| 引用本文: | 刘子腾,徐永海,陶顺.基于SHIBSS方法和数据优选的系统侧谐波阻抗估算方法[J].电力自动化设备,2021,41(2): |
| LIU Ziteng,XU Yonghai,TAO Shun.Estimation method of harmonic impedance on system side based on SHIBSS method and data optimization[J].Electric Power Automation Equipment,2021,41(2): |
|
| 摘要: |
| 针对盲源分离类方法和传统数据筛选方法的不足，提出一种基于盲源分离的移位阻断(SHIBSS)方法和数据优选的系统谐波阻抗估计方法，对公共连接点处的谐波电压和谐波电流的采样数据采用SHIBSS方法分离出系统侧和用户侧谐波发射电流并计算谐波阻抗；针对谐波阻抗筛选步骤，证实了传统数据筛选方法具有失效率高的问题，提出依据数据优选法辨识系统侧谐波阻抗最优解的方法。仿真分析结果表明，与波动量法、线性回归类常用谐波阻抗估计方法及盲源分离类方法中应用较广泛的快速独立分量分析法相比，SHIBSS方法具有更高的估算精度，数据优选方法克服了传统数据筛选法存在的高失效率的问题，所提系统侧谐波阻抗估算方法具有较高的估计精度。 |
| 关键词: 谐波 谐波阻抗 盲源分离 SHIBSS 数据筛选 |
| DOI：10.16081/j.epae.202011019 |
| 分类号:TM711 |
| 基金项目:国家自然科学基金资助项目(51777066) |
|
| Estimation method of harmonic impedance on system side based on SHIBSS method and data optimization |
|
LIU Ziteng, XU Yonghai, TAO Shun
|
|
State Key Laboratory of Alternate Electrical Power System with Renewable Energy Sources, North China Electric Power University, Beijing 102206, China
|
| Abstract: |
| Aiming at the disadvantages of blind source separation and the traditional data filtering methods, an estimation method of harmonic impedance on system side based on SHIBSS(SHIfted blocks for Blind Source Separation) method and data optimization is proposed. Firstly, SHIBSS method is used to separate the harmonic emission current of system side and consumer side from the sampling data of harmonic voltage and harmonic current of PCC(Point of Common Coupling) to calculate the harmonic impedance. Then, the problem of high failure rate of traditional data filtering methods in data filtering process is proved, and a method for selecting the optimal solution of harmonic impedance on system side based on data optimization is proposed. The results of simulation analysis show that, compared with the fluctuation quantity method, common harmonic impedance method based on linear regression and commonly used FastICA(Fast Independent Component Algorithm) method in blind source separation methods, the SHIBSS method has higher estimation accuracy, the problem of high failure rate of traditional data filtering methods is overcome by the proposed data optimization method and the proposed estimation method of harmonic impedance on system side has high estimation precision. |
| Key words: harmonics harmonic impedance blind source separation SHIBSS data filtering |
