Differentially Private Real-Time Pricing Control for Smart Grids

Smart meters provide fine-grained power usage profiles of consumers to various utility providers, thus facilitating multiple grid functionalities such as load monitoring, Real-Time Pricing (RTP), demand response, and so on. However, information leakage from such usage profiles reveals consumers’ private day-to-day life patterns and their home presence/absence, as the state-of-the-art metering strategies lack adequate security and privacy measures. Since Smart grid communication infrastructure supports low bandwidth, it prohibits the usage of computation-intensive cryptographic solutions. Among different privacy-preserving smart meter streaming methods, data manipulation techniques can easily be implemented in smart meters and do not require installing any storage devices or alternative energy sources. For this purpose, Differential Privacy (DP) is widely adopted in the literature due to its solid mathematical foundation. However, the effect of such manipulations on the RTP control is worth exploring since pricing signals operate in a closed-loop between consumers and utilities. This brings up the privacy-utility tradeoff problem between the user’s achieved privacy and the performance of the pricing loop of Smart grids, an area where the characterization between privacy and pricing control performance is not yet established. We analytically highlight such privacy-utility tradeoff in the closed-loop RTP systems in terms of achieved privacy and the overall generation scheduling errors. We utilize the notion of \(w\) -event privacy and present a RTP aware DP scheme that promises strong user privacy and guarantees pricing signal stabilization irrespective of the privacy level of the DP mechanism. Finally, we show the efficiency and robustness of our scheme by performing extensive experimental validation on MATLAB and, subsequently, on an in-house smart meter test bed.