AAA=J THETA(J.173TJUPI.(HHHIAAA)/CCC 551')!I1=SSTO(I)OATO(J)‘COSCTHETA1J.I))08Tn1J)*SIN(THETA(JoI)) 5111l)IST11I)¢uTX(J)*COS(THL'TA(Jo1))OBTI(J)*SIN(THETA(J 1)) IVYEGER OPT.OPP UPTIJ 0?r:2 ALUAINIUN TILE IF‘OPTQEUQ‘)G” TO 414 CLAY TILE 1r¢npr.£u.2) GJ To 555 CUHEUCAFED ASBESTOS TILE lrtdvt.fiu.3)co T0 666 XCII. Go To 6c? xc=1.0139 60 TU 607 AC81.O274 Du )33 Jllox STI(J)83T1(J)OXC CUHIC DHJILUIHC 1F(OPF.LU.1)GU TO 115 RETAVGULAR BUILDING.LONC IF10PF.i 'U.2)G 0 T0 556 RETAVGULAR BUILDING.HIGH 1F(UPF.SN.3)GO TO 668 BLTANGULAN BUILDING.81NGLE SLOPE 1F(OPF.EU.1)CU T0 778 xc-1.015 60 TU 869 1:816 63 TU UB9 XCI1.016 GO TO 089 XCI1.UO14 DU 999 J81.K 5T11J)3ST1(J)'XC cuflPUTE INDIVIDUAL TERMS OF THE HEAT BALANCE DO 73 1:1.K 00(1):META-(GANHA&STI(1)) 01(1)8(VENTMOCPAIR105TI(I) 001I)x(VEuTMOCPAIR).5TO(I) DU 72 1319"" 051118(AN11100d11)1*(AVETI-AVET5111) DU 71 1:1.K ur(1)-u.n DO 71 J81.L AkGa(SuRT(FREU(J)/(1.!ALPHAP)))6(((AFIKF)OAA(J))01.0)O(1.0.0.0) ARG=(ARGICEXP((0.0.1.0)O(PHIEP(J)oPlFOUR-BB(J))))DT1(J) A9812.0)*REAL(ARG) Bu:(-2. 01-AZMAG(ARG) QF(I)IUF(I)+A09C05(THETA(J0I33+80981NITHETA(J011) 00 7o 1:1,Nw 0° 10 J31.“ 01(‘OJ7'00 14¢» no 70 JJ-1.L ura(tA«(1)-xn(1.JJ))~TS(I.JJ))~C€XP((0.0.1.0)sOWEGA(I.JJ)) ersnr¢((~fl(1)~4J(1.JJ))¢TI(JJ))-csxv((o.o.1.0)»9erptda)) A082.OOREAL(EF) aqa-2.0oAIMAc(EF) 7o 61¢I.J)sor(I.J)oauoc03(TntTA(JJ.J))oao»SIN(ruerA(JJ.J)) DU 175 1A:I¢K 5rU(IA)xo.5r(STotIA)-)2.) 55TO1IA):0.50(85T0(IA)-32.) 125 STI(1A):0.SI(5TI(IA)-32.) TYPE 777115711170131021) 777 FORHAT11X.21F5.0) ”U ‘26 11131.11.“ 00 126 11:1.K 126 515(10.1A):0.50(ST5(IB.IA)-32.) ' no 127 10:1.K uPtlA):JP(1A)&0.252 00(1A)=QO(IA)00.252 ul(lh):al(IA)¢0.252 05(1A)IJS(IA)GJ.257 127 UF(IA)8]F(IA)QU.252 DO 173 10:1.HH DO 179 1A:1.K 120 01(1M.IA)qu(IG.IA)oo.252 00 179 131.30 AH(I)=0.073'A~(I) Ud([)84.9OOUJ(1) La(l)=z.51~Ld(L) K~(IJ=U.5“'Kd(I) 129 ALPHAH(I):H.193'ALPHAN(I) Alana5-1.ASBIAIRHAS BETA80.252908TA GAHWLIIS.ROGAH4A R030,076Rfl "180.67.”! RHUF:16.02ORHOF KF:1.490KF “F'1049.NF AFI0.09JOAF ALPHAFIO.4930ALPHAF veuruan,qss.veurn c no. PRINT THE OUTPUT URIT€(6.4n01)1XK.nH 4606 FORWAT11H1.11.'DATE :‘.lJ.'