PI Control Of Distributed Delay System And Real-time Simulation Verification For Aero-Engines

The aero-engine distributed control system is a complex system with great potential.It has great advantages and also brings new problems and difficulties such as system modeling,network-induced delay,and packet loss.In this paper,we explore the method of establishing real-time simulation model of distributed control system and carry out research on control method for time-delay system based on a turbofan engine.Taking the aero-engine component-level model as the object,the Matlab/True Time box is used to establish node model for each sensor,controller and actuator.According to the needs of different component,we initialize each node and establish tasks for each node.By setting the network characteristic,we finally obtain a real-time simulation model of the aero-engine distributed control system.This model is the basic of the stability analysis and simulation verification.The current aero engine still using the traditional PI control.Considering the problems of network-induced delay and packet loss in distributed control system,we search the upper bound of system delay based on the established model by considering packet loss into delay.Using the transfer function model of turbofan engine,the stability domain ofK_p is obtained based on Nyquist stability criterion and delay.Then the equivalent open-loop transfer function and the stability region ofK_pare used to find the stability region ofK_i.Finally,we obtain the PI controller parameter stability domain of system with time.Based on the established control system model,the integral criterion in multiplication of time by absolute error is selected as the performance index.And the simplex method is used to search the controller parameters for the speed and pressure ratio circuits in the stability region.The simulation results show that the optimized control system has better dynamic performance and steady-state accuracy.