Design of a Decentralized PID Controller for Poultry House System using Genetic Algorithm

Abstract

Climate control for livestock building is of considerable significance but it is additionally a difficult and convoluted task. The chickens are mainly affected by the variation of temperature and relative humidity. The arrangement of these parameters is accomplished by selecting proper control techniques. In this paper, an optimized controller for the stabilization of the poultry house system has been designed in order to reduce the heat stress of broilers and to achieve the preservation of chickens’ health and comfort. A hygro-thermal model describing the behavior of the poultry house is decomposed into two single loops based on the theory of the effective open-loop transfer function. Adopting a model reduction technique, the equivalent transfer function of each loop is then designed by developing an independent multi-loop PID controller. The initial stability of the reduced model is assured via the Hermite-Biehler theorem. Then, the Genetic Algorithm is adopted to search the optimal gain values that contribute to the desired indoor climate monitoring. An extensive numerical simulation is tested with original experiments measures to show the effectiveness of the design control and the results are compared to those of the ant colony optimization and Ziegler-Nichols methods.