To mitigate the impact of singular values in voltage and current signals on regression results and achieve a more accurate delineation of harmonic responsibilities for both the system and users， this paper proposes a method for calculating system harmonic impedance based on partial least squares （PLS） equivalent weight regression. Utilizing harmonic voltage and current data from common connection points， the paper formulates a regression equation containing the sought-after parameters. The PLS is then employed to determine regression coefficients， which correspond to the harmonic impedance and harmonic voltage on the system side. These coefficients are used as initial values for iterative equivalent weight calculations. Then， the user’s harmonic emission level is deduced based on the iterative results. This approach overcomes the drawbacks of using the least squares solution as an initial value for iterative calculations， enhancing the precision of user’s harmonic emission level calculations. The accuracy of the proposed method is validated through simulation analysis in case studies.
Full Text Finder