ASSESSING THE HARMONIC EMISSION LEVEL BASED ON ROBUST REGRESSION METHOD

This paper presents a new method for assessing the harmonic emission level, which based on the robust regression. By utilizing the signals of harmonic voltage and harmonic current which have been measured synchronously at PCC, it applies an iteratively reweighted least squares algorithm to solve regression coefficient: The idea is to use Huber function as the influence function and to compute iteratively with weights. The weights are functions of the residuals from the previous iteration such that the estimator is not influenced by data with error. Then the harmonic impedance of system can be got, consequently the harmonic emission level of customer can be calculated. Through the simulation analysis of laboratorial circuit and the analytic result of practical engineering road-testing data, the method proved its validity.