Achieving Optimality and Fairness in Autonomous Demand Response: Benchmarks and Billing Mechanisms

Abstract

Autonomous demand response (DR) programs are scalable and result in a minimal control overhead on utilities. The idea is to equip each user with an energy consumption scheduling (ECS) device to automatically control the user's flexible load to minimize his energy expenditure, based on the updated electricity pricing information. While most prior works on autonomous DR have focused on coordinating the operation of ECS devices in order to achieve various system-wide goals, such as minimizing the total cost of generation or minimizing the peak-to-average ratio in the load demand, they fall short addressing the important issue of fairness. That is, while they usually guarantee optimality, they do not assure that the participating users are rewarded according to their contributions in achieving the overall system's design objectives. Similarly, they do not address the important problem of co-existence when only a sub-set of users participate in a deployed autonomous DR program. In this paper, we seek to tackle these shortcomings and design new autonomous DR systems that can achieve both optimality and fairness. In this regard, we first develop a centralized DR system to serve as a benchmark. Then, we develop a smart electricity billing mechanism that can enforce both optimality and fairness in autonomous DR systems in a decentralized fashion.