Vehicle-to-Building (V2B) Charging Optimization

Overview

Developing intelligent systems for Vehicle-to-Building (V2B) charging that optimize energy usage, reduce costs, and improve grid stability through advanced algorithms and machine learning techniques.

Key Projects

1. OPTIMUS: Discrete Event Simulator

A comprehensive discrete event simulator for optimizing V2B charging operations, enabling detailed analysis of charging strategies and their impact on building energy systems.

Publication: OPTIMUS: Discrete Event Simulator for Vehicle-to-Building Charging Optimization, IEEE SmartComp 2024

2. Reinforcement Learning for V2B

Developed reinforcement learning approaches for heterogeneous agents with long-term reward optimization in V2B charging scenarios.

Publication: Reinforcement Learning-based Approach for Vehicle-to-Building Charging with Heterogeneous Agents and Long Term Rewards, AAMAS 2025

3. Online Decision-Making Under Uncertainty

Created online decision-making algorithms that adapt to real-time uncertainty in V2B systems, improving reliability and efficiency.

Publication: Online Decision-Making Under Uncertainty for Vehicle-to-Building Systems, ACM/IEEE ICCPS 2025

4. CONSENT: Negotiation Framework

An innovative negotiation framework that leverages user flexibility to optimize V2B charging under uncertainty, balancing user preferences with grid constraints.

Status: Accepted, AAMAS 2026

Technologies

• Mixed Integer Linear Programming (MILP)
• Reinforcement Learning
• Stochastic Optimization
• Discrete Event Simulation
• Python, MATLAB

Impact

These systems enable more efficient integration of electric vehicles into building energy management, reducing costs and environmental impact while maintaining grid stability.