/'.I2.I) NR1751002903K “RIT€(D.237)(KOUNT1J).J=1.K) 2&0 , FUKHAT(1N .zsuru. OUTSIDE AIR TEMPERATURE (.12.8H VALUES).//) 201 PUR"AT(1H .5X.2415.l.5X.2115./) 0R1T£(6.2JR)(5TU(J).J81.K) dRITE(o.20])K - 203 FUR"AT(1N0.25HTS. SOL-AIR TEMPERATURE (.12.17H VALUES PER WALL).II h) 00 9o I-I.nw 99 dRIT£(b.200)Ip(515(10J19J810K7 254 PORNAT(1H .13.!X.2OF5.0.6X.2§F5.0) uRlT€(6.244)K 214 219 221 265 09 2:6 224 272 207 223 217 239 231 232 39 233 234 c lb. 500 400 544 145 ruewartxuo. leCALCULATED OUTSIDE AIR TEMPERATURE (c .12.8H VALUES).II) uRIT£(6.237)(KOUNT(J).J=1.K) dRITF1Oo2JF)(5570(J39J31'K) IRITK(6.719) FOHHATtlun.iUHIUSIOS Ala TEUPERATUNr AT SPECIF TIUESoI) wkITn(6.737)(ROUNT(J).J:1.K) uRITR(o.71*)(5TI(JJ.J=1.K) d“l?€(00221) FOHwAT(1H1.25uanLL AREAS AuO PROPERTIES) nulra(o.705) rdnwfir(1uo,3d I .2x,2nAu,sx,2HUw,«x,2HLw.4x.2HKN.2x,6HALPHAu./) DU 89 1310““ “Hird100200)I:“~(I).UU(I).LH(I).K0([J.ALPHAH(I) FONWAT11H .12.1&.F6.0.4Fu.3) quTs(o.224) FJHWAV(1N0. 29HADDITIUNAL PROBLEM PARAHETERS) unlra(o.222) FUNHDT(1HH.2X.1SHAIRMASS In BDLG.5X.SUCPAIR.11X.4HBETAo10X. .SMGAHHA.12x,2unu.13x.2HRI.10x.7flCPFLOOR) «uITE(O.207)AIH115-CPAIRoBETA'GAHHA.R0.NIoCPF FUH9A711N .1P7515.4./) uKIT5(6.?21) PURWAT11H .3X.13HFLOOR DEUSITY.2X,1SHFLOOR K VALUE.1X. LISHFLUUR SURFACE H.3X.1OMFLDUR AREAo4XoIIHFLOOR ALPHA95X0 ~9HVENT RATE) URITH(6.207)RHOF.KF.HF.AF.ALPHAF.VENT“ FJHHAT11M1) uRITE(6.239) FORMAT(1H1.23X.76HBUILDIJG (CONTROL VOLUME) HEAT GAIN AT HOURLY IN .TERVALS. LISTED AS KCAL/HR../lll) uk1T£(6.231) FuuHAT11H .37HAdIHAL HEAT PRODUCT10h DURING THE DAY./) ufilrtto.237)(KJUNT(J).J=1.K) uKITd(6.?3H)(09(J39J81oK) HHITE(6.232) FORMAT(1H .21HVENTILATIUN HEAT LUSSo/J DU 39 J81.K UU"(J).')I(J).U\J(J) NRITE(6.237)(£UUNT(J).J=1.K) uRITE(6.238)(0"H(J).J:1.K) wnIT£(o.233) FukHAT11H .46H3TEADY STATE HEAT LOSSoLISTED BY HALL;KCALIHR./) NKINI41 «RIT316.237)(KOUNT(J).J:2.NK) deTH16.2Ja)(03(J)oJ-1.uu) IR1P816.234) PUNHAT(1N0.22HHSAT L055 TO THE FLUUH./) «RITE(6.237) (KUUHT(J).J=1.K) “RAT§(O'2392(JF(J)QJ310K) TOTAL HEAT BALANCE DJ 4)” J:1.K OUH1J)IO. 00 50" 181.0d OUM(J):OUW(J)+02(IoJ)-US(I) OUH(J):UP(J)-0P(J)-OI(J)+uO(J)+OUM(J) d812£16.514) FORMAT(1N .IBHTUTAL HEAT BALANCEoI) dRITB(b.237)(KOUNT(J).J=1.K) 2 2 (\fhflfifhnli 37 39 146 “RITE(6'238)(UQW(J)OJ=1QK) FUH“\T(IH 94(7K¢12110,/,1X)) FUN"AT(‘H . 