A Data-driven AC Optimal Power Flow Using Extreme Learning Machine. Issue 1 (1st February 2023)

• Record Type: Journal Article
• Authors: Liang, Weichen; Wang, Yajuan; Zhao, Zhiyu; Liu, Bo; Li, Xuan
• Is Part Of: Journal of physics. Volume 2418:Issue 1(2023)
• Publication Date: 2023-02-01
• Subjects: Physics -- Congresses; 530.5
• DOI: 10.1088/1742-6596/2418/1/012105
• Languages: English
• ISSNs: 1742-6588

Abstract

Abstract: With the increasing integration of renewable energy (RE), AC Optimal Power Flow (AC OPF) becomes a necessary foundation for electricity system which has a high level of renewable energy generation. However, most current studies of AC OPF are not applicable due to the requirements of high computation speed in practical applications. To this end, we propose a data-driven method using Extreme Learning Machine (ELM) for getting the AC OPF optimal solution in a faster way. This approach can map the relationship between the optimal operation results and variations of REs and loads, avoiding the time-consuming solving process of AC OPF. Moreover, an ELM network structure suitable for AC OPF is designed to significantly improve the computational speed of ACOPF with acceptable accuracy. This method was applied to the RTS-79 system for improving the calculation efficiency of the AC OPF.