Model predictive control of nonlinear hybrid system based on neural network optimization

This paper presents Model predictive control (MPC) of nonlinear hybrid system based on neural network (NN) optimization. Multiple model method is used to modeling of nonlinear hybrid system and these models are combined using Bayes theorem. NN optimization combined gradient NN with recurrent NN is proposed to solve optimization problem of each sample time in MPC. An example of benchmark three spherical tank system demonstrates the effectiveness and efficient of the proposed recurrent neural network based MPC. Simulation results show that this approach can utilize fast converge property and the parallel computation ability of NN and be applied to real-time industrial process control.