A parallel and asynchronous state estimation for coupled transmission-distribution networks

• Published in: International journal of electrical power & energy systems Vol. 141; p. 108163
• Main Authors: Tang, Yingqi, Tang, Kunjie, Dong, Shufeng
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.10.2022
• Subjects: Asynchronous state estimation; Communication delays; Coupled transmission-distribution networks; Parallel mechanism; Sensitivity analysis; Parallel mechanism; Sensitivity analysis; Asynchronous state estimation; Communication delays; Coupled transmission-distribution networks
• ISSN: 0142-0615; 1879-3517
• DOI: 10.1016/j.ijepes.2022.108163

•A parallel and asynchronous state estimation method for power system is proposed.•The method contains four improvements to speed up the iteration process.•A sensitivity analysis mechanism to obtain key distribution network is proposed.•The method proposed can effectively handle with communication delays.•The method proposed has low sensitivity to parameter changes. Considering the increasing scale of power systems and gradually coupling between transmission networks (TNs) and distribution networks (DNs), a parallel and asynchronous state estimation (par&async-SE) algorithm is proposed to deal with communication delays. First, the coupled transmission-distribution network state estimation problem is reformulated in line with engineering practice. Then, a TN broadcast transmitting strategy, a synchronous preprocessing technique, and a receive-compute-transmit parallel mechanism are applied to improve conventional asynchronous algorithm. Moreover, a ‘key DN’ concept is introduced and a sensitivity analysis method is proposed to realize key DNs selection. These improvements make the par&async-SE algorithm proposed not only have a higher convergence rate and efficiency but also be less sensitive to parameter settings. Numerical experiments demonstrate that the par&async-SE algorithm has the same accuracy as conventional algorithms. It can also achieve better convergence, reduce computation time, and be less sensitive to parameter settings than the synchronous state estimation algorithm and conventional asynchronous state estimation algorithm when communication delays exist, so that parameter tuning for this self-adaptive algorithm is not necessary in engineering practice. Besides, the par&async-SE algorithm also has bad data robustness when choosing appropriate methods to solve extended-TNSE and DNSE subproblems.