Privacy-Preserving and Approximately Truthful Local Electricity Markets: A Differentially Private VCG Mechanism

Privacy-aware market participants care about the leakage of their private information via statistical releases of local electricity markets outputs. This kind of privacy breach would have major implications on the future transactions of the market participants and unauthorized observers’ beliefs about them. To address this challenge, we introduce the notion of noisy electricity markets in the framework of Differential Privacy (DP) for preserving the privacy of individuals and maintaining the utility of their data for social good. In this regard, this paper proposes a novel differentially private mechanism for local electricity markets that releases a near-optimal solution while guarantying the outputs of the market would reveal almost nothing about any individual’s input data. To do so, we implement the exponential mechanism for privatizing the baseline Vickrey–Clarke–Groves (VCG) mechanism in the proposed local electricity market. Moreover, we provide an upper-bound on the social welfare loss incurred by the privacy constraint and analyze the inherent trade-off between the privacy and suboptimality. In the end, numerical case studies for reflecting the theoretical properties of the proposed mechanism are provided.