SciELO - Scientific Electronic Library Online

 
vol.42 issue1Methods of incorporation of the enzyme phytase in diets for laying hensIncorporation of n-3 in muscular tissue of Nile tilapia fed with shrimp head silage author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand

Journal

Article

  • English (pdf)
  • Article in xml format
  • How to cite this article
  • SciELO Analytics
  • Curriculum ScienTI
  • Automatic translation

Indicators

Related links

Share


Ciência Rural

Print version ISSN 0103-8478

Cienc. Rural vol.42 no.1 Santa Maria  2012

http://dx.doi.org/10.1590/S0103-84782012000100027 

ARTICLES
ANIMAL PRODUCTION

 

A fuzzy system for cloacal temperature prediction of broiler chickens

 

Sistema fuzzy para a predição da temperatura cloacal de frangos de corte

 

 

Leandro FerreiraI; Tadayuki Yanagi JuniorII, 1; Wilian Soares LacerdaIII; Giovanni Francisco RabeloII

IPrograma de Pós-graduação em Engenharia de Sistemas, Universidade Federal de Lavras (UFLA), Lavras, MG, Brasil
IIDepartamento de Engenharia, UFLA, 37200-000, CP 3037, Lavras, MG, Brasil. Email: yanagi@deg.ufla.br
IIIDepartamento de Ciência da Computação, UFLA, Lavras, MG, Brasil

 

 


ABSTRACT

Cloacal temperature (CT) of broiler chickens is an important parameter to classify its comfort status; therefore its prediction can be used as decision support to turn on acclimatization systems. The aim of this research was to develop and validate a system using the fuzzy set theory for CT prediction of broiler chickens. The fuzzy system was developed based on three input variables: air temperature (T), relative humidity (RH) and air velocity (V). The output variable was the CT. The fuzzy inference system was performed via Mamdani's method which consisted in 48 rules. The defuzzification was done using center of gravity method. The fuzzy system was developed using MAPLE® 8. Experimental results, used for validation, showed that the average standard deviation between simulated and measured values of CT was 0.13°C. The proposed fuzzy system was found to satisfactorily predict CT based on climatic variables. Thus, it could be used as a decision support system on broiler chicken growth.

Key words: expert system, fuzzy sets, thermal environment.


RESUMO

A temperatura cloacal (TC) de frangos de corte é um importante parâmetro para classificar a sua condição de conforto, portanto, a sua predição pode ser usada no suporte à decisão de acionamento de sistemas de climatização. Objetivou-se com a presente pesquisa desenvolver e validar um sistema, utilizando a teoria dos conjuntos fuzzy para predição da TC de frangos de corte. O sistema fuzzy foi desenvolvido com base em três variáveis de entrada: temperatura do ar (T), umidade relativa (UR) e velocidade do ar (V), tendo, como variável de saída, a TC. A inferência fuzzy foi realizada por meio do método tipo Mamdani, que consistiu na elaboração de 48 regras e a defuzzificação por meio do método do Centro de Gravidade. O sistema fuzzy foi desenvolvido no ambiente computacional MAPLE® 8. Resultados experimentais, usados para a validação, mostraram que o desvio padrão médio entre os valores simulados e medidos de TC foi de 0,13°C. O sistema fuzzy proposto prediz satisfatoriamente a TC com base nas variáveis climáticas, podendo ser utilizado como suporte à decisão em sistemas de criação de frangos de corte.

Palavras-chave: sistema especialista, conjuntos fuzzy, ambiente térmico.


 

 

INTRODUCTION

Thermal comfort of broiler chickens can be characterized through its physiological responses, such as cloacal temperature (CT), and it is influenced by the climatic conditions, such as air temperature (T), relative humidity (RH), and air velocity (V), among others. To obtain suitable climatic conditions for animal production, the thermal environment inside of a broiler house should be into the thermoneutral zone (TNZ) (CURTIS, 1983). In this zone, the animal reaches its maximum potential and body temperature is maintained into acceptable level with minimal use of thermoregulation mechanisms (SMITH, 1964). However, when the thermal environment is out of TNZ, the environment becomes uncomfortable and losses in performance are imminent (BORGES et al., 2003).

