Long-term optimal planning for renewable based distributed generators and plug-in electric vehicles parking lots toward higher penetration of green energy technology
Ahmad K. ALAhmad,Renuga Verayiah,Hussain Shareef,Agileswari Ramasamy
DOI: https://doi.org/10.1016/j.est.2024.114057
IF: 9.4
2024-10-15
Journal of Energy Storage
Abstract:Due to the extensive pollution generated by conventional fuel-based power systems, there has been a significant shift in global focus toward increasing the adoption of renewable energy sources (RESs) through renewable-based distributed generation (DG), particularly wind and solar photovoltaic (PV) systems. Additionally, the electrification of the automotive sector, aimed at reducing pollution, is driving a rapid increase in electric vehicles (EVs). A critical element of this transition is the development of efficient infrastructure for plug-in electric vehicle parking lots (PEV-PLs). A collaborative planning model is essential to address the impact of integrating RESs and PEV-PLs into the electric power distribution system (DS) over the long term. This paper introduces a long-term mixed-integer non-linear (MINL) optimization planning model designed to optimize the planning and operation of RESs, including wind and PV sources, alongside PEV-PLs infrastructure. The goal is to increase the penetration of renewable energy and EVs within the DS while adhering to security constraints. The optimization model features three non-linear, incompatible objective functions: minimizing overall strategic expected investment, maintenance, emission, and operational costs; long-term power loss; and voltage deviation. Moreover, to ensure realism, the model incorporates uncertainties related to stochastic variables such as the intermittent nature of RESs, EV energy and time variables, loads, and energy price fluctuations, using Monte Carlo Simulation (MCS) and the backward reduction method (BRM). A hybrid optimization algorithm addresses the proposed objectives, combining the non-dominated sorting genetic algorithm (NSGA-II) and multi-objective particle swarm optimization (MOPSO) to minimize the three distinct objective functions concurrently. The effectiveness of the planning model is validated using the 69-bus benchmark test system, with four configurations tested: case 1 (the base case), case 2 (the base case with RESs (wind and PV)), case 3 (the base case with RESs and PEV-PLs), and case 4 (the base case with RESs, PEV-PLs, and a higher number of EVs). The impact of RESs on DS operation, PEV-PLs on RES penetration levels and DS operation, and the effect of increased EV penetration on the integrated capacity of RESs and DS operation are thoroughly investigated. Simulation results demonstrate that the optimal integration of 5 PEV-PLs, accommodating a fleet of 107 PEVs with wind and PV DGs, increases the RES penetration level from 3.35 MVA to 3.85 MVA compared to the case with RESs alone. Moreover, integrating PEV-PLs with RESs results in a 51.00 % reduction in overall operational costs, a 37.55 % reduction in overall planning and operation costs, a 52.82 % reduction in total carbon emissions, and a 45.85 % reduction in total voltage deviation.
energy & fuels