基于关联规则挖掘与组合赋权-云模型的整流变压器异常跳闸故障诊断方法  

Fault Diagnosis Method for Rectifier Transformer Abnormal Trip Based on Association Rule Mining and Combination Weighting-cloud Model

作　　者：徐明磊 XU Minglei(Qingtongxia Aluminium Co.,Ltd.,Wuzhong 751603,China)

机构地区：[1]青铜峡铝业股份有限公司,宁夏吴忠751603

出　　处：《微型电脑应用》2024年第12期237-240,共4页Microcomputer Applications

摘　　要：整流变压器作为电力工业的主要元件,关乎整流系统运行的可靠性和可用性。当整流变压器发生故障时,若不能及时控制会造成电源中断等问题。为此,提出基于关联规则挖掘与组合赋权—云模型的整流变压器异常跳闸故障的诊断方法。该方法是通过关联规则挖掘技术设定分析流程,以数据支持度和关联规则置信度,定义整流变压器运行数据之间联系性。在组合赋权—云模型中划分数据类型,获取整流变压器异常跳闸的具体特征。应用移相滤波原理计算功率损耗,利用对应特征向量诊断变压器故障,实现异常跳闸故障诊断。实验结果表明,以引起整流变压器异常跳闸的5种情况及正常情况为测试条件,应用所提方法能够及时进行故障判断,并正确分类不同故障类型,具有应用价值。As the main component in the power industry,the rectifier transformer is related to the reliability and availability of the rectifier system operation.When the rectifier transformer has natural or artificial fault,if it can not be controlled in time,it causes power interruption and other problems.Therefore,the diagnosis method of abnormal tripping fault of rectifier transformer based on association rule mining and combined weighting-cloud model is proposed.The analysis process is set by association rule mining technology and the connection between the operation data of rectifier transformer is defined by data support and association rule confidence.The data types are divided in the combined weighting-cloud model to obtain the specific characteristics of abnormal trip of rectifier transformer.The power loss is calculated under the principle of phase-shifting filter and the corresponding eigenvector is used to diagnose the transformer fault,realizing the design of fault diagnosis method for abnormal trip.The experimental results show that under the test conditions of five conditions that cause abnormal trip of rectifier transformer and normal conditions,the method in this paper can form judgment in time and correctly classify different fault types,which has application value.

关 键 词：关联规则 整流变压器 异常跳闸 故障诊断 

分 类 号：TP311.5[自动化与计算机技术—计算机软件与理论]