首页> 外国专利> Oil-filled electrical equipment abnormality diagnosis system, decision tree generation system, decision tree generation method, and decision tree generation program

Oil-filled electrical equipment abnormality diagnosis system, decision tree generation system, decision tree generation method, and decision tree generation program

机译:充油电气设备异常诊断系统,决策树生成系统,决策树生成方法和决策树生成程序

摘要

PROBLEM TO BE SOLVED: To provide a decision tree generation system capable of generating a decision tree suitable for oil-filled electrical equipment, a decision tree generation method, a decision tree generation program, and an abnormality diagnosis system of the oil-filled electrical equipment.SOLUTION: A decision tree generation system includes: a case data storage part 31 for storing case data where a plurality of pieces of attribute information X including density information is associated with abnormality rank information; a purpose group setting part 21 for setting each purpose group by grouping each rank of the abnormality rank information to grouping patterns capable of grouping it into a small abnormal group and a large abnormal group; an information gain ratio calculation part 22 for calculating an information gain ratio G(X) based on each purpose group when dividing an aggregation D into a plurality of subsets by using the threshold value of each piece of attribute information X as a reference, correspondingly to the combination of each piece of attribute information and each grouping pattern; and a node generation part 23 for generating a node by using a piece of attribute information and a threshold value corresponding to a maximum information gain ratio as an attribute and a branch threshold value respectively.
机译:解决的问题:提供一种能够生成适用于充油电气设备的决策树的决策树生成系统,一种决策树生成方法,一种决策树生成程序以及该充油电气设备的异常诊断系统解决方案:决策树生成系统包括:案例数据存储部分31,用于存储案例数据,其中包括密度信息的多个属性信息X与异常等级信息相关联;以及目的组设定部21,通过将异常等级信息的各等级分组为能够将异常等级组分为小异常组和大异常组的分组模式,来设定各目的组。信息增益比计算部分22,用于当通过将每条属性信息X的阈值作为参考将聚集D划分为多个子集时,基于每个目的组来计算信息增益比G(X),对应于每个属性信息和每个分组模式的组合;节点生成部分23,用于通过分别使用一条属性信息和对应于最大信息增益比的阈值作为属性和分支阈值来生成节点。

著录项

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号