基于RBFNN优化干扰观测的电机转速控制  

Motor Speed Control Based on RBFNN Optimized Interference Observation

作　　者：颉宏宇 JIE Hongyu(Southwest China Research Institute of Electronic Equipment,Chengdu,Sichuan 610036,China)

机构地区：[1]西南电子设备研究所,四川成都610036

出　　处：《自动化应用》2024年第16期8-11,15,共5页Automation Application

摘　　要：为了有效降低负载转矩波动甚至突变对永磁同步电机转速控制系统运行平稳性的影响,提高系统的抗负载扰动性能,提出了一种基于径向基函数神经网络(RBFNN)优化干扰观测的转速控制方法。该方法首先将负载转矩作为复杂运行工况对系统产生的干扰,基于系统逆设计干扰观测器;其次考虑到手动整定观测器滤波时间常数效率低、定值参数对复杂工况适应性差、抗负载扰动性能欠佳等问题,基于RBFNN自动优化观测器参数,实现负载转矩的精确快速观测;最后基于观测值进行前馈补偿以实现对负载扰动影响的抑制。结果表明,所提方法在提高整定效率的同时可实现负载转矩的精确快速观测,具有良好的抗负载扰动性能,与手动整定参数方式相比,其负载转矩观测速度更快、转速波动更小、响应更快。In order to reduce the effects of random fluctuations or sudden changes of load torque on the speed control performance of permanent magnet synchronous motor effectively and improve the anti-disturbance performance,a speed control method based on the disturbance observer optimized by radial basis function neural networks(RBFNN)is proposed in this paper.Firstly,considering the load torque as the interference of complex work conditions to the system,a disturbance observer based on the inverse model is designed to observe the load torque.Secondly,considering the low efficiency of manual parameter tuning,the poor adaptability and anti-disturbance performance of the fixed time constant,the RBFNN is used to automatically optimize the filter time constant to achieve accurate and fast observation.Finally,the observed value is used as a feedforward compensation term to suppress the influence of load disturbance.The simulation results indicate that the load torque is observed accurately and quickly,and the efficiency of parameter tuning is improved through the proposed method.Meanwhile,the simulation results demonstrate that the proposed method is with better anti-disturbance performance of faster observation,smaller speed fluctuation and faster speed response than the manual parameter tuning.

关 键 词：永磁同步电机 负载转矩 抗扰动性能 径向基函数神经网络 干扰观测器 

分 类 号：TM341[电气工程—电机]