Predictive Control of Nonlinear System Based on Neural Networks

Model predictive control (MPC) is an important industrial control technique. Most conventional MPC schemes use linear models. However, the use of linear models can result in a serious deterioration of control performance with many types of nonlinear plants. Feedback linearisation is an important nonlinear control technique which can transform a nonlinear system into a linear system. Dynamic neural networks have the ability to approximate multi-input multi-output general nonlinear systems and have the differential equation structure. This book presents a hybrid control strategy integrating dynamic neural networks and feedback linearisation into a predictive control scheme. This book can be used as a course textbook, a source for practising control engineers with an interest in nonlinear control techniques and also a reference material for academic researchers in nonlinear control theory.

Autorentext

Dr. Jiamei Deng is an internationally established researcher, who is currently a Lecturer in Loughborough University in the United Kingdom. Dr. Deng received her Ph.D. degree in Cybernetics from the University of Reading in 2005. She was an Associate Professor in University of Shanghai between 1998 and 2002.