A broad learning-based state estimation method for power system. (August 2022)

• Record Type: Journal Article
• Authors: Luo, Lizi; Wang, Jie; Zhou, Suyang; Lou, Guannan; Sun, Jinsheng
• Is Part Of: Energy reports. Volume 8(2022)Supplement 5
• Page Start: 1227
• Page End: 1235
• Publication Date: 2022-08
• Subjects: Broad learning system -- Incremental learning -- Power system -- State estimation; Power resources -- Periodicals; Energy industries -- Periodicals; Power resources; Periodicals; Electronic journals; 621.04205
• DOI: 10.1016/j.egyr.2022.02.299
• Languages: English
• ISSNs: 2352-4847

Abstract

Abstract: With the fast growth of developing smart grid technology in the last several years, traditional state estimation methods have proven insufficient for the real-time and precise study of emerging complex power systems. The state assessment of electric grid depending on deep learning and other related intelligent algorithms often encounter problems such as time-consuming model training and easy to fall into local optimum. In this background, this research provides a broad learning-based state estimation approach for power system, which is quite different from the existing artificial intelligence algorithms. Due to the operation theory of matrix pseudo-inverse, the broad learning system can not only swiftly compute the connection weights between various network layers, but also has the advantage of incremental learning. Finally, the proposed method is verified based on IEEE 39-node system combined with actual load data.