In the TNZ, the CT is maintained between 41.2°C and 42.2°C (TAO & XIN, 2003). According to MEDEIROS et al. (2005), for broiler chickens the thermal comfort is characterized by values of T between 18°C and 28°C in conventional houses, with RH between 50% and 70% and V around 1.0 to 2.5m s-1. However, MEDEIROS (2001) affirms that the maximum productivity is obtained when the T is in the range from de 21 to 27°C, with 50 to 70% RH and V from 0.5 to 1.5m s-1.

As thermal environment influences broiler chicken comfort and, consequently, the performance, thus a decision support system can be drawn to control the thermal environment inside the broiler houses, considering the animal physiological responses, such as CT. Therefore, expert systems based on fuzzy sets theory are an alternative for the management of uncertainties in the rearing environment of broilers.

Fuzzy sets are also called misty sets or diffuse sets, and it is an extension of the classic logic. It was first studied by Lofti Zadeh in the University of California Berkeley, in 1965, when he published a paper titled "fuzzy Sets" in the journal Information and Control (ZADEH, 1965).This methodology is one of the more recent specialties of the artificial intelligence area that aims to generate techniques to solve problems in several knowledge areas, approaching the computational decision to the human decision. fuzzy set theory uses approximate instead of exact information, imitating the human thinking. Nowadays fuzzy set is used in control systems and in decision support systems where the problem description approach cannot be precise.

A fuzzy system is formed of output and input variables. For each variable, fuzzy sets that characterize those variables are formulated, and for each fuzzy set a membership function is built. After that, the rules that relate the output and input variables to their respective fuzzy sets are defined. The computational evaluation of a fuzzy system is formed of fuzzification (construction of output variables that define the study), inference (fuzzy reasoning application on fuzzy output) and defuzzification (translation of linguistic value to numerical value). The fuzzy reasoning can be implemented by a direct method or indirect method (TANAKA, 1997; PEDRYCZ & GOMIDE, 1998; MENDEL, 2001).

fuzzy set theory has been used in several areas, such as biomedicine and biomathematics (ORTEGA, 2001; BARROS &BASSANEZI, 2006). For animal application specifically, the following studies can be citied: on poultry environment (YANAGI JUNIOR et al., 2006; SCHIASSI et al., 2008) and for swine (PANDORFI et al., 2007); and in the detection of estrus in dairy cows (FERREIRA et al., 2007; BRUNASSI, 2008).

This work aimed to develop a decision support system based on fuzzy set theory for cloacal temperature prediction of broiler chickens as a function of thermal environment variables.

 

MATERIAL AND METHODS

Development of fuzzy system

The development of the fuzzy system was based data from research developed by SEVEGNANI (2000), YAHAV et al. (2004), SOUZA et al. (2005) and MEDEIROS et al. (2005).The following were defined as input variables: air temperature (T, °C), relative humidity (RH, %) and air velocity (V, m s-1). Selection of the variables and their range were based on the availability of the data in the literature to validate the model. Based on these variables, the fuzzy system predicts the cloacal temperature (CT, °C). The fuzzy sets of input and output variables are graphically represented by triangular membership curves (Figure 1). Triangular membership curves are the most common and suitably represent the behavior of input data, according to the literature (AMENDOLA et al., 2005; YANAGI JUNIOR et al., 2006; FERREIRA et al., 2007; SCHIASSI et al., 2008).

Considering, for example, the value of T = 25°C, it is possible to observe which fuzzy sets that this value is a member of (Figure 1), as well as its membership functions (T1 and T2). The fuzzy sets that contain T = 25°C were . Therefore, and µT1(25) = 0.17 and µT2(25)=0.83. The decision process admits that this value of T belongs to the fuzzy set T2 (greater membership degree).

The fuzzy inference was comprised of a rules system (Table 1) based on average data obtained by SEVEGNANI (2000), YAHAV et al. (2004), SOUZA et al. (2005) and MEDEIROS et al. (2005), as well as, by specialists. The rules system was built based on a database and on opinions of experts. Each rule is composed of logical connectives (if, and, or, then) and both antecedent and consequent parts. For example, IF x is A AND y is B THEN z is C, where A, B and C are fuzzy sets, x and y are the input variables, z is the variable output, 'IF x is A AND y is B' is the antecedent part and 'THEN z is C' is the consequent part.

