关联规则分类的数据流挖掘方法在水电机组故障诊断的应用

摘要

引进一种数据流关联规则分类法(AC-DS),并应用该方法对UCI机器学习库中标准数据集进行分类验证,验证结果表明该方法准确且有效.然后将该方法应用到水电机组故障的诊断分类中,证明该方法的分类精度随着测试样本的增加而增加.该方法对现场不同类型机组故障分类有一定意义.%This paper introduces a kind of data stream associative rule classification method, and this method is used to carry out the classification test of the standard datasets in the UCI machine learning repository. The test results indicate that this method is accurate and effective. And then, this method is used in the fault diagnosis classification of hydro-turbine generating unit, whereby proving that the classification accuracy of this method is improved with an increase in the testing samples. Accordingly, this method is of the certain significance for the in-situ fault classification of generating units of different types.