Assessing the Harmonic Impedance Based on Least Squares Support Vector Machine
Yankun Xia, Wenzhang Tang, Xinyi Lin · 2021 4th International Conference on Energy, Electrical and Power Engineering (CEEPE) · 2021
This paper presents a new method based on least square support vector machine regression for estimating harmonic impedance. The regression model is built by using least squares support vector machines, Lagrange multiplier optimization model is introduced to get Lagrange function, and Lagrange function is solved to get model parameters. The harmonic voltage and current signals measured at PCC are substituted into the model of least squares support vector machine to estimate the harmonic impedance of the system. The results of simulation analysis of equivalent circuit show the accuracy of the method of LSSVM and error analysis proves that the method is more robust and accurate than other methods such as binary linear regression and SVM. Finally, the effective-ness of this method has been proved through the simulation results of equivalent circuit and the analysis results of engineering measured data, and the results are compared with those of other harmonic impedance estimation methods.