4(10X.12F10.2./.IX)II) RETURN END SUHROUTIHE FUURIE(F,NDP,NH,A0.A,H) SUHROUTINE T0 COMPUTE FUURIER SERIES HEAL COEFFICIENTS F IS THE DATA POINT INPUT VaCTUR HOP Is THE no. UF DATA POINTS IN F H“ Ib THE NO. OF HARMONICS DESIRED AU Is THE SAVERAGE DATA-VALUE (OUTPUT) AJ.8J ARE THE COEFFICIENTS (OUTPUT) DIMKUSIUN FINUP).A(NH).O(HH),AA(10).Hn:10) X=NUP CISC”S(6.28319IX) s:=51~(b.28319’£) Cal.U 580." Nulzuuol OJ 3 J:1.NH1 U;=U.0 0120.0 NSNDP UO:F(N)9(2.00C)9UI-U2 02:0! UlsUfl Nan-I IFIU.GT.I)GO To 9 AA!J)=(2.01X)'(F(I)+C¢UI-UZJ BH(J)=(2.0IA)¢(S|UI) ' UflCI'C'SI’S 5=C1059515C Clu AOIAA(I)12.U DO 7 J:2.HH1 “(J'l)ih|(J) 8(J-I)IBB(J) RETURN END 147. Appendix C Example of the yearly weather simulation output. 148 co. n.c.: .7.1. ..o ,..~ ..nw ~._ 5.9a ...V a... Ixxua :NJ 2222222 222222222 222222222 22222222 22222222222222222222 22222222222222.22222 22222222222222-2222- 9:.I :2..u<.; a4u. wx:xr‘1; ... ... . up: 1&3 I»: .r .us .a .c:;.:« xca:a 2.. 11;:n 35a xo‘3;,a o. 3. I_. . .:2 up. ..N: .x: :4::I.;;. .J.:;x;~u <;:.c.n:a ”.9: .tamumrxwu... ....n...1 .I.e:.p a? .24 z n 222222222222222222222222 l "I ll 2 I D 2222222I ll2. I ll" .1 9 ..zo ~::.: .._ o:.a .c.:- n.n. no.o ..= . —+.ou ..a I a:x; .5“: ... .s;.: ..z.. eo.: »¢.: «.0 nu.= o.n~ re~.¢ o.»— I «aux a—gh c.. .:;.; .r:.. ,c.: In._ ‘.a :s.: ..»u :~.~ «.9. I .u. .: I 2222222 222222222 222222222 22222222 22222222222222-22222 22222222222222222222 22222222222222222222 3x.I :aa_ac‘. atu¢ uranaazz .a: :.: I»: an: awn 4.: ,4. .. fl I:¢xu.¢ Icaan c.¢ xtann I.I IJIxa.I onc:;;¢,x .:.:;:a:. carat<<a :3: can a;« >4. 2.5 a»: docx.,¢ 344:4 2:4 xcacn u»¢ wc¢1a>< orgua ~._;stuco In::;~x<7 142:0.xac a ontcsa ca:zn.;ao says: ...1uu;cut n. zawnsee.a I..u~ut a: u;< r a z 8 ..3. .~:.: .x-. :=.: ...:u :.eu :_.a ..:a .5. o u... I 13:; . : ~o. n~n.: .:o:n cc.