Local Differential Privacy for Smart Meter Data Sharing with Energy Disaggregation

Smart meter data with energy disaggregation can be used to identify energy usage of appliances, offering valuable insights for consumers and energy companies to better manage energy. However, these techniques also present privacy risks, such as the potential for behavioral profiling. Local differential privacy (LDP) methods provide strong privacy guarantees with high efficiency in addressing privacy concerns. Exiting LDP methods primarily focus on securing aggregated energy data, ignoring the streaming nature of smart meter data and individual appliance data. In this paper, we propose a novel LDP approach (named LDP–Energy) to facilitate the sharing of appliance-level energy data over time while not revealing individual users’ appliances and their usage patterns. By incorporating a sliding window concept, LDP–Energy is able to efficiently handle the streaming smart meter data. Our evaluations indicate that LDP–Energy performs efficiently compared to other methods, striking a proper balance between privacy and data utility for effective analysis.