Predictive control of turbofan engine model based on improved elman neural network

Due to the complex and variable working process of aero-engines, and the strong nonlinearity, multi-control variables, time-varying and complex structural features, neural networks have self-learning, adaptive uncertainty system dynamics and approximation of arbitrary complex nonlinear systems. the neural network-based nonlinear predictive control combining the advantages of neural network and predictive control has become an important method to solve the nonlinear system control problem. This paper first introduces the development of nonlinear predictive control, and points out that predictive control is proposed for linear systems, and the control effect of nonlinear systems is often not ideal. Therefore, the local dynamic feedback network Elman neural network with good approximation ability is introduced to identify the nonlinear system. Based on this, the improved Elman neural network is applied in predictive control. Firstly, the neural network is used as the predictive model for multi-step prediction, and the future output value is output. The improved particle swarm optimization algorithm integrated with GuoA algorithm is used as the optimization algorithm to design the predictive controller. The simulation results show that the nonlinear predictive control based on improved Elman neural network is obtained a good control effect.