Partial Least-Squares Regression Based Harmonic Emission Level Assessing at the Point of Common Coupling

This paper proposes a new method, called PLSR (partial least-squares regression), which is used for assessing harmonic impedance and the harmonic emission level at the point of common coupling (PCC). Utilizing the signals of harmonic voltage and harmonic current measured synchronously at PCC, regression coefficients are worked out through partial least-squares algorithm. By separating and filtering the measured data, the comprehensive variables that can explain dependent variables most effectively are distilled, and then the useful information and noise are identified from the measured data. Therefore, the disadvantage of variable dependence in establishing of the system model is overcome. After that, the supply harmonic impedance is got, consequently the harmonic emission level of customer is calculated. The simulation analysis and the analytic result of practical engineering road-testing data indicate that the proposed method is valid and feasible.