Distributed modeling and control of large scale systems using neural networks

This paper investigates distributed modeling and control of large scale nonlinear systems with uncertainty using neural network models. Since the neural networks can learn the dynamical system, the authors use the neural networks for modeling and control of large scale systems which consist of interconnected subsystems with nonlinear interaction. The distributed neural network model is composed of several neural networks for subsystems and prediction of interaction. Two neural networks are employed for modeling and control of each subsystem with local identification and control goals, and the interaction prediction neural network is used for prediction of interaction between the subsystems. The learning algorithm is given in detail, and simulation results show that the proposed method is capable of identification and control of large scale systems.