摘要
现代电力系统中,电容式电压互感器(CVT)的绝缘性能对系统运行的稳定性、安全性至关重要。文章提出了一种电容式电压互感器绝缘故障识别方法,构建了包含正常运行状态、已知故障状态的全面数据集,根据同相关系,运用PCA分析法提取关键特征,基于随机森林算法构建故障识别模型,通过实际数据验证其有效性。研究结果显示所提出的方法能够准确识别电容式电压互感器的绝缘故障,且模型经过参数优化后,预测准确率达到100%,运算效率显著提升。
In modern power systems,the insulation performance of capacitive voltage transformers(CVTs)is crucial for the stability and safety of system operation.The article proposes a method for identifying insulation faults in capacitive voltage transformers,constructing a comprehensive dataset that includes normal operating states and known fault states.Based on the same phase relationship,PCA analysis is used to extract key features,and a fault identification model is constructed using the random forest algorithm.The effectiveness of the method is verified through actual data.The research results show that the proposed method can accurately identify insulation faults in capacitive voltage transformers,and after parameter optimization,the prediction accuracy of the model reaches 100%,and the computational efficiency is significantly improved.
出处
《电力系统装备》
2024年第8期93-95,共3页
Electric Power System Equipment
关键词
电容式电压互感器
绝缘故障
故障识别
capacitive voltage transformer
insulation fault
fault identification
