Differentially Private Optimal Power Flow for Distribution Grids

Although distribution grid customers are obliged to share their consumption data with system operators (DSOs), a possible leakage of this is often disregarded in operational routines DSOs. This paper introduces privacy-preserving optimal power flow (OPF) mechanism for grids that secures customer privacy from unauthorised access OPF solutions, e.g., current and voltage measurements. The based on the framework differential allows control participation risks individuals dataset by applying carefully calibrated noise output computation. Unlike existing private mechanisms, does not apply optimization parameters or its result. Instead, it optimizes variables as affine functions random noise, which weakens correlation between loads variables. To ensure feasibility randomized solution, makes use chance constraints enforced limits. further extended optimality loss induced well variance shows differentially solution leak up specified parameters.