Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Dynamic Electric Vehicle Charging Pricing for Load Balancing in Power Distribution Networks based on Collaborative DDPG Agents

Leloko J. Lepolesa, Kayode E. Adetunji, Khmaies Ouahada, Zhenqing Liu, Ling Cheng

TL;DR

The paper tackles grid stability under high EV adoption by proposing a dynamic EV charging pricing framework driven by collaborative deep reinforcement learning to balance and distribute EV demand across neighboring distribution networks. It uses XGBoost for day-ahead conventional load forecasting and polynomial regression for EV load forecasting, followed by a DRL-based pricing stage that compares DDPG, SAC, and PPO, with DDPG showing superior performance. The approach enables peak shaving, valley filling, and load balancing across networks, improving overall grid utilization compared to traditional ToU pricing. The findings demonstrate that inter-network load balancing via the proposed PVB pricing strategy can significantly reduce utilization disparities and smooth the aggregate demand trajectory, offering a scalable path for integrating EVs into distribution grids.

Abstract

The transition from the Internal Combustion Engine Vehicles (ICEVs) to the Electric Vehicles (EVs) is globally recommended to combat the unfavourable environmental conditions caused by reliance on fossil fuels. However, it has been established that the charging of EVs can destabilize the grid when they penetrate the market in large numbers, especially in grids that were not initially built to handle the load from the charging of EVs. In this work, we present a dynamic EV charging pricing strategy that fulfills the following three objectives: distribution network-level load peak-shaving, valley-filling, and load balancing across distribution networks. Based on historical environmental variables such as temperature, humidity, wind speed, EV charging prices and distribution of vehicles in different areas in different times of the day, we first forecast the distribution network load demand, and then use deep reinforcement learning approach to set the optimal dynamic EV charging price. While most research seeks to achieve load peak-shaving and valley-filling to stabilize the grid, our work goes further into exploring the load-balancing between the distribution networks in the close vicinity to each other. We compare the performance of Deep Deterministic Policy Gradient (DDPG), Soft Actor-Critic (SAC) and Proximal Policy Optimization (PPO) algorithms for this purpose. The best algorithm is used for dymamic EV pricing. Simulation results show an improved utilization of the grid at the distribution network level, leading to the optimal usage of the grid on a larger scale.

Dynamic Electric Vehicle Charging Pricing for Load Balancing in Power Distribution Networks based on Collaborative DDPG Agents

TL;DR

The paper tackles grid stability under high EV adoption by proposing a dynamic EV charging pricing framework driven by collaborative deep reinforcement learning to balance and distribute EV demand across neighboring distribution networks. It uses XGBoost for day-ahead conventional load forecasting and polynomial regression for EV load forecasting, followed by a DRL-based pricing stage that compares DDPG, SAC, and PPO, with DDPG showing superior performance. The approach enables peak shaving, valley filling, and load balancing across networks, improving overall grid utilization compared to traditional ToU pricing. The findings demonstrate that inter-network load balancing via the proposed PVB pricing strategy can significantly reduce utilization disparities and smooth the aggregate demand trajectory, offering a scalable path for integrating EVs into distribution grids.

Abstract

The transition from the Internal Combustion Engine Vehicles (ICEVs) to the Electric Vehicles (EVs) is globally recommended to combat the unfavourable environmental conditions caused by reliance on fossil fuels. However, it has been established that the charging of EVs can destabilize the grid when they penetrate the market in large numbers, especially in grids that were not initially built to handle the load from the charging of EVs. In this work, we present a dynamic EV charging pricing strategy that fulfills the following three objectives: distribution network-level load peak-shaving, valley-filling, and load balancing across distribution networks. Based on historical environmental variables such as temperature, humidity, wind speed, EV charging prices and distribution of vehicles in different areas in different times of the day, we first forecast the distribution network load demand, and then use deep reinforcement learning approach to set the optimal dynamic EV charging price. While most research seeks to achieve load peak-shaving and valley-filling to stabilize the grid, our work goes further into exploring the load-balancing between the distribution networks in the close vicinity to each other. We compare the performance of Deep Deterministic Policy Gradient (DDPG), Soft Actor-Critic (SAC) and Proximal Policy Optimization (PPO) algorithms for this purpose. The best algorithm is used for dymamic EV pricing. Simulation results show an improved utilization of the grid at the distribution network level, leading to the optimal usage of the grid on a larger scale.

Paper Structure

This paper contains 21 sections, 18 equations, 12 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Background
  3. Extreme Gradient Boosting (XGBoost)
  4. Polynomial regression
  5. Deep Reinforcement Learning (DRL) - Deep Deterministic Policy Gradient (DDPG) algorithm
  6. Distribution network load forecasting and dynamic EV charging pricing
  7. Distribution network conventional load forecasting
  8. Conventional load data description and pre-processing
  9. Charging load of electric vehicles
  10. Dynamic pricing
  11. Results and Discussion
  12. Conventional load prediction
  13. Electric vehicles load prediction
  14. Electric vehicles' charging dynamic pricing
  15. Conventional ToU pricing
  16. ...and 6 more sections

Figures (12)

  • Figure 1: Distribution network load forecasting and dynamic EV charging pricing
  • Figure 2: Total power consumption for 21 days
  • Figure 3: Vehicles' travel behaviour between residential and commercial regions based on U.S. NHTS data
  • Figure 4: RMSE and r2 results of conventional load forecasting regression algorithms
  • Figure 5: Predicted EV load demand
  • ...and 7 more figures