Privacy-Preserving Distributed Energy Management for Battery Energy Storage Systems over Time-Varying Networks

Abstract

This article addresses the privacy-preserving energy management problem of battery energy storage systems (BESSs). An autonomous privacy-preserving distributed optimization (APPDO) scheme is developed over time-varying networks with the aim of regulating the power output of local BESS to fulfill the total load demand at the minimum economic cost under battery capacity constraints without privacy leakage. To this end, a linearly convergent distributed algorithm is proposed by combining the gradient descent algorithm with leaderless and leader-following consensus schemes. This algorithm is applicable to both islanded and grid-connected modes of BESSs. Furthermore, a novel privacy-preserving approach is constructed by injecting well-designed perturbation sequences into the data exchanged between neighboring nodes, making it effective against malicious eavesdroppers. Furthermore, a comprehensive analysis framework is established to evaluate the convergence, optimality, and privacy-preserving performance of the APPDO algorithm. Finally, numerical studies are conducted to demonstrate the effectiveness of the developed APPDO scheme.