A hierarchical control scheme for residential air-conditioning loads to provide real-time market services under uncertainties

Recently, there has been growing interest in the provision of market services from distributed energy resources (DERs). In pursuing this goal, demand response (DR) aggregators continue to face challenges in retaining privacy and comfort for end-users, mitigating scalability issues while controlling a large cohort of DERs and handling uncertainties which are inevitable in a practical setting. This paper presents an end-user privacy and comfort preserving, scalable, hierarchical control scheme for inverter-type air conditioners to provide real-time market services in the presence of uncertainties. Privacy and scalability are achieved thorough the adoption of the alternating direction method of multipliers (ADMM) framework which ensures minimal reliance on local information whilst ensuring desired reference tracking without compromising the end-user comfort. Benefiting from the proposed non-conservative robust MPC design, the local control is able to account for mismatches in outdoor temperature predictions. The overall scheme is validated using real data obtained from the Australian Energy Market operator. The results demonstrate that the proposed approach can achieve desired tracking of the reference signal while regulating indoor temperature within a narrow range (& PLUSMN;1 & DEG;C) from the nominal set-point. Besides, the robustness to uncertainties is achieved without compromising computational performance and therefore the approach is scalable. (C)& nbsp;2022 Published by Elsevier Ltd.