基于参数跟踪的感应电机转速与磁链强跟踪滤波估计方法

Strong Tracking Filter Estimation Method for Speed and Flux of Induction MotorBased on Parameter Tracking

针对感应电机高性能矢量控制中的转速与磁链估计问题,提出了一种包含参数跟踪的转速与磁链联合估计强跟踪滤波(STF)方法。考虑电机参数变化及模型不确定性,采用粒子群迭代学习动态优化算法实现对感应电机参数的在线跟踪,修正的STF算法用于实现对感应电机转速与转子磁链的联合估计。仿真结果表明,包含参数跟踪的STF算法能够有效实现对感应电机转速与转子磁链的高性能估计;与扩展卡尔曼滤波算法相比,包含参数跟踪的STF算法在估计精度、跟踪速度和稳定性方面得到大幅度提高,并且能够快速跟踪突变状态,尤其在低速段仍能保持较好的估计性能,有效提高了状态估计精度和模型鲁棒性。

According to the speed and flux estimation of induction motor in high-performance field-oriented vector control,a strong tracking filter(STF)method for speed and flux joint estimation with parameter tracking was proposed.Considering time-varying parameters and model uncertainties of induction motor,the particle swarm iterative learning dynamic optimization algorithm was used to realize online parameters tracking of induction motor,and the modified STF algorithm was used to realize speed and flux joint estimation of induction motor.Simulation results show that STF algorithm with parameter tracking can effectively estimate the speed and rotor flux of induction motor accurately.STF algorithm with parameter tracking is better than extended Kalman filter(EKF)on the estimating accuracy,tracking speed and stability,and it can quickly track jumping state,especially in the lowspeed,and the state estimating accuracy and robustness against model uncertainties are improved effectively.