= _~._ a... 72.; \.~n a... owx I sane .. a a.. «no.: .-to ca.) .:.. ..I, .B.: a... .o.« ..~« u;;u» .. 4 2222222 222222222 222222222 22222222 22222232222222222222 22222322222222222222 22 "U 34—: :3—_u¢2. ;:u‘ ax:n:;x; ..L .2. . ..3 :Pt c.¢ . .m .> sacs..< x233» ..s n...& 2" u);.¢.< .411: .2; b...n 4:. .<.: .z; r1:.a~.‘. 1.4 . ~. .:.:u. I. . xux¢2 sauwuupwnhdhn 13~¢aatnw sutuIaI so uxIrxzo 18:. ¢.: ~o=.:o .o.vu 30.: I..: a..- ago: «as. n~.o do: I 2535 mun: o.. can.: .p..n no.9 o~.= a... ~a.o ..an as.o o.n~ I :43» ouzu n.. -o.a .n~_o s:.o 3a.: 0..“ oa.o ~.=n an.o n.n~ Itxup 92:4 2222222 222222222 222222222 22222222 I 1 31.; 23.nu43. aaux uxanwaan ,2: an» a>< um; can u,< pun spa u»¢ 302:..I z¢434 s>« acaou w>¢ uUwa>¢ IquIau>< hz—oaxuao In:=:—x¢ a4:aouxao uza: gum m>< ,wo sun u»: >45 can a>¢ awe;.,¢ c443» ..a xIaaa 3,2 wecua>a .manxu». ~=_caxuco awr=r~x¢a caacouzo. onuoaau.< aa:no_¢oo Izza math hddadx or hauzu m4 Wuxh Dov—O IFJ< >x4szaa saunuuhmuhauo zauaarrna «u:~._ nJI.: ...._I ..D—_U21 EFT. gun “—.o0 n.: ma.— c.on c«.. p.on >mc :hn w,- 2 . 2.222.222.22222222.22....0...‘.2.I. — c. 338 «n.00 0.6 I 3:35 35.9 a... 0..“ I saga aux» on.— non~ Irauh oa:a >3: aha u>¢ .¢.I.>I xgasa ..I I.4:m ..4 .u.a.>4 .:.:.:m>z .n.:a .4. oar: .xa: ca;:u»;c. .mw u=uan :J::qu:I :x:.xuq; :.;nu.umqru»u : nuae_.n 2.:uaxI a: -:4 Ian 2222222 I IIII2.222222222222222222222222222222222222222222222222222222222222222222222222222222I0III 2222 I "I" a a ;.e- .=:.a. .II. 35.: :p.: c.no m:.on ~.¢o pn.no ..o: I 13;; ...o ..~ 9...: .0... a:.: r:.; p.np mp.c ¢.9« «x.o a.- I z9: chm 23¢ >.: ehn s>< ..II :43;a .;< x< 2043;:4 Itusc; .t;mu~pn«hmun :r.«ar:~a 2.:h<.r a: u:¢.x a 2 n..2222222222220 IIIII II III I II II22 I II I I0I2222222I0I II I I I I I I :33: c.o s_:.oo Imo.l oo.oa o—I—u oInI no.0! 0.0 nnooo coco I lack than ~.~ 3...: ..san oo.o ...c n..« n..o ..on 9a.: e.n~ I IIu» w.:~ . ~.~ 9...: .M... oo.= oo.. o.o— on.o ..o. a... c.n« Iaamh uzaa 2220022 222200222 222222222 22222222 IflI I III I 52.. ::__uI;; gap; uzacmuxa >wc as» u>¢ >u: cum u>< >u: own u>< IJII.