Distributed Optimal Energy Management for Integrated Energy Systems

This article is concerned with the multiobjective energy management problem of integrated energy systems (IESs) in a distributed fashion. A multiobjective model is established where the operating costs and gas emissions are treated as competing objectives simultaneously. To achieve optimal energy management with reasonable operating costs and gas emissions, a primal-dual-based distributed algorithm with dynamic weights is proposed to assign the multiple energy sources. By this algorithm, each participant only needs to compute locally and share information among neighbors, and thus it offers better flexibility, reliability, and adaptivity, and lower communication burden compared with some centralized algorithms. Besides, the proposed algorithm can solve the multiobjective problem with no special initialization conditions and lower computational complexity different from existing distributed optimization algorithms for IESs. Furthermore, it is proved that each participant can converge to the global optimal solution asymptotically. Simulation studies demonstrate that the proposed algorithm can coordinate the two conflicting objectives of economy and environment simultaneously and obtain the entire Pareto front. The effectiveness and scalability of the proposed algorithm are also validated on a 39-32 power-heat IESs and the integrated energy unit system.