Optimal Design of On-Line PID Controller Using Model Predictive Control and GMDH-type Neural Networks

In this paper Model Predictive Control (MPC) is used for optimal selection of PID controller gains. In conventional tuning methods a history of response error of the system under control in the passed time is measured and used to adjust PID parameters in order to improve performance of the system in proceeding time. But model predictive control obviates this characteristic of classical PID. In fact MPC tries to tune the controllerby predicting system’s behavior in some time steps ahead. In this way PID parameters are adjusted before any real error occurs in the system response. The whole approach which also uses a GMDH-type neural network modeling is an on-line method of tuning PID controllers. Results mentioned at the end obviously declare that the proposed method surpasses conventional tuning methods for PID controller

کلیدواژه ها:

Model Predictive control ، PID control ، Online PID ، Genetic Algorithms ، GMDH-type NeuralNetworks