基于改进FastICA及偏最小二乘法的系统谐波阻抗估计

System Harmonic Impedance Estimation Based on Improved Fast ICA and Partial Least Squares

系统谐波阻抗的精确估计可用于合理区分系统与用户的谐波污染责任。提出一种基于改进快速独立成分分析（Fast ICA）及偏最小二乘法的系统谐波阻抗估计方法。基于公共连接点处的谐波电压和谐波电流数据,采用基于修正的三阶收敛牛顿迭代的Fast ICA算法实现混合信号的分离,相比传统二阶牛顿迭代提高了算法收敛速度;其次,基于解混出的变量,采用偏最小二乘法进行主成分分析求得混合系数,降低了解混变量之间的弱相关性带来的计算误差。最后依据系数间的线性关系计算系统谐波阻抗。仿真和实际案例分析表明,所提方法的系统谐波阻抗估计结果相比现有方法精度更高。

Accurate assessment of system harmonic impedance can be used to distinguish harmonic pollution contributions from system side and customer side. A harmonic impedance assessment method based on improved fast independent component analysis（Fast ICA） and partial least squares（PLS） is proposed. With the harmonic voltage and current data measured at the point of common coupling（PCC）, it adopts Fast ICA algorithm based on modified Newton iteration with the third order convergence to obtain decomposed components to improve iteration speed of conventional Newton iteration with the second order convergence. PLS is used to perform principal component analysis（PCA） to obtain mixed coefficients, decreasing the error due to low correlation between the decomposed components. The system harmonic impedance is assessed according to linearity of the mixed coefficients. Results of simulation and practical case analysis show that assessment accuracy of the system harmonic impedance obtained with the proposed algorithm is higher than that of existing methods.