Research On Transient Air-Fuel Ratio Control Of Aviation Piston Engine

With the development of aero engine such as unmanned aerial vehicle,as its power plant the aviation piston engine has been more considerable.As can be seen,whether the air-fuel ratio control is accurate or not has a significant influence on fuel economy and power performance,as well as flight safety and reliability of aero vehicle,hence,the research of the air-fuel ratio control under transient condition of aviation piston engine is of great importance.In this paper,a variation particle swarm optimization(VPSO)has been proposed by analyzing the theoretical of particle swarm optimization algorithm.The result of performance test shows that the VPSO can maintain diversity of population well during searching process,and the VPSO performance is significantly improved in convergence precision and convergence speed.Besides,the VPSO is used to optimize Elman Neural Network to obtain a VPSO-Elman prediction model which is used in identifying air-fuel ratio.The control of transient air-fuel ratio of aero-engine has some problems such as strong nonlinearity,time delay,uncertainty.In order to control transient air-fuel ratio of engine accurately,a model predictive control strategy based on particle swarm optimization algorithm and Elman neural network is proposed.This strategy combines the adaptive ability and the characteristics of the nonlinear mapping with arbitrary accuracy of neural network,and the advantages of the model predictive control for dealing with uncertain system.An engine model being used in simulation has been established in AMESim.And the model has been demarcated by experimental data.A model predictive control system has been established in MATLAB/Simulink.Coupling simulation is carried out to test the effect of algorithm of model prediction control.The simulation result shows that the differentiation degree of the predict model of air-fuel ratio in this paper is sufficiently high,and the average error is much little.And the differentiation degree can meet the requirement of the predicted air-fuel ratio.Compared with PID control,model predictive control can achieve more accurate control of air-fuel ratio,and it significantly improves the overshoot and the callback time of air-fuel ratio.It proves that used to control the air-fuel ratio,model predictive control is more effective and superior.Under the acceleration and deceleration condition,for different changing rate of the throttle opening,the control effectiveness of air-fuel ratio turns out to be more ideal when the changing rate of the throttle opening tends to be slow.