ADMM Based Multi-Objective Distributed Optimization of AC/DC Distribution System

With increasing proportion of DC-driven loads and the growing user demand for DC distribution systems, coupled with addition of high proportion of clean energy sources and energy storage devices, traditional distribution system is transforming from a unidirectional power source to a bidirectional energy-flowing AC-DC distribution system. Aiming at problems of poor real-time, poor embeddability and poor privacy protection of current centralized optimization methods, this paper suggests a multi-objective distributed optimization scheduling strategy based on the alternating direction method of multipliers (ADMM) algorithm. Firstly, an AC/DC distribution network optimization model is assembled with aims of minimizing substation active output, minimizing network loss, and minimizing the penalty for wind and solar disposal; Secondly, the distribution network is decoupled into a number of AC and DC sub-regions according to the converter connection, and a distributed optimal scheduling strategy based on the ADMM is established. Finally, through analysis of examples, it is verified that the method suggested can quickly converge to same result as the centralized optimization method, and realize the 100% absorption of distributed new energy and the effective reduction of substation output and network loss cost.