为解决传统故障诊断需要先验信息干预的问题,提出了一种基于模糊免疫网络聚类分析的故障诊断方法。同其它聚类算法相比,该算法能克服动态确定聚类数和分类边界模糊问题,是一种较好的无监督算法,并通过实例验证了该算法在风机故障诊断中的有效性。 In order to solve the problem of prior information intervening in traditional fault diagnosis, a fault diagnosis method based on fuzzy immune network clustering analysis is proposed. Compared with other clustering algorithms, the proposed algorithm overcomes the problem of dynamically determining the number of clusters and the classification of boundary ambiguities. It is a better unsupervised algorithm and validates the effectiveness of the algorithm in fan fault diagnosis through examples.