永磁同步电机神经网络调速系统的研究

摘要

为改善永磁同步电机在电梯无齿轮曳引复杂工况下的调速动态性能,使其满足快速性、舒适性的同时,对频繁的负载变化有较强的鲁棒性和稳定性,提出将CMAC神经网络控制应用于电梯行业永磁同步电机调速系统中.通过建立永磁同步电机动态模型,搭建内模控制的电流内环及CMAC网络控制的速度外环仿真控制系统,并对比传统PI控制方式进行仿真实验,结果表明,本调速系统在动态性能、抗扰性能及鲁棒性能上均有显著改善;并根据电梯运行加速度要求,跟踪电梯运行的速度抛物线,验证了在CMAC神经网络的学习和记忆能力下有逐渐提高的稳态精度.%In order to improve the dynamic performance of the permanent magnet synchronous motor(PMSM) under the complicated conditions of gearless traction of elevator,while satisfying the performances of rapidity and comfort,and having strong robustness and stability under frequent load changes,the CMAC network control be applied to the PMSM control system in the elevator industry is proposed.The dynamic model of PMSM was established,and the simulation system was built,in which the current inner loop was controlled by the internal mode and the speed outer loop was controlled by the CMAC network,as well as the simulation experiment was compared with the traditional PI control mode.The results show that the dynamic performance,disturbance restraint performance and robust performance of the system are significantly improved.According to the elevator acceleration requirement,the velocity parabola of the elevator operation is tracked,and the conclusion is verified that the steady-state accuracy of this system has been gradually improved in the learning and memory capabilities of the CMAC neural network.