加权聚类分析在设备运行监控中的应用

The Application of Feature Weight's Clustering Analysis to Monitor Equipments's Circulating

论文采用数据挖掘中的聚类分析算法对流程企业的大量而正常的历史数据进行分析,首先采用基于欧几里德距离的加权K-means算法建立了参数的聚类模型,然后用相关系数法计算每个簇团中的参数和中心参数的相似度,得到了相似度阈值。以此为基础,可以对设备的运行状况进行监控,从而起到设备运行优化和故障预警的作用。

To monitor flowing industry's production,the l ar ge and right history data are analyzed by clustering algo-rithm.The equipment 's parameters clustering models are built by Feature Weight's K-means algorithm .Similarity relations are calculated between equipment's parameters in clusteri ng cluster and the cluster centroid by correlation coefficient method,then the key threshold is got.Based on the models and similarity relations,we can opt imize equipments process and monitor faults.