基于LTS初值的稳健回归的谐波发射水平评估方法

A method for assessing harmonic emission level based on robust regression of LTS initial value

提出了一种基于最小截尾二乘(LTS)的稳健回归的谐波发射水平评估方法。在系统公共连接点测量电压与电流,经过傅立叶分析,利用最小截尾二乘算法估计系统谐波阻抗的迭代初始值。通过反复加权迭代计算最终的系统谐波阻抗估值。利用系统谐波阻抗评估公共连接点两侧的谐波发射水平,从而衡量谐波源对电力系统的谐波危害程度。最后,通过仿真分析验证了该方法的改进效果。与普通稳健回归法相比,所提出的改进方法具有较稳健的迭代初始值,可以有效减小异常值的影响,且具有更强的背景谐波波动适应能力。

This paper presents a new method for assessing harmonic emission level, which is based on robust regression method of the LTS initial value. Utilizing Fourier to analyze the data of voltage and current measured at PCC（point of common coupling）. By using least trimmed squares algorithm, the iterative initial value of system harmonic impedance is worked out. The final value of system harmonic impedance can be got by iteratively reweighted least squares algorithm. The system harmonic impedance is used to evaluate the harmonic emission level of the both side of PCC, consequently the harmonic pollution of power system can be quantified. Finally, the simulation analysis verifies the improvement effect of this method. Compared with ordinary robust regression method, the improved method has a robust iterative initial value, can effectively reduce the influence of the outliers, and has stronger ability to adapt the fluctuation of background harmonic.