A distributed consensus based algorithm for economic dispatch over time‐varying digraphs

doi:10.1049/cth2.12526

A distributed consensus based algorithm for economic dispatch over time‐varying digraphs

2023-07-17

0

OA

AI

DOI：10.1049/cth2.12526

原文链接

查看原文

摘要

En 中文

AbstractIn this paper, a consensus based fully distributed optimization algorithm is proposed for solving economic dispatch problem (EDP) in smart grid. Since the incremental cost of all buses reach consensus when the optimal solution is achieved, it is selected as a consensus variable. An additional variable at each bus, called “surplus” is added to record the local power mismatch, which is used as a feedback variable to purse the balance between power supply and demand. Different from most of the existing distributed methods which require the communication network to be balanced, the algorithm uses a row random matrix and a column random matrix to precisely steer all the agents to asymptotically converge to a global optimal solution over a time‐varying directed communication network. Due to the use of a fixed step size, the proposed algorithm also outperforms other algorithms in terms of convergence speed. The graph and eigenvalue perturbation theories are employed for the algorithm convergence analysis, and the upper bound of the parameters required for convergence is given theoretically. Finally, the performance and scalability of the proposed distributed algorithm are verified by several case studies conducted on the IEEE 14‐bus power system and a 200‐node test system.

A distributed consensus based algorithm for economic dispatch over time‐varying digraphs

摘要

AI总结

期刊

机构

引用论文