>I sage» .>I xaaom w>4 ween»). caucaxw>q hznaaxwoo Insaruxcx cazcouon Inmuexm>< mango-mac «uterus souwu-n~h¢nn mahqaezun «wtuawt s: uznzxao 3318 ..c ~.:.o. .moco ao.a :o.cu a... ~o.o ..eo no.o 9.9. I Iasu 55—0 ..~ ~.n.o ....n .o.o a... o... ~n.. I.on an._ a.- I :4u~ aura ~.~ apo.: .Iozn .o.e o».— I.n. en._ o.On oo.c n.- Ixxwh uaaa 2222222 222222222 222222222 22222222 "I I I I """I I II I II a... 2:..u‘;. aau‘ xazanuaa >¢o own u>< >ua can v.4 >wc awn m>4 .uII..I zeaau .>a aIaou 3.. weaau>< Ina042u>¢ ~:_ou1woc Ia7=r_x< >ma aka u>¢ sun as» u>< 1.2:, ..I ;:.:t ..c guaza.¢ Imuc¢ hr_ca.ch our::_x¢: ranco>zao Inuonsms« m4:xI-;:I ..I? ~—;IT sI_ 0.209 I \I s v.1 t o—I- I...-I.I.I.I& 5J¢LQ)Q .......h I... I). ..I.. Izhr. .22222222 aIUI 15...! I)... ....I..2 €on0 ’0\2 \-.I ..I. F... x... .;.. c.,. IOIIOIIO IDIOIOOIIOIOIOIIOIOO .41.}.vkn... :31 ......V I... f...:.;>a 23.13:: It... ._. —._.‘ ..Z. kv3v72 atunu .:I:l :I.2 ~;.. ..c~ .... ¢.c« >u: :wu v>p In... ._..uw... .4..t|s::I was: ufunu ~»w~ruuw c:»«na:~w a._~«vr be ~;¢;I;o 3.: .o. _.:o I :nxu ”5.9 .h.. :.I" I Imuy .— h ru.. o.r— Isaak Ja;a >.r ppm «I. .m. u. a. s .. . 31.0.23 “Inqk.hs m7h.19. n . .;.ar 5: u:a.::n ..cn _.;.: .2. '9'...’ .“‘ owsa. :.I V\I.l won! «.10 ;01 "—0.2 c..- ..o— ‘2'..\ 3.: a .u r... I 13:: .5”: . a ..s. I save —.h 150 33 ha arxs..,......,.._. «.94? ungz ax. .. 1.7.p.;.su 2w.a.a za:s:n axr.. n::~..;.7u :n.nwa za.zun OIOILH—f naquaL::w gekacx~xfiz a #2 .2ub44:W.H :¢::1 Japaa .zh Iv 3:4.247 .3 a. ~.:«h n mark as. ;;nyer.m:xr .::...c .t» an”......_u\...:2€. W17. 2......» 1&7“. tank u..m_—H£.N.UL3.HF 1574‘...) d..P IIII1LJ:M¢;3;xuh OQIIIIIIIIIIIIrIIIIIIIIOIOIOIOI. 2 2 . yx¢u5=~ 5a¢=a= 54 co.x..n5.. ncuzoco¢.. u5.x 5:5» «:54. mesa. .ouucoqn.. .oouac5o.v 55 a; 55.55 na.on n5.on 05.05 55.55 5n.n~ 55 a. 5. a 5 o 55.55 05.55 .o.m5 oo.c5 5¢.55 v5.0. o5. a. a. c 5 o ..5 .55 .o5. .on .on .55 ..5 ..5 .55 .en .55 .55 ..5 ..5 .55. .c5 .en .55 .25 .o5 .55 .3. .an .55 ..5 .55 .55 .:n .on .55 .I5 .95 .55 .cn .an .55 5. e. a. o. n. 5. oo..5 53.55 ...o5 :5.55 .a.o5 «v.5. a5 a. a. I 5 o 25 qax5zouv uznoaunn «conncan.e amm< «cog. nouwaoo¢.. nncmooaoIO mno.u.5q5.5 : uu.5.