Adjustment of controller PID's parameters of genetic algorithms

Authors

  • Francisco José Grandinetti Universidade de Taubaté
  • Luiz Eduardo Nicolini do Patrocínio Nunes Universidade Estadual Paulista
  • Victor Orlando Gamarra Rosado Universidade Estadual Paulista

Abstract

The objective of this work is the application of Genetic Algorithms in the automatic adjustment of controller PID'S parameters. The Genetic Algorithms (GA’s) constitute a search and optimization technique, highly parallel, inspired in the beginning of evolution of Darwin. The natural beginnings, on which AG's was inspired, are simple. The selection beginning privileges the most capable individuals and, therefore, with larger reproduction probability. For the development of this work the model of a motor of continuous current was used. The obtained results show that the use of GA in this model drove it an optimization of the controller's parameters and, consequently, to the minimization of the error, converging for the great values with less than 100 evolution cycles, demonstrating that the Genetic Algorithm is efficient for this application type, because it gets to find satisfactory values for the controller in comparison with traditional methods.

Published

2008-10-12

Issue

Section

Articles