Privacy-Aware Load Ensemble Control: A Linearly-Solvable MDP Approach

TL;DR

This paper introduces a privacy-preserving control method for demand response load ensembles using a linearly-solvable MDP that balances operational efficiency with participant privacy, validated on real-world data.

Contribution

It develops a novel MDP framework incorporating differential privacy for ensemble load control, enhancing privacy protection in demand response programs.

Findings

Effective privacy protection of load transition data.

Trade-off between control optimality and privacy guarantees.

Validated approach using real-world microgrid data.

Abstract

Demand response (DR) programs engage distributed demand-side resources, e.g., controllable residential and commercial loads, in providing ancillary services for electric power systems. Ensembles of these resources can help reducing system load peaks and meeting operational limits by adjusting their electric power consumption. To equip utilities or load aggregators with adequate decision-support tools for ensemble dispatch, we develop a Markov Decision Process (MDP) approach to optimally control load ensembles in a privacy-preserving manner. To this end, the concept of differential privacy is internalized into the MDP routine to protect transition probabilities and, thus, privacy of DR participants. The proposed approach also provides a trade-off between solution optimality and privacy guarantees, and is analyzed using real-world data from DR events in the New York University microgrid in New York, NY.