Identification of Power Quality Disturbances Considering New Energy Intake

• Published in: 2023 3rd International Conference on New Energy and Power Engineering (ICNEPE) pp. 886 - 889
• Main Authors: Liu, Chang, Ji, Xiu, Meng, Xiangdong, Guo, Zhongqi
• Format: Conference Proceeding
• Language: English
• Published: IEEE 24.11.2023
• Subjects: Compounds; decision tree; Decision trees; Feature extraction; new energy; Optimization; Power quality; Power systems; Random forests
• DOI: 10.1109/ICNEPE60694.2023.10429731

Power quality can be one of the most important indicators for assessing the sensitivity of new energy intake. With the increasing access to renewable energy sources in the power system, the issue of power quality has become an important consideration. This paper proposes an identification method based on an integrated decision tree to identify and analyze the perturbations in power quality correctly. Firstly, real-time data of power quality compound disturbances are collected, and through preprocessing as well as feature extraction, we obtain a set of characteristic vectors. Then, utilizing the idea of integrated learning, an integrated model consisting of multiple decision trees is constructed. In the integrated model, each decision tree independently classifies the feature vectors, and the final classification results are obtained according to the voting mechanism. Finally, the effectiveness and accuracy of the method are verified through experiments, which can accurately identify and classify power quality compound disturbances with high identification accuracy and robustness. Therefore, applying the method in the power system has important practical significance.