 

 

The analysis of fuzzy system was developed using the software MAPLE® 8. It also employed the Mamdani inference system due its simplicity and satisfactory adaptation to fuzzy controllers. The defuzzification was done using the Center of Gravity Method that considers all the possibilities of the output fuzzy set (ORTEGA, 2001). This method can be understood as a weighted average in which weights are represented by µA(Xi), indicating the membership degree of the value Xi with the concept modeled by the output fuzzy set A.

 

RESULTS AND DISCUSSION

For the thermal environment defined by T=25°C, RH=55% and V=1.5m s-1, two rules were reached using fuzzy sets T1 (used memberships function and degree: and µT1(25) = 0.17, respectively), T2 (used memberships function and degree: and µT2(25)=0.83, respectively), RH2 (used memberships function and degree: µRH2(RH)=1 and µRH2(55)=1, respectively) and V2 (used memberships function and degree: µV2(v)=1 and µV2(1.5)=1, respectively) , as follows: a) rule 6 - IF (T é T1) AND (RH is RH 2) AND (V is V2) THEN (CTis RT1) and b) rule 18 - IF (T isT2) AND (RH is RH 2) AND (V is V2) THEN (CTis RT1).

The minimum membership degree of the antecedent part of each rule was obtained via Mamdani inference method through the minimum operator (logical operator AND - intersection) using rules 6 and 18. Thus, for the rules 6 and 18 the minimum operator () result in 0.17 11 = 0.17 and 0.8311 = 0.83, respectively.

Membership degrees of 0.17 and 0.83 obtained in the inference phase by the minimum operator for rules 6 and 18, respectively, were set as the cut values of the consequent part of each rule. The cuts were performed in the fuzzy set RT1 because both rules presented this set as a consequent. Afterwards, using the maximum operator (U), fuzzy set was obtained as an output.

The resultant fuzzy set presented 0.83 as a maximum membership degree. The next step was the defuzzification of this fuzzy set, which consequently had a numerical value of 41.2. Thus, considering a thermal environment with T=25°C, RH=55% and V=1.5m s-1, a resulting CT = 41.20°C was observed. The mean deviation (MD) and mean percentage error (MPE) between the results obtained by the fuzzy system and those measured experimentally by SEVEGNANI (2000), YAHAV et al. (2004), SOUZA et al. (2005) and MEDEIROS et al. (2005)were 0.13°C and 0.31%, respectively, obtaining a coefficient of determination (r²) equal to 0.9318. The MD, MPE and r2 obtained show that the fuzzy system satisfactorily simulates the cloacal temperature of broiler chickens. After adjustment of a neural network to predict CT of broiler chickens, LOPES (2009) obtained MPE of 0.78% and r2 of 0.8860 for training, and 1.02% and 0.8205 for validation, respectively.

Figure 2 shows simulations of CT as a function of T and RH, with V held constant at 0.6 and 2.4m s-1. Based on the results obtained by the fuzzy system, CT increases with increasing values of T. The V influences the CT when the animal is subjected to hot environments, and thus, in agreement with the results found by YAHAV et al. (2004). In addition, the profiles observed in figure 2 are in concordance to the expected physiological responses of broiler chickens, agreeing with the results observed in other studies, such as MEDEIROS (2001) and TAO & XIN (2003). The results show that CT increases with an increasing RH, but less so in magnitude than with an increase of T.

When thermoneutral conditions are considered, T ranging from 20 to 26.3°C (MEDEIROS et al., 2005), the CT simulated by the fuzzy system were in the interval from 41.2°C and 42.2°C, considered as typical values under comfortable environment (LACEY et al., 2000;TAO & XIN, 2003; LIN et al., 2005). On the other hand, for T equal to 35°C the CT simulated by the fuzzy system is 43.50°C, whereas MARCHINI et al. (2007) found 42.9 when the animals were subject to high cyclic values of T. Exposition of broiler chickens to high values of T results in reduction of performance (COOPER & WASHBURN, 1998), provoking decrease of feed intake, growth rate, chest productivity and meat quality, in addition to the use of energy to promote heat loss instead of production (DOZIER et al., 2006; LU et al., 2007).

 

CONCLUSION

The fuzzy system developed for the prediction cloacal temperature (CT) of broiler chickens, based on the thermal environment, characterized by air temperature (T), relative humidity (RH) and air velocity (V), provided a low standard deviation (±0.13°C), and satisfactorily simulated CT, helping in the decision-making process. Moreover, the theory of fuzzy sets is a promising tool in predicting the body temperature of broiler chickens, and can be used to help make decisions about the farming system to be used.

