基于IPSO-BP算法的燃油系统气压控制优化

Optimization of Gas Pressure Control of Fuel System Based on IPSO-BP Algorithm

针对燃油系统地面模拟试验气箱压力控制所存在的非线性、时变性问题,提出了一种基于改进粒子群算法(IPSO)优化的BP-PID控制器。该控制器结合改进粒子群算法全局搜索能力强和收敛速度快的特点,对BP神经网络的连接权值不断地调整,输出最优的PID控制参数。依据地面模拟试验参数,建立了主要器件的数学模型,利用Simulink软件对各器件的仿真模块进行搭建,构成一个完整的气箱控制系统进行研究。结果表明,IPSO-BP-PID控制比BP-PID控制响应速度快,稳定性好,超调量小,大大提高了气箱压力控制过程的精确性与鲁棒性。

Aiming at the non-linear and time-varying problems existed in the gas tank pressure control of fuel system analogue experiment on ground,a optimized BP-PID controller based on improved particle swarm optimization(IPSO)algorithm is proposed.The controller constantly adjusts the connection weights of BP neutral network with the characteristics of strong global search ability and fast convergence by IPSO algorithm,and outputs the optimal parameters of PID control.According to the parameters of analogue experiment on ground,the mathematical models of the main components are established,Simulink software is used to build the simulation modules of each component and form a complete gas tank control system to research.The results show that IPSO-BP-PID control has faster response speed,better stability and smaller overshoot than BP-PID control,the controller can greatly improve the accuracy,robustness in the gas tank pressure control.