摘要
基于小波包-模糊C均值聚类诊断柴油机曲轴轴承故障的方法,通过对测试数据进行小波包分频滤波,提取特征参数,用模糊C均值聚类对所选评价指标进行聚类,得到优化的分类矩阵和聚类中心,建立故障诊断的标准样本。计算待识样本与标准模式样本的贴近度,实现故障模式识别。该方法应用于曲轴轴承故障的诊断,取得了良好的效果。
A method based on Wavelet Packet - Fuzzy C Mean clustering algