带有扰动补偿的永磁同步电机自适应预测控制

摘要

为保证永磁同步电机(PMSM)控制系统的稳定性和鲁棒性,提出一种基于小波神经网络的自适应Smith预测控制器(WNNSPC),通过对模型误差的补偿,消除模型不确定性的影响,保证了预测输出的准确性,同时针对电机运行中突加负载等的扰动,采用小波神经网络进行前馈补偿,有效抑制了各类扰动对系统的影响,而且具有良好的动、静态品质.仿真结果证明了提出方法的有效性.%An adaptive Smith predictive controller based on wavelet neural network ( WNN) was proposed to guarantee the stability and robustness of the PMSM control system. It could eliminate the influence of model uncertainties by compensating the model error and make a accurate predictive output. Feed forward compensation using WNN was also utilized to reject kind of disturbances like a sudden load, which effectively restrains the affect of disturbances and provides good static and dynamic performance. The simulation results proved the validity of the proposed control scheme.