=n «ace. 2:445 5 gang. 55.ngua «sods “(VFHOUIO ‘2-2 a 8‘ oeoueco..n ne.ea5.m.. .cuwo5o..5 no.aasoo.5 «tr.o .5xa c.45u can: a. 554x554 wxwhuzaxan amaaazs a<=3555204 Ivo.c c55.. 355.. 5aa.55.5:e o.o.e :55...e~5.. 5oa.55.¢:v aoa.e 5...: .zn.e55an.o .o.. 5.5.: 5...: .en.e5uan.o .:5 3.5.5 ..—.e .=5.555o..o ..5. o.c.o 5...e .rn.e55o..a .oe n<=53< 1: 14 I: 34 nwnpawgrza :24 n v5. 95:5‘auazu5 x54 uc.55=a au554=u44u ..5 .a5 .55 .55 .05 .5. .5. .o.. .5. .5. .5. ..5 .55. .55 .55 ..o5 .5. .5. .55 .5. .5. ... ..5 .05 .55 .55 .35 .5. .5. ... .5.. .5. ... ..5 .c5 .55 .55 .c5 .55. .5. .a. .5. .5. .5. ..5 ..5 .55 .55 .o5 .5. .5. .o. .5. .5. ... .55 .95 .55 .55 .55 .5. .5. .o. .5. .5. .a. .. o. e. a 5 o n . 5 5 . .44.: a»; Ina..> e5. 5525452.:55 5.4uaau .5. .55.55 5e..5 .5.:5 a=.:5 59.55 5..:. 55.». 55.5. 55.0. 55.5. 5. a. 5. e. w. m e 5 . anu344> en. ucahdxusxuh and aaunha3 Ia» \‘ . uh4o 1.513 5.....5 5...5... o~ ~— 5.000 95.5— v5 «— .ooou‘.n 55.5005. on «- o..o.5o.. 55....59. on 5. 5..5... ....... ....... 55.5.51. oe.naono 02.5550. .5 55 . .5 .. a. o 5.... co... cc... 5... .5.. a... .5 55 .5 .. a. . no.e..55 05...:55 «v.5..n5 «c.5055. ..._qu. oo.no.n. an an .N a. :5 o ‘ 55.55.55. 5......5. 55.05559. vo.555oo. 5.....50. v5.5...o. 55 55 .5 .. a. o 5...... .a.o..o .o.oovc 55....» .5..5vo. .o.5.¢o. 55.¢o.e.. 5....oo. .5 .o. a. 5. c. 5 . n a... 55.. 5... ..... .5... 9.... co.... ...... .5 a. .. 5. . 5 o m ......5. .5.5..5. 5 . 5......5 .5..5.55 o..55o.5 ....5... 50...... .5.5.c55 a..e.v55 ov.55.55 on 05 3. ha c 5 o r ...o5..o. 5......0. .......o. a..5.5.o. ....5..o. ......o.. 5..5..... ......5.. on a. a. h. a 5 o n .5.vn.n ....5... a. v 55... 5..... o— nhonon n .5..5.5 .5...5. n— n .5... o..n.. «— 55.55 v ....55. 5e.ooa. 0. .5.5.5 . .e.595. 55.5.9. n— . . duzia‘n h43:.4<.35 5.... 5a.... a. . .xaa.. 2:5 on «no. 54.: 55... 5 5¢:\a¢ux~ad o..5..5a. .5.55.5e. o..o..5o. ....5.5a. .o.5..5.. ......5.. .5...5... ......... c. o n. n v. N n— . x