Dynamic-Based Privacy Preservation for Distributed Economic Dispatch of Microgrids

Abstract

With an escalating emphasis on distributed economic dispatch within microgrid systems due to its inherent adaptability, scalability, and sustainability, an extensive focus on the confidentiality of this field is pronouncedly emerging. The primary emphasis of this study is the safeguarding of power-sensitive information in the distributed economic dispatch issue prevalent in microgrids. This pursuit leads us to the development of a distributed optimization algorithm that preserves privacy, tailored for directed networks. The algorithm strives to secure a balance between supply and demand at the lowest economic cost, all while adhering to real-world constraints and maintaining the confidentiality of power-sensitive information. To fulfill this objective, we propose a novel privacy-preserving distributed algorithm that capitalizes on the inherent resilience exhibited by system dynamics toward uncertainty. Specifically, to ensure privacy preservation, we strategically incorporate randomness into the mixing weights, thereby generating a degree of uncertainty in communication messages during the initial iteration. Rigorous analysis is built to delineate that our method can achieve exact convergence and ensure the confidentiality of power-sensitive information. Further, additional numerical trials conducted on an IEEE 14-bus system substantiate the algorithm's practical efficiency.