Work

Voltage Stability Enhancement in IEEE 14-Bus System Using Deep Deterministic Policy Gradient for EV Charging Management

Year: 2025

Type: conference-paper

Abstract: The increasing integration of Electric Vehicles (EVs) into power systems presents significant challenges for voltage stability due to their unpredictable charging behaviors and potential grid disturbances. This paper proposes a novel Deep Reinforcement Learning (DRL) framework to coordinate EV charging and voltage control in distribution networks using the Deep Deterministic Policy Gradient (DDPG) algorithm. A two-layer control architecture is introduced, where the upper layer handles charging scheduling while the lower layer focuses on voltage regulation. The framework is modeled as a Markov Decision Process (MDP) and trained using an Actor-Critic structure capable of managing mixed discrete-continuous action spaces. Simulation results on a modified IEEE 14-bus system validate the effectiveness of the proposed method. Compared with conventional methods like Deep Q-network (DQN), Proximal policy optimization (PPO), and Soft Actor Critic (SAC), our approach achieves faster convergence and superior voltage stability across varying levels of EV penetration, ensuring all nodal voltages remain within safe operational limits. The results demonstrate the potential of DDPG-based strategies in enhancing grid reliability and supporting smart EV integration.

Source: IEEE Workshop on the Electronic Grid (eGRID)

DOI: 10.1109/egrid63452.2025.11255115

Authors: Arash Mousaei, Yahya Naderi, Syed Muhammad Arif, I. Safak Bayram

Institutions: University of Tabriz, Scottish Power (United Kingdom), University of Strathclyde

Language: English

Topic: Electric Vehicles and Infrastructure

Subfield: Electrical and Electronic Engineering

Field: Engineering

Domain: Physical Sciences

SDGs: Affordable and clean energy, Industry, innovation and infrastructure

Open Access status: green

OpenAlex ID: W4416727397