Parameters optimization of relay self-oscillations sampled data controller based on particle swarm optimization

Abstract

The applied task of synthesis of digital self-oscillating control system is considered. Problem of correction devices parameters search is considered as the task of finite-dimensional optimization task. For its solution parallel swarm optimization algorithm is worked out. Algorithm is applied to parametric optimization of actuator digital control self-oscillating system. The proposed approach had superior features, including easy implementation, stable convergence characteristic, and good computational efficiency. When control system digital realization synthesis the exact algorithm of determination of periodical movement in relay system is worked out. Method of self-oscillations stability analysis in such a system is proposed. Optimization task is solved with taking into account the criterion of stability of self-oscillations in the system under consideration. Effectiveness of proposed methods and PSO algorithms was shown on the example of synthesis of digitally controlled relay system for an electro-pneumatic servo-mechanism. On the model example it is shown that the use of the PSO algorithm as compared with the genetic algorithm (GA) have better convergence.