G151052 Design and Application of a Data-Oriented Nonlinear PID Controller
Shin Wakitani, Toru Yamamoto · The Proceedings of Mechanical Engineering Congress Japan · 2013
PID control schemes have been widely used in most industrial control systems.However,it is difficult to determine a suitable set of PID gains,because most industrial systems have nonlinearity.On the other hands,Cerebellar Model Articulation Controller(CMAC) classified as neural networks has been proposed,and design scheme of an intelligent PID controller by using the CMAC-PID tuner has been proposed.Moreover,the conventional CMAC-PID tuner needs an on-line learning to obtain a set of optimum PID gains.However,the issue has been solved by combining with the FRIT which is the method to determine control parameters by using operating data,thus the CMAC-PID tuner can train in an off-line manner using a set of operating data.In this paper,the reduction memory method of the CMAC-PID tuner by using the Group Method of Data Handling(GMDH) is proposed.According to the method,the required memory to configure the controller is reduced and the algorithm can be added to the low memory capacity computer like a micro controller.The effectiveness of the proposed method is evaluated by applying to a magnetic levitation device.