Power grid fault detection method based on cloud platform and improved isolated forest
Weiliang Li, Yan Li, Jiawei Shi, Kai Chao, Xiaoliang Zhang · 2021
In order to avoid equipment failures that may occur during power generation and transportation, real-time fault detection and diagnosis are required. Based on the cloud computing architecture, this paper proposes a fault detection method suitable for the power grid cloud platform. The method first improved the isolated forest model. After that, a power grid cloud platform was built, which was divided into three layers. The field layer collects the data generated by the various monitoring systems at the equipment layer and uploads it to the cloud center layer after data preprocessing; The cloud center layer uses the previously stored data to parallelize and iteratively train the improved isolated forest model, and the trained model can classify the faults that occur. Finally, the accuracy of the method is proved through experiments.