 

ACKNOWLEDGMENTS

The authors express their thanks to Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES), Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), and Fundação de Amparo à Pesquisa do Estado de Minas Gerais (FAPEMIG) for funding this research.

 

REFERENCES

AMENDOLA, M. et al. Using fuzzy sets to analyze environmental condition in order to improve animal productivity. Biomatemática, Campinas, n.15, p.29-40, 2005. Available from: <http://www.ime.unicamp.br/~biomat/bio15art3.pdf>. Accessed: Aug. 26, 2011.         [ Links ]

BARROS, L.C.; BASSANEZI, R.C. Topics of fuzzy logic and biomathematics. Campinas: IMECC-UNICAMP, 2006. 354p.         [ Links ]

BORGES, S.A. et al. Heat stress physiology and eletrolites for broilers. Ciência Rural, Santa Maria, v.33, n.5, p.975-981, 2003. Available from: <http://www.scielo.br/scielo.php?script=sci_arttext&pid=S0103-84782003000500028&lng=pt&nrm=iso>. Accessed: Aug. 26, 2011. doi: 10.1590/S0103-84782003000500027.         [ Links ]

BRUNASSI, L.A. Development of a fuzzy system for estrous detection. 2008. 101f. Dissertation (Master in Agricultural Engineering) - Campinas University, Campinas, SP.         [ Links ]

COOPER, M.A.; WASHBURN, K.W. The relationships of body temperature to weight gain, feed consumption, and feed utilization in broilers under heat stress. Poultry Science, Champaign, v.77, n.2, p.237-242, 1998. Available from: <http://ps.fass.org/cgi/reprint/77/2/237.pdf>. Accessed: Aug. 26, 2011.         [ Links ]

CURTIS, S.E. Environmental management in animal agriculture. Ames: The Iowa State University, 1983. 410p.         [ Links ]

DOZIER III, W.A. et al. Growth responses of male broilers subjected to high air velocity for either twelve or twenty-four hours from thirty-seven to fifty-one days of age. Journal Applied Poultry Research, St. Joseph, v.15, p.362-366, 2006. Available from: <http://japr.fass.org/cgi/reprint/15/3/362.pdf>. Accessed: Aug. 26, 2011.         [ Links ]

FERREIRA, L. et al. Development of algorithm using fuzzy logic to predict estrus in dairy cows: part I. Agricultural Engineering International: The CIGR Ejournal, Dublin, v.9, p.1-16, 2007. Available from: <http://www.cigrjournal.org/index.php/Ejounral/article/view/898/892>. Accessed : Aug. 26, 2011.         [ Links ]

LACEY, B. et al. Monitoring deep body temperature responses of broilers using biotelemetry. Journal Applied Poultry Science, St. Joseph, v.9, p.6-12, 2000. Available from: <http://japr.fass.org/cgi/reprint/9/1/6.pdf>. Accessed: Aug. 26, 2011.         [ Links ]

LIN, H.et al. Thermoregulation responses of broiler chickens to humidity at different ambient temperatures. II. Four weeks of age. Poultry Science, Champaign, v.84, n.8, p.1173-1178, 2005. Available from: <http://ps.fass.org/cgi/reprint/84/8/1173.pdf>. Accessed: Aug. 26, 2011.         [ Links ]

LU, Q. et al. Effect of chronic heat exposure on fat deposition and meat quality in two genetic types of chicken. Poultry Science, Champaign, v.86, p.1059-1064, 2007. Available from: <http://ps.fass.org/cgi/reprint/86/6/1059.pdf>. Accessed: Aug. 26, 2011.         [ Links ]

MARCHINI, C.F.P. et al. Respiratory frequency and cloacal temperature in broiler chickens submitted to high cyclic ambient temperature. Archives of Veterinary Science, Curitiba, v.12, p.41-46, 2007. Available in: <http://ojs.c3sl.ufpr.br/ojs2/index.php/veterinary/article/view/9227/6453>. Accessed: Aug. 26, 2011.         [ Links ]

MEDEIROS, C.M. mAdjustment of models and determination of environmental thermal index of productivity for broiler chickens. 2001. 115f. Thesis (Doctor's degree in Agricultural Engineering) - Federal University of Viçosa, Viçosa, MG.

