At present, the power wagon is confronted with several problems: small brake torque at low speeds, output instability of random loading and slow response speed. Aiming at the existing problems of current control performance for loading system of power wagon, the paper proposes a back propagation (BP) neural net PID algorithm. The dynamic analysis of power wagon loading system was presented. A mathematical model of the loading system is established by use of the classical control theory. Through simulation and analysis, the dynamic response of the loading system is researched by using step signal, sine signal and field-load spectrum stochastic signal as system input respectively. Meanwhile, through MATLAB simulation experiment, compares and analysis with superiorities and characteristics between BP neural net PID control method and traditional PID control method. On the basis of this study, the power wagon was modified based on YTO-LX1304. The road test of the YTO-MF554 tractor traction property was performed on the concrete by using the power wagon. Experimental results show that the system output traction has a good follow effect in comparison with input load. The BP neural net PID algorithm can improve the system’s dynamical performance and its static accuracy in comparison with the traditional PID control. The simulation results show that system delay time is 0.12s and maximum overshoot is 3.8%. Road test results show that system delay time is 0.22s and maximum overshoot is 4.2%. The BP neural net PID algorithm has better control performance for second-order plant with time-delay time-varying system.

This content is only available via PDF.
You do not currently have access to this content.