基于时序聚类及改进Pettitt的系统谐波阻抗估计

Utility Harmonic Impedance Estimation Based on Time Series Clustering and Improved Pettitt Method

针对谐波电压与电流数据出现严重无序性,导致常规非干预式方法难以对系统谐波阻抗进行准确的实时估计,提出一种基于时序聚类及改进Pettitt的系统谐波阻抗估计方法。首先,运用时序聚类中互相关系数原理使谐波电压和电流序列平移对齐并筛选出强相关性的数据。然后,为了规避阻抗突变的影响,提出基于二元分割的Pettitt方法对阻抗突变点进行检验。最后,使用再生权最小二乘法,通过赋权重系数进一步减小异常值对结果的影响。仿真与实例结果表明,本文方法可以有效降低背景谐波波动和阻抗突变对系统谐波阻抗估计带来的误差。

Aimed at the serious disorder in harmonic voltage and current data,which makes it difficult to estimate the utility harmonic impedance accurately in real time using conventional non-interventional methods,a utility harmonic impedance estimation method based on time series clustering and an improved Pettitt method is proposed.First,the cross-correlation coefficient principte in time series clustering is applied to shift and align harmonic voltage and current sequences and further select the data with a strong correlation.Second,to avoid the effect of impedance mutation,the Pettitt method based on binary partitioning is proposed to test the impedance mutation points.Finally,the self-born weighted least squares method is used to further reduce the effect of outliers on results by assigning weighting coeffi⁃cients.The simulation results and an example show that the proposed method can effectively reduce the errors in utility harmonic impedance estimation brought by background harmonic fluctuations and impedance mutations.