基于BP神经网络的异步电机转子磁链定向矢量复合控制
|
摘要
针对异步电机在传统PI控制策略中,参数固定且容易超调,IP控制策略跟踪响应慢的特点,提出神经网络矢量PI-IP复合控制策略。在转子磁链定向矢量控制中,转速环引入新型的PI-IP控制器,同时利用神经网络进行参数整定,综合PI-IP控制器,神经网络以及转子磁链定向矢量控制的优点,完成神经网络矢量PI-IP控制器对异步电机的优化控制,提高系统的动、静态性能。最后基于仿真效验,仿真结果表明,与常规控制方法相比,该方法能有效提高控制精准性,抑制扰动,神经网络矢量PI-IP复合控制器具有更强的稳态精度和应对负载扰动能力。
In view of the characteristics of the asynchronous motor in the traditional PI control strategy, the parameters are fixed and easy to overshoot, and the response of the IP control strategy is slow, the neural network vector PI-IP compound control strategy is proposed. In the directional vector control of rotor flux linkage, the speed loop introduces a new PI-IP controller. At the same time, the parameter tuning of the neural network is used, the advantages of the PI-IP controller, the neural network and the rotor flux oriented vector control are integrated, and the optimal control of the neural network vector PI-IP controller to the asynchronous motor is completed and the system movement is improved. Static performance. The simulation results are based on the simulation results. The simulation results show that, compared with the conventional control method, the proposed method can effectively improve the precision of control and suppress the disturbance. The neural network vector PI-IP composite controller has a stronger steady state precision and the ability to respond to the load disturbance.
In view of the characteristics of the asynchronous motor in the traditional PI control strategy, the parameters are fixed and easy to overshoot, and the response of the IP control strategy is slow, the neural network vector PI-IP compound control strategy is proposed. In the directional vector control of rotor flux linkage, the speed loop introduces a new PI-IP controller. At the same time, the parameter tuning of the neural network is used, the advantages of the PI-IP controller, the neural network and the rotor flux oriented vector control are integrated, and the optimal control of the neural network vector PI-IP controller to the asynchronous motor is completed and the system movement is improved. Static performance. The simulation results are based on the simulation results. The simulation results show that, compared with the conventional control method, the proposed method can effectively improve the precision of control and suppress the disturbance. The neural network vector PI-IP composite controller has a stronger steady state precision and the ability to respond to the load disturbance.
引文
[1] Blaschke F.The principle of field orientation as applied to the new TRANSVECTOR-closed loop control systems for rotating field machines[J].Power Electronics,2004,2(1):39.
[2] Bojoi R,Levi E,Farina F,et al.Dual three-phase induction motor drive with digital current control in the stationary reference frame[J].Power Engineer,2006,153(1):129-139.
[3] 郭新华,温旭辉,赵峰,等.基于电磁转矩反馈补偿的永磁同步电机新型IP速度控制器[J].中国电机工程学报,2010,30(27):7-13.
[4] 陈鹏展,唐小琦,任清荣.基于IITAE评价指标的交流伺服系统参数自动整定研究[J].微电机,2010,43(2):70-73.
[5] 邵珠荣,王宁.基于IP控制和扩展卡尔曼滤波的PMSM转速控制[J].大连海事大学学报,2016,42(1):90-95.
[6] 付立华,王刚,张晓玫.基于神经网络PID的三相异步电机矢量控制仿真研究[J].煤矿机械,2015,36(1):63-66.
[7] 周佳,卢少武,周凤星.基于RBF神经网络的永磁同步电机速度PI-IP控制[J].组合机床与自动化加工技术,2017(1):116-118.
[8] Rojas C A,Rodriguez J,Villarroel F,et al.Predictive Torque and Flux Control Without Weighting Factors[J].IEEE Transactions on Industrial Electronics,2013,60(2):681-690.
[9] 曹朋朋,张兴,杨淑英,等.异步电机基于MRAC的转子时间常数在线辨识算法的统一描述[J].电工技术学报,2017,32(19):62-70.
[10] 刘洪玮.遗传算法在模糊PID交流电机矢量控制系统中的应用[J].工业仪表与自动化装置,2018(1):120-123.
[11]王阔厅.BP神经网络在船舶推进电机故障预测中的应用[J].中国科技信息,2014(z2):30-32.
[12] 鲁建东,鲁啸.基于BP神经网络的煤矿电机故障诊断预测[J].机电设备,2013(5):17-20.
[13] 龙泉,鲁志平,张羽.基于BP神经网络的多变量风力发电机绕组温度预测[C].中国电机工程学会年会,2016.
[14] 王新,候风艳.基于改进的PSO-BP神经网络的无刷直流电机控制[J].电子测量技术,2017,40(2):10-14.
[2] Bojoi R,Levi E,Farina F,et al.Dual three-phase induction motor drive with digital current control in the stationary reference frame[J].Power Engineer,2006,153(1):129-139.
[3] 郭新华,温旭辉,赵峰,等.基于电磁转矩反馈补偿的永磁同步电机新型IP速度控制器[J].中国电机工程学报,2010,30(27):7-13.
[4] 陈鹏展,唐小琦,任清荣.基于IITAE评价指标的交流伺服系统参数自动整定研究[J].微电机,2010,43(2):70-73.
[5] 邵珠荣,王宁.基于IP控制和扩展卡尔曼滤波的PMSM转速控制[J].大连海事大学学报,2016,42(1):90-95.
[6] 付立华,王刚,张晓玫.基于神经网络PID的三相异步电机矢量控制仿真研究[J].煤矿机械,2015,36(1):63-66.
[7] 周佳,卢少武,周凤星.基于RBF神经网络的永磁同步电机速度PI-IP控制[J].组合机床与自动化加工技术,2017(1):116-118.
[8] Rojas C A,Rodriguez J,Villarroel F,et al.Predictive Torque and Flux Control Without Weighting Factors[J].IEEE Transactions on Industrial Electronics,2013,60(2):681-690.
[9] 曹朋朋,张兴,杨淑英,等.异步电机基于MRAC的转子时间常数在线辨识算法的统一描述[J].电工技术学报,2017,32(19):62-70.
[10] 刘洪玮.遗传算法在模糊PID交流电机矢量控制系统中的应用[J].工业仪表与自动化装置,2018(1):120-123.
[11]王阔厅.BP神经网络在船舶推进电机故障预测中的应用[J].中国科技信息,2014(z2):30-32.
[12] 鲁建东,鲁啸.基于BP神经网络的煤矿电机故障诊断预测[J].机电设备,2013(5):17-20.
[13] 龙泉,鲁志平,张羽.基于BP神经网络的多变量风力发电机绕组温度预测[C].中国电机工程学会年会,2016.
[14] 王新,候风艳.基于改进的PSO-BP神经网络的无刷直流电机控制[J].电子测量技术,2017,40(2):10-14.