Distributed Optimization for Integrated Energy Systems with Secure Multi-Party Computation

With increasing distributed energy resource integration, future power and energy systems will be more decentralized using advanced Internet of Things (IoT) technologies. Integrated energy systems (IES) boost the whole energy efficiency by coordinating multi-regional energy resources and networks. However, distributed coordination of the IES requires different subregions or energy hubs (EHs) to share their sensitive information (e.g., energy demands and operation status) explicitly, which poses serious privacy leakage. To this end, secure multi-party computation (SMPC) is innovatively introduced to the distributed optimization of the IES in this paper. First, standardized modeling of multiple interconnected EHs with linearized network models is formulated to analyze the IES’s inherent energy and information interaction comprehensively. Then, a privacy-preserving distributed optimal energy flow algorithm is proposed by combining the Paillier Cryptosystem mechanism with the alternating direction multiplier method (ADMM). Theoretical analysis proves the proposed method is convergent without sharing sensitive information in plaintext. Numerical experiments on a three-subregions IES validate that the proposed method has better convergence performance than the differential privacy-based method. Results show that the maximum relative error of the distributed optimal solutions with various step sizes is no more than 0.072% compared with the centralized method.