Applied intelligent system used for controlling acclimatization of dairy housing / Sistema inteligente aplicado ao acionamento do sistema de climatização em instalações para bovinos leiteiros

AUTOR(ES)
DATA DE PUBLICAÇÃO

2007

RESUMO

The objective of this study was to develop a specialist system for controlling the acclimatization system in free-stall dairy housing. The modeling procedure used animal physiological data (rectal temperature and respiratory track) as function of environmental data inside housing (air temperature and relative humidity, and temperature and humidity index). Data base was constructed using two distinct environment as: subtropical climate (São Pedro, SP country) and Mediterranean climate (city of Évora, Portugal). Holstein multipart cows in lactation selected according to their weight and milk yield were used for building the data base with the physiological data. The cows were maintained during all day with free access to the beds and the other áreas in the housing, however without access to pasture. Body core temperature was measured with a clinical thermometer inserted un the rectum, and the respiratory frequency was measured by counting the flank movements for 15s. Meteorological stations were used for registering air temperature and relative humidity every 30 min along 24 hours, for the analysis of the physical environment both inside and outside the housing. In the initial process of analysis of the data base Dataming technique was used with the objective to form a decision tree in order to induce the base of rules. In this stage the software Waikato Environment for Knowledge Analysis (WEKA®), version 3-4, was used, and executed the task of classifying using the algorithm J48 applying cross validation technique for building up the decision tree. Results obtained were used afterwards applying fuzzy logic using the software Fuzzy Logic Toolbox do MATLAB® 6.1. The data base analysis indicated a good capability of the model to represent the classes of thermal comfort for the analyzed data. According to the found results it is possible to adapt the temperature and humidity index (THI) using Datamining of meteorological and physiological data. The use of the association of Datamining and fuzzy logic allowed establishing some ideal thermal comfort parameters for Holstein dairy cattle during lactation housed in total confinement. It was observed that the superior limit for the thermoneutral zone (ZTN) is located near dry bulb temperature of 26ºC. It was also found that values of THI around 77 may result in normal comfort conditions for the confined animals, and when the THI reaches values above 80 it indicates high heat stress.

ASSUNTO(S)

bioclimatologia animal ambient thermal comfort instalações para animais bioclimatologia mineração de dados. animal welfare data mining fuzzy logic. bovinos leiteiros comportamento animal fuzzy conforto térmico

ACESSO AO ARTIGO

Documentos Relacionados