交流电机直接转矩控制系统新型闭环磁链观测器研究  

Research on a Novel Closed-loop Flux Observer for Direct Torque Control System of AC Motors

作　　者：侯湘庆 刘珊[1] Hou Xiangqing;Liu Shan(Hunnan Electrical College of Technology,Xiangtan,Hunan 411101,China)

机构地区：[1]湖南电气职业技术学院,湖南湘潭411101

出　　处：《绿色科技》2023年第14期230-234,251,共6页Journal of Green Science and Technology

基　　金：2021年湖南省教育厅科学研究项目(编号:21C1015);2020年湖南省自然科学基金项目(编号:2020JJ7021);2021年湖南省教育厅科学研究项目(编号:21C1016)。

摘　　要：针对电机参数变化与负载扰动影响电机磁链辨识结果的问题,提出了一种新型的闭环磁链观测器,通过DTC系统采样数据进行离线训练,构建了BP神经网络逆磁链模型,与电机磁链模型构成“可调整系统”,以实际电机作为参考模型,建立了具有自适应能力的新型闭环磁链观测器。新型闭环磁链观测器应用于直接转矩控制(DTC)系统中,通过电机参数变化与负载扰动仿真结果表明:系统性能优于传统DTC系统,磁链辨识精度高、转矩脉动小、转速波动小,具有较强的鲁棒性和控制精度。In response to the problem of the impact of motor parameter changes and load disturbances on the motor flux identification results,this paper proposes a new closed-loop flux observer.The observer uses a DTC system to sample data for offline training,constructs a BP neural network inverse flux model,and forms an"adjustable system"with the motor flux model.By using the actual motor as a reference model,a new type of closed-loop flux observer with adaptive capability is established.The new closed-loop flux observer is applied to the DTC system,and simulation results of motor parameter changes and load disturbances show that the system performance is better than that of traditional DTC systems.The flux identification accuracy is high;the torque ripple and speed fluctuation are small with strong robustness and control accuracy.

关 键 词：闭环磁链观测器 逆磁链模型 BP神经网络 直接转矩控制 

分 类 号：TM343[电气工程—电机]