Aircraft Engine Based On Neural Network Pid Control

Aero engine is complicated and nonlinear control system. For such controlled object, a lot of control theories are limited. Neural networks can be thought as a kind of mathematic tool which is independent on model. Neural networks have learning and strongly adaptive capability and are suitable for aero engine which have uncertainty and non-linearity. In this paper, the turbo-fan engine control system is established on the basis of the combination of neural network and PID.First, the one-variable control system was designed. The weights of the neural network PID controller were adjusted based on RBF Neural Network identification. The identifier and controller were used grads descent method. The principle and configuration of identifier were given. Simulation of the control system was performed, then excellent tracking performance and robustness was obtained. The system was suitable for aero-engine control.Next, on the basis of the one-variable control system, to solve the problem of coupling in turbofan engine double-variable control system, double variable PID decoupling control for turbo-fan engine method based on RBF neural network identification was presented in this paper. The structure of engine decoupling control system, as well as its decoupling principle, was given. Simulation of the control system was performed, and then excellent tracking performance and robustness were obtained. The simulation results show that the method can effectively reduce the coupling influence of each control loop and assure satisfactory transient performance.Finally, using neural network control and PID control, the accelerated control system of engine was designed. The C++ engine nonlinear model program was done as S-function. Simulation of the control system was performed, the result indicated that neural network control was suitable for engine transient state control.