Tuning of fractional PID controllers by using radial basis function neural networks

A new tuning method for designing fractional order PID controllers based on radial basis function (RBF) neural networks is presented in this paper. Though fractional order PID controllers can provide better control for dynamical systems, the difficulties of designing them increase. This paper deals with the design of fractional order PID controllers by first using an existed tuning method, and then taking advantage of good approximation ability of RBF neural networks to establish a mapping relationship between plant parameters and the parameters of fractional order PID controller. The greatest advantage of this method is that the controller could be easily obtained from plant parameters with good performance of closed-loop system. Simulation results for both integer-order plant and fractional-order plant show the proposed method is highly effective.