Multi-objective Photovoltaic Sizing with Diverse Inverter Control Schemes in Distribution Systems Hosting EVs

Worldwide, photovoltaic (PV) and electric vehicles (EVs) have intensively been integrated into distribution systems. As a result, different operational issues can be observed due to PV generation variability and EV stochastic characteristics. In this work, an optimal sizing approach of multiple PVs in the existence of EVs is proposed. The proposed approach minimizes both the total voltage deviations and overall energy losses, prevents active PV power curtailment, and considers numerous constraints of PV, EV, and the distribution system. The features of the proposed approach are the considerations of PV, EV, and load uncertainties via incorporating their probabilistic models. Besides, it models arrival/parting times of EVs, the required state of charge (SOC) of EV batteries based on initial SOCs and remaining parking periods, and controlled/uncontrolled charging. Furthermore, diverse control schemes of the interfacing PV inverter are formulated in the proposed optimization model. To effectively solve this comprehensive model with conflicting sub-functions and variables, a two-level multi-objective evolutionary algorithm based on decomposition with fuzzy sets is developed. The upper optimization level accurately optimizes the sizes of multiple PVs while the lower one optimizes charging/discharging of EV batteries, PV inverter oversize, and PV reactive power. The results prove the effectiveness of the proposed approach.