MEDEIROS, C.M. et al. Effects of temperature, relative humidity and air speed on broiler chickens. Engenharia na Agricultura, Viçosa, v.13, n.4, p. 277-286, 2005. Available from: <http://www.ufv.br/dea/reveng/arquivos/Vol13/v13n4p277-286.pdf>. Accessed: Aug. 26, 2011.         [ Links ]

MENDEL, J.M. Uncertain rule-based fuzzy logic systems: introduction and new directions. New Jersey: Prentice Hall PTR, 2001. 576p.         [ Links ]

ORTEGA, N.R.S. Applications of the fuzzy logic theory to problems of biomedicine. 2001. 152f. Thesis (Doctor's degree in Science) - Universidade de São Paulo, São Paulo, SP.         [ Links ]

PANDORFI, H. et al. The use of fuzzy logic for the productive environment characterization for pregnant sows. Revista de Engenharia Agrícola, Jaboticabal, v.27, n.1, p.83-92, 2007. Available from: <http://www.scielo.br/scielo.php?script=sci_arttext&pid=S0100-69162007000100001&lng=en&nrm=iso>. Accessed: Aug. 26, 2011. doi: 10.1590/S0100-69007000100001.         [ Links ]

PEDRYCZ, W.; GOMIDE, F. An introduction to fuzzy sets: analysis and design. Cambridge: MIT, 1998. 461p.         [ Links ]

SCHIASSI, L. et al. Metodologia fuzzy aplicada à avaliação do aumento da temperatura corporal em frangos de corte. Engenharia na Agricultura, Viçosa, v.16, n.2, p.180-191, 2008. Available from: <http://www.seer.ufv.br/seer/index.php/reveng/article/view/15/5>. Accessed: Aug. 26, 2011.         [ Links ]

SEVEGNANI, K.B. Avaliação dos efeitos fisiológicos causados pela ventilação artificial em frangos de corte, em dispositivos de simulação climática. 2000. 96f. Tese (Doutorado em Engenharia Agrícola) - Universidade de Campinas, Campinas, SP.         [ Links ]

SMITH, C.V.A. Quantitative relationship between environment, comfort and animal productivity. Agricultural Meteorology, Amsterdam, v.1, p.249-270, 1964.         [ Links ]

SOUZA, B.B. et al. Effect of the energy level and supplementation with potassium and ammonium chlorides in the diet on the physiological parameters and the performance of broiler in the summer. Ciência e Agrotecnologia, Lavras, v.29, n.1, p.185-192, 2005. Available from: <http://www.scielo.br/scielo.php?script=sci_arttext&pid=S1413-70542005000100023&lng=en&nrm=iso>. Accessed: Aug. 26, 2011. doi: 10.1590/S1413-70542005000100023.         [ Links ]

TANAKA, K. An introduction to fuzzy logic for practical applications. Tokyo: Springer, 1997. 138p.         [ Links ]

TAO, X.; XIN, H. Acute synergistic effects of air temperature, humidity, and velocity on homeostasis of market-size broilers. Transactions of the ASABE, Saint Joseph, v.46, n.2, p.491-497, 2003. Available from: <http://asae.frymulti.com/abstract.asp?aid=12971&t=1>. Accessed: Aug. 26, 2011.         [ Links ]

YAHAV, S. et al. Ventilation, sensible heat loss, broiler energy, and water balance under harsh environmental conditions. Poultry Science, Champaign, v.83, n.2, p.253-258, 2004. Available from: <http://ps.fass.org/cgi/reprint/83/2/253.pdf>. Accessed: Aug. 26, 2011.         [ Links ]

YANAGI JUNIOR, T. et al. fuzzy logic model to predict laying hen body temperature rise during acute heat stress. In: CONGRESSO BRASILEIRO DE ENGENHARIA AGRÍCOLA, 35., 2006, João Pessoa. Proceedings… João Pessoa: SBEA, 2006. V.1, p.1-4.         [ Links ]

ZADEH, L.A. fuzzy sets. Journal Information and Control, Berkeley, v.8, p.338-353, 1965.         [ Links ]

 

 

Received 11.16.10
Approved 09.12.11
Returned by the author 12.05.11
CR-4377

 

 

1 Autor para correspondência.

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License