A statistical method for predicting the net harmonic currents generated by a concentration of electric vehicle battery chargers

This paper presents a method for predicting the net harmonic currents produced by a large number of electric vehicle (EV) battery chargers. The problem is stochastically formulated in order to account for randomness in individual charger start-time and battery state-of-charge. We introduce a model that allows for partial harmonics cancellation due to diversity in magnitudes and phase angles. A general solution technique is presented along with an example using data -from a commercially available EV charger. Our results show that a limiting distribution of 7-10 chargers is adequate for accurately predicting harmonic injection currents using the Central Limit Theorem. We also show that the expected values of net harmonic currents are considerably less than the peak values that would have been realized if the same number of chargers were operated in unison.