Speed-Regulation System of PMLSM Based on BP Neural Network PID Control
For the characteristics of permanent magnet linear synchronous motor (PMLSM) hoisting system’s nonlinear, time-varying and vulnerable to disturbance, based on the established PMLSM d-q axis dynamic model, designed of an improved BP neural network PID speed controller. Modified the fixed learning rate in BP neural network to adaptive adjustable, and added the momentum to reduce the oscillation tendency in the learning process, greatly improved the convergence speed and avoided the network into a local minimum. Compared with the simulations of traditional PID and the improved BP network PID speed controller, the results showed that the improved BP neural network PID speed controller had the high quality and it can make the system with better dynamic performance and robustness.