On the Optimality of Procrastination Policy for EV charging under Net Energy Metering

Minjae Jeon,Lang Tong,Qing Zhao

DOI: https://doi.org/10.1109/CDC49753.2023.10383822

2024-04-30

Abstract:We consider the problem of behind-the-meter EV charging by a prosumer, co-optimized with rooftop solar, electric battery, and flexible consumptions such as water heaters and HVAC. Under the time-of-use net energy metering tariff with the stochastic solar production and random EV charging demand, a finite-horizon surplus-maximization problem is formulated. We show that a procrastination threshold policy that delays EV charging to the last possible moment is optimal when EV charging is co-optimized with flexible demand, and the policy thresholds can be computed easily offline. When battery storage is part of the co-optimization, it is shown that the net consumption of the prosumer is a two-threshold piecewise linear function of the behind-the-meter renewable generation under the optimal policy, and the procrastination threshold policy remains optimal, although the thresholds cannot be computed easily. We propose a simple myopic solution and demonstrate in simulations that the performance gap between the myopic policy and an oracle upper bound appears to be 0.5-7.5%.

Systems and Control

What problem does this paper attempt to address?

This paper mainly discusses the joint optimization problem of electric vehicle (EV) charging, rooftop solar, battery energy storage, and flexible electric devices (such as water heaters and HVAC systems) under the net energy metering (NEM) policy. The researchers consider a finite time window and aim to maximize the surplus. Both EV charging demand and solar production are stochastic. They find that when EV charging and flexible demand are jointly optimized, the optimal strategy is procrastination strategy, which means delaying EV charging until the last moment as much as possible. In addition, they propose a simple myopic policy to solve the battery operation problem and prove that myopic policy can be optimal when the battery state does not reach the boundary.

The main contributions of this paper include:
1. Demonstrating that the procrastination threshold strategy is optimal for the coordinated scheduling of EV charging, flexible demand, and stochastic renewable energy.
2. Proposing a myopic solution to handle battery operations and showing that decision-making under this strategy is still optimal when the battery charge state does not reach the boundary.
3. Showing that myopic battery operation is a piecewise linear function generated by renewable energy and that the procrastination threshold strategy remains effective when energy storage participates in coordinated optimization, although determining the threshold is not easy.
4. Providing numerical simulation results that demonstrate the superiority of the procrastination threshold strategy and myopic battery operation over other non-coordinated optimization strategies and their performance is close to the ideal strategy.

The related work section of the paper reviews the research on the coordinated scheduling of EVs, energy storage, and flexible loads, pointing out the limitations of existing literature, such as not considering EV charging deadline and deferability or not simultaneously optimizing multiple flexible loads. The authors' work builds on the recently revealed characteristics of the NEM tariff structure.

In summary, this paper introduces new strategies for the coordinated optimization problem of home energy management systems, which perform well in practical scenarios and are of great significance in reducing power consumption and improving grid efficiency.