Identification of the plant dynamic using genetic algorithms

Abstract

For the purpose of research, modelling and simulation of control plants, linear models are commonly used. There are many methods of creating linear models of plant dynamics. Artificial intelligence methods are used in many areas of science and technology. There are also several known applications of artificial intelligence in automatic controllers. The authors have developed a method of linear modelling by means of evolutionary algorithms, which enable searching for optimal values of the transfer function coefficients. For a method validation, an experiment with a real thermal plant was conducted. This paper presents identification of a linear transfer function of a thermal plant.