ANFIS-DRIVEN FLEXIBLE CHARGING CONTROL OF PLUG-IN ELECTRIC VEHICLES FOR SMART GRID LOAD MANAGEMENT

Abstract

The large-scale integration of plug-in electric vehicles (PEVs) into smart grids introduces significant challenges in load management due to uncoordinated charging behaviour, which can lead to peak demand escalation, voltage deviations, and increased operational stress on distribution networks. To address these challenges, this paper proposes an Adaptive Neuro-Fuzzy Inference System (ANFIS)–driven flexible charging control strategy for PEVs aimed at effective smart grid load management. The proposed controller dynamically adjusts PEV charging power based on grid load conditions, electricity price signals, and battery state-of-charge (SOC), enabling load curve smoothing and peak demand reduction. The neuro-fuzzy framework combines the learning capability of neural networks with the reasoning ability of fuzzy logic, ensuring adaptability under uncertain and time-varying grid conditions. Simulation results demonstrate that the proposed ANFIS-based control strategy significantly reduces peak-to-average load ratio, improves load profile smoothness, and enhances overall grid operational efficiency compared with conventional uncoordinated and rule-based charging schemes