Power loss and harmonic minimizanon at distribution system with electric vehicle by passive filter using modified lightning search algorithm

Abstract

The drastic rise in the usage of electric vehicles (EV) in recent years has negatively impacted the distribution system, especially on the apparent losses and harmonic distortion. In order to mitigate these issues, optimal placement and sizing of multiple passive filters for medium and low voltage network are determined using proposed modified lightning search algorithm (MLSA). Weight summation approach is used to identify the best location for the filter in the design stage, while Pareto with the assistance of fuzzy technique is used to determine a suitable sizing for the passive filter switching that will operate 24 hours a day based on EV consumer behavior. In addition, the proposed method also considers fast charging station (CS) and slow CS analysis for medium and low voltages, respectively. All power system harmonic flow, load profile, EV charging pattern, passive filter, CS and battery modelling are programmed in MATLAB. The performance of MLSA is compared with other metaheuristic techniques, such as particle swarm optimization (PSO), firefly algorithm (FA) and lightning search algorithm (LSA). From the results, the MLSA is able to minimize the apparent losses and harmonic distortion at 33 bus radial distribution system (medium voltage) by considering worst harmonic distortion scenario from all 17 units of fast CS. The MLSA has provided superior result compared to PSO, FA, and LSA. Next, the proposed method was tested at 449 bus radial distribution system (medium and low voltage) with variance load and EV charging pattern for 24 hours, with fifteen minutes interval, using slow CS. The analysis shows that the optimal placements and sizes of variable passive filters were able to reduce the maximum total harmonic distortion (THD) for voltage, current and also the total apparent losses up to 39.14%, 52.5%, and 2.96 %, respectively. Furthermore, the results prove that the variation of passive filter is able to provide superior solution compared to single sizing. Therefore, it can be concluded that the multiple passive filters with an assistance of the MLSA algorithm is suitable to be implemented in minimizing overall apparent losses and harmonic distortions. This study is very useful as a guide for distribution network owners to control the impact of large-scale CS deployment in the future distribution system.