Nonlinear model based predictive control using multiple local models

Abstract

The chapter describes nonlinear generalised predictive control (GPC) where the internal linear plant model is replaced by a local model (LM) neural net representation, which represents the process by a set of locally valid and simpler submodels. The LM network structure, interpretation and training are considered. The network was constructed from local autoregressive with external input (ARX) models and trained using hybrid learning. Two methods of exploiting the LM network for long-range, nonlinear model predictive control are described. One consists of a network of predictive controllers, each designed around one local models. The output of each controller is passed through a validity function and summed to form the plant input. The other uses a single predictive controller, which extracts a local model from the net to represent the process at each controller sample instant. Simulation studies for pH neutralisation show excellent nonlinear modelling characteristics. Both nonlinear model predictive controllers gave excellent tracking and disturbance rejection results and improved performance compared with conventional linear GPC.