Charging control of electric vehicles using contextual bandits considering the electrical distribution grid

TL;DR

This paper proposes a contextual bandit-based control mechanism for electric vehicle charging to prevent grid overloads, integrating mobility and electrical network simulations under realistic conditions.

Contribution

It introduces a novel approach combining contextual bandits with detailed simulations for EV charging control, outperforming traditional reinforcement learning methods.

Findings

Conditional bandit learning outperforms context-free reinforcement learning.

The approach effectively prevents electrical grid overloads during EV charging.

Simulations demonstrate the method's suitability for real-world scenarios.

Abstract

With the proliferation of electric vehicles, the electrical distribution grids are more prone to overloads. In this paper, we study an intelligent pricing and power control mechanism based on contextual bandits to provide incentives for distributing charging load and preventing network failure. The presented work combines the microscopic mobility simulator SUMO with electric network simulator SIMONA and thus produces reliable electrical distribution load values. Our experiments are carefully conducted under realistic conditions and reveal that conditional bandit learning outperforms context-free reinforcement learning algorithms and our approach is suitable for the given problem. As reinforcement learning algorithms can be adapted rapidly to include new information we assume these to be suitable as part of a holistic traffic control scenario.