Heuristic Greedy Scheduling of Electric Vehicles in Vehicle-to-Grid Microgrid Owned Aggregators

Area of Science:

• Electrical Engineering
• Renewable Energy Systems
• Artificial Intelligence

Background:

• On-grid microgrids require efficient electric vehicle (EV) scheduling for cost-effective operation.
• Stochastic parameters, large numbers, and correlations complicate EV scheduling in hybrid microgrids.
• Existing scheduling methods often require models or training, limiting their real-time applicability.

Purpose of the Study:

• To introduce an innovative, model-free, and training-free greedy approach for EV scheduling in hybrid microgrids.
• To reduce total operational costs for on-grid EVs within hybrid microgrids.
• To assess the approach's efficiency, scalability, and tolerance to uncertainties.

Main Methods:

• A heuristic greedy approach was developed for optimal EV scaling and scheduling.
• The approach considers electricity prices, EV arrival/departure times, and revenue for load demand fulfillment.
• A hardware-in-the-loop experiment and a data simulator were used for real-time validation and uncertainty assessment.

Main Results:

• The greedy scheduler effectively manages on-grid EVs as energy storage within photovoltaic, wind turbine, and utility grid hybrid systems.
• Simulations over 500 years show a 50-85% reduction in energy consumption costs, outperforming state-of-the-art methods.
• The approach demonstrated applicability, scalability, efficiency, and tolerance to system uncertainties.

Conclusions:

• The proposed model-free greedy strategy offers a robust and efficient solution for EV scheduling in hybrid microgrids.
• Increased EV penetration significantly decreased total expenses, highlighting the economic benefits of the approach.
• The method successfully addresses the challenges posed by stochastic parameters and complex correlations in microgrid operations.