基于多信息融合的换流站直流设备健康状态决策树评价模型  

A Decision Tree Evaluation Model for Health Status of DC Equipment in Converter Station Based on Multi-information Fusion

作　　者：石延辉[1] 杨洋[1] 阮彦俊 张博 洪乐洲[1] SHI Yanhui;YANG Yang;RUAN Yanjun;ZHANG Bo;HONG Lezhou(EHV Power Transmission Company Guangzhou Bureau of China Southern Power Grid Co.,Ltd.,Guangzhou 510000,China)

机构地区：[1]中国南方电网有限责任公司超高压输电公司广州局,广东广州510000

出　　处：《微型电脑应用》2024年第7期97-101,共5页Microcomputer Applications

摘　　要：针对换流站直流设备间交互信息融合能力较差、小样本与大样本难以均衡的问题,设计基于多信息融合的换流站直流设备健康状态决策树评价模型。使用同质多传感器多信息数据融合方法对多信息实施融合处理;利用k近邻方法和层次聚类方法划分换流站直流设备健康状态量,提取健康状态量的有效信息作为输入,构建决策树评价模型,输出换流站直流设备健康状态评价结果。实验结果表明,该模型具备较好的多信息融合能力和样本均衡能力,且可从不同角度实现换流站直流设备健康状态评价,评价结果较为准确。Aimed at the problems of poor mutual information fusion ability between DC equipment in converter station and difficulty in balancing small samples and large samples,a decision tree evaluation model for health status of DC equipment in converter station is designed based on multi-information fusion.The homogeneous multi-sensor and multi-information data fusion method is used to fuse them.The k-nearest neighbor method and hierarchical clustering method are used to divide the health status quantities of DC equipment in the converter station,and the effective information of the health status quantities is extracted as input to build a decision tree evaluation model,and the health status evaluation results of DC equipment in the converter station is output.The experimental results show that the model has better multi-information fusion ability and sample balance ability,and can realize the health status evaluation of DC equipment in converter station from different angles,and the evaluation results are more accurate.

关 键 词：多信息融合 换流站 直流设备 健康状态 决策树 评价模型 

分 类 号：TM72[电气工程—电力系统及自动化]