A Privacy-Preserving Distributed Control of Optimal Power Flow

We consider a distributed optimal power flow formulated as an optimization problem that maximizes nondifferentiable concave function. Solving such by the existing algorithms can lead to data privacy issues because solution information exchanged within be utilized adversary infer data. To preserve privacy, in this paper we propose differentially private projected subgradient (DP-PS) algorithm includes encryption step. show sequence generated DP-PS converges expectation, probability, and with probability 1. Moreover, rate of convergence expectation is affected target level chosen user. conduct numerical experiments demonstrate preservation DP-PS.