一种融合RST与模糊聚类的故障诊断方法

摘要

In order to process the abundant information in fuzzy clustering method, one fusion fault diagnosis method is pro-posed based on attribution reduction algorithm of rough set theory and fuzzy equal relationship clustering. The attribution reduc-tion algorithm is applied to reduce the redundant information, and then the fuzzy equal relationship clustering method is used to get the diagnosis clustering results. This method can not only effectively reduce the iteration numbers of calculating the fuzzy equal relationship matrix, but lower the clustering numbers comparing with the only fuzzy equal relationship clustering method. The examples are applied to test its validity and high efficiency.%　　针对模糊聚类方法中存在冗余信息的问题，提出一种融合粗糙集属性约简和模糊等价关系的故障诊断方法，通过应用粗糙集属性约简算法对冗余数据的处理后再应用模糊等价关系聚类获取聚类结果。该方法与单采用模糊等价关系聚类法相比，不仅能够有效减少模糊等价关系矩阵求解中的迭代次数，而且获得聚类数也得到有效降低，并通过实例验证了该方法的有效性和高效性。