A Distributed Time-Varying Inherent Privacy-Preserving Consensus Algorithm for Integrated Energy Systems
Wenlong Fu, YiHeng Gao, Yunning Zhang, X. L. Tu, Songlin Hu, Yu S. Huang · IEEE Transactions on Industrial Informatics · 2024
This article addresses the issue of privacy-preserving distributed economic dispatch in integrated energy systems (IESs). We propose the time-varying inherent privacy-preserving consensus algorithm (TVIPPCA) without involving a third party, aimed at safeguarding sensitive information during communication and ensuring optimal energy provision at minimal cost. The proposed TVIPPCA is designed with inherent privacy protection for each private variable, featuring a two-layer security strategy. The first layer of security is achieved through an intrinsic node decomposition mechanism coupled with random mixing coefficients, effectively shielding key system parameters from potential adversaries. The second layer enhances privacy preservation by designing a time-varying weight matrix strategy, further obfuscating the information transmitted. Through rigorous theoretical analysis, the proposed TVIPPCA is proven to converge to the optimal value while strictly protecting the privacy of sensitive information. Simulations in a 39-32 node IES validate the algorithm's efficacy in cost optimization and privacy protection.