Intermittently differential privacy in smart meters via rechargeable batteries

• Published in: Electric power systems research 2021-10, Vol.199, p.107410, Article 107410
• Main Authors: Liu, Xing, Wang, Huiwei, Chen, Guo, Zhou, Bo, Rehman, Aqeel ur
• Format: Article
• Language: English
• Subjects: Algorithms; Batteries; Cost saving; Differential privacy; Energy consumption; Machine learning; Measuring instruments; Privacy; Rechargeable batteries; Reinforcement learning; Smart meter; Studies
• ISSN: 0378-7796; 1873-2046
• DOI: 10.1016/j.epsr.2021.107410

•We propose a battery-based intermittently differential privacy (IDP) scheme to achieve differential privacy at certain times with taking the rate restriction of battery into account for smart meter readings.•We develop a RL-based scheme to get the optimal policy for battery control that serves two goals: maintaining the battery power level and realizing cost savings.•We combine our IDP scheme with RL-based scheme to form a complete RL-IDP scheme for smart meter readings, experimental results show that RL-IDP has better performance in both privacy-preserving and cost savings. In this paper, we first propose a battery-based intermittently differential privacy (IDP) scheme and prove our IDP scheme can achieve differential privacy that is a powerul privacy-preserving mechanism. We then develp another scheme–a reinforcement learning (RL) algorithm aiming to guide the battery control policy for two purposes: 1) cost savings; 2) Solving battery constraint. Finally, we integrate the IDP scheme with the RL scheme into a complete RL-IDP scheme. With this scheme, the users can have a privacy-preserving and an economic environemnt-a “green” environment. [Display omitted] The smart meters have been widely deployed to monitor customer usage profiles around the world, which can read the energy load of residents at the rate of per minute or per second. This fine-grained data would expose personal behaviors and other sensitive information to malicious adversaries. To address this privacy concern, battery-based load hiding (BLH) scheme was proposed and had been explored for several years. The differential privacy is a provable method to preserve the privacy against the adversaries of arbitrary computational power. At first, a battery-based intermittently differential privacy (IDP) scheme is proposed in this article and also prove that the IDP can achieve the differential privacy. Then, we develop another scheme — a Reinforcement Learning (RL) algorithm to guide the battery control policy to match the requirement of battery constraints and cost-saving in a better way. Finally, we integrate the IDP scheme with the RL algorithm into a complete RL-IDP scheme. The experimental results show that the privacy-preserving level performs well by our RL-IDP scheme while it can achieve cost-saving.