The study of application of FCMAC neural network in the industrial process on-line identification and optimization

In this paper, a kind of online identification and optimization method based on FCMAC (fuzzy cerebellar model arithmetic controller) neural network is proposed. This method is composed of learning procedure and optimization procedure. For some practical applications without process models, only observation of sensor information is needed. The simulation results prove the effectivity and superiority of this method, thereby offering a new frame for implementing optimal schemes in practical applications.