AUTHOR=Chen Ran , Zhou Li , Xiong Chuanyu , Xu Hanping , Zhang Zhaoyang , He Xuhui , Dong Qingguo , Wang Can TITLE=Islanding detection method for microgrids based on CatBoost JOURNAL=Frontiers in Energy Research VOLUME=10 YEAR=2023 URL=https://www.frontiersin.org/journals/energy-research/articles/10.3389/fenrg.2022.1016754 DOI=10.3389/fenrg.2022.1016754 ISSN=2296-598X ABSTRACT=

The occurrence of unintentional islanding will seriously threaten the stable operation of a microgrid (MG). Therefore, detecting the islanding of an microgrid timely is an important premise to ensure the microgrid operates safely and stably. However, the problem of dead zone exists in the traditional islanding detection process because the threshold of various electrical feature quantities of the point of common coupling (PCC) cannot be determined effectively. To solve this problem, an islanding detection method based on CatBoost is proposed for an microgrid. The novelty of this method lies in two aspects: 1) To reduce the error brought by the electrical feature quantities with weak correlation in the process of islanding detection, an analysis method based on the Spearman correlation coefficient is used to extract the electrical feature quantities closely related to islanding detection. 2) To determine the threshold of the electrical feature quantities more accurately and reduce the dead zone of island detection, an integrated learning machine is used to dig out correlations between the electrical feature quantities and the operation of an microgrid. The performance of the proposed islanding detection method is verified based on the modified IEEE13-bus system. The results of the example verify that the proposed islanding detection can achieve higher detection accuracy in cases of grid-connected interference and line faults.