A Unified Approach for Identification and Control of Delta Operator Systems using Neural Networks

Abstract

Delta operator modelling of dynamical systems provides a unified framework for identification and control of continuous time systems from discrete time input–output data. This paper proposes applications of neural network for identification and control of dynamical system modelled with delta operator. Delta operator models are recasted into a few realizable neural network structures using the properties of delta operator. The theoretical framework and analytical methods are presented for identification and control of single input single output systems using the numerical robustness properties of delta operator. Simulation examples are presented to demonstrate that the proposed methods work very well.