Probability Distribution Approach of EMI Filter Performance for Photovoltaic Applications

This paper proposes a probability distribution approach to investigate the uncertainty of Electromagnetic Interference (EMI) filter performance based on filters used for PV applications. In general, a Line Impedance Stabilization Network (LISN)is utilized to measure conducted EMI. Typically, the filter performance is measured in a $50\ \Omega\ /\ 50\ \Omega$-System or a $0.1\ \Omega\ / 100 \ \Omega$ or $100\ \Omega\ /\ 0.1\ \Omega$ system (see CISPR 17). In real power installations, the impedances vary significantly from this value. Empirical data had been derived and published in several papers. Based on that, Monte-Carlo simulations have been performed in order to determine the stray of a margin below the disturbance voltage limits $U_{cispr}$ of test laboratory voltage $U_{lab}$ to take into account the parameter tolerance and measurement uncertainties. Based on statistical results for a representative PV application, the tolerance intervals used to recommend the performance of an EMI filter design to satisfy EMC standards.