Uncertainty-aware stochastic assessment of fully integrated EV charging infrastructure in distribution networks under critical load conditions

The widespread deployment of Electric Vehicle Charging Stations (EVCS) contributes considerable unpredictability to Distribution Networks (DNs), frequently intensifying power losses and voltage instability during peak demand periods. This research proposes an uncertainty-aware stochastic evaluation methodology for the optimal planning and operation of a fully integrated network that includes EVCS, Distributed Generators (DGs), and Shunt Capacitors (SCs). To improve the computational accuracy of distribution analysis, a new zero-bus based forward-backward load flow (FBLF) technique is used. Furthermore, to address the competing trade-offs in network performance, a Multi-Objective Modified Discrete Particle Swarm Optimization (MDPSO) technique based on a Fuzzy Max-Min approach is presented. The optimization approach aims to reduce active power loss, Voltage Deviation Index (VDI), environmental emissions, and total operating expenses simultaneously. The propsed methodology is tested on an IEEE 33-bus DN under a variety of critical loading scenarios and probabilistic uncertainty. Monte Carlo Simulation (MCS) is used for uncertainty modeling in order to effectively address the probabilistic nature of load demands and EV charging patterns. According to simulation studies, the fully integrated infrastructure’s coordinated control greatly reduces the negative effects of stochastic EV loads, resulting in better voltage profiles and more cost-effective network operations.