电工技术学报  2024, Vol. 39 Issue (22): 7099-7110    DOI: 10.19595/j.cnki.1000-6753.tces.231617
电机及其系统 |
基于在线参数辨识及自适应谐波提取滤波器的改进型直线振荡电机无位置传感器控制
葛健1, 宫逸凡1, 徐伟2,3, 廖凯举2,3, 苏诗湖4
1.强电磁技术全国重点实验室(华中科技大学) 武汉 430074;
2.高密度电磁动力与系统重点实验室(中国科学院电工研究所) 北京 100190;
3.中国科学院大学 北京 100049;
4.中车株洲电机有限公司 株洲 412001
Improved Sensorless Control of Linear Oscillatory Machine Based on Online Parameter Identification and Adaptive Harmonic Extraction Filter
Ge Jian1, Gong Yifan1, Xu Wei2,3, Liao Kaiju2,3, Su Shihu4
1. State Key Laboratory of Advanced Electromagnetic Technology Huazhong University of Science and Technology Wuhan 430074 China;
2. Key Laboratory of High Density Electromagnetic Power and Systems Institute of Electrical Engineering Chinese Academy of Sciences Beijing 100190 China;
3. University of Chinese Academy of Sciences Beijing 100049 China;
4. CRRC Zhuzhou Motor Co. Ltd Zhuzhou 412001 China
全文: PDF (2803 KB)   HTML
输出: BibTeX | EndNote (RIS)      
摘要 传统直线振荡电机(LOM)无位置传感器控制方法存在参数敏感性差、行程观测精度低等问题,进而会降低电机运行的可靠性。为减小预设参数不准确及参数非线性变化对行程观测精度的影响,该文提出一种基于在线参数辨识及自适应谐波提取滤波器的改进型直线振荡电机无位置传感器控制策略。首先,采用频域法分析了直线振荡电机行程观测方法的参数敏感性,明确了影响行程观测精度的主要电机参数。然后,根据直线振荡电机数学模型提出了基于高频电压注入法的在线参数辨识方案,并设计了一种基于自适应谐波提取滤波器的高频谐波提取方法。最后,通过大量仿真和实验,验证所提方法的有效性和先进性。
服务
把本文推荐给朋友
加入我的书架
加入引用管理器
E-mail Alert
RSS
作者相关文章
葛健
宫逸凡
徐伟
廖凯举
苏诗湖
关键词 直线振荡电机参数辨识高频电压注入无位置传感器控制    
Abstract:Sensorless control is an important control strategy for linear oscillatory machines (LOM). The traditional position observer often has low accuracy and poor anti-interference ability. The deviation of stroke observation affects the closed-loop control of piston stroke and resonance frequency tracking control of LOM. Therefore, this paper provides an improved sensorless control method for LOM based on online parameter identification and adaptive harmonic extraction filters.
Firstly, the amplitude-frequency characteristic is applied to analyze parameter sensitivity for LOM. The impacts of different circuit parameters on stroke observation are analyzed. It is proved that the inductance parameter has a more prominent impact on stroke observation. Therefore, the main object of the parameter identification algorithm is the inductance parameter. Secondly, a parameter identification strategy based on a high-frequency injection method is designed. The influence of high-frequency voltage excitation on LOM is studied by analyzing the forced vibration system, and the feasibility of the algorithm in online parameter identification is demonstrated. Then, a new adaptive harmonic extraction filter is designed to solve the harmonic extraction problem, and the fundamental signal can be effectively decayed. Furthermore, an amplitude calculation method based on an orthogonal generator is designed to accelerate data update speed and improve stability, which can obtain accurate piston amplitude information. Finally, the simulation and experimental results show that this method can accurately identify the motor parameters, thereby improving the accuracy of stroke observation.
The following conclusions can be drawn. (1) The impact of a 5% deviation in inductance parameters on stroke observation is much greater than the 10% deviation in resistance parameters, indicating that the accuracy of the stroke observation algorithm is mainly affected by the inductance parameter. (2) Injecting high-frequency voltage signals during LOM operation has little impact on the movement of the piston. (3) The designed adaptive harmonic extraction filter can extract harmonic signals with smaller amplitudes. The extraction of harmonic signals is more accurate and stable than second-order generalized integrators. (4) Compared with the traditional stroke observation method, the accuracy of the stroke observation after parameter identification and correction is significantly improved.
In conclusion, the proposed stroke observation algorithm solves the traditional algorithm's high parameter sensitivity problem. It can obtain accurate stroke observation results even in the case of inaccurate parameters and dynamic changes in parameters.
Key wordsLinear oscillatory machine (LOM)    parameter identification    high-frequency voltage injection    sensorless control   
收稿日期: 2023-10-06     
PACS: TM32  
基金资助:国家自然科学基金面上项目(52277050)、深圳市协同创新计划国际科技合作项目(GJHZ20210705142539007)资助
通讯作者: 徐 伟 男,1980年生,教授,博士生导师,研究方向为直线电机设计及控制。E-mail: weixu@hust.edu.com   
作者简介: 葛 健 男,1994年生,博士,研究方向为新型直线电机电磁设计与特性分析。E-mail: gejian1994@hust.edu.cn
引用本文:   
葛健, 宫逸凡, 徐伟, 廖凯举, 苏诗湖. 基于在线参数辨识及自适应谐波提取滤波器的改进型直线振荡电机无位置传感器控制[J]. 电工技术学报, 2024, 39(22): 7099-7110. Ge Jian, Gong Yifan, Xu Wei, Liao Kaiju, Su Shihu. Improved Sensorless Control of Linear Oscillatory Machine Based on Online Parameter Identification and Adaptive Harmonic Extraction Filter. Transactions of China Electrotechnical Society, 2024, 39(22): 7099-7110.
链接本文:  
https://dgjsxb.ces-transaction.com/CN/10.19595/j.cnki.1000-6753.tces.231617          https://dgjsxb.ces-transaction.com/CN/Y2024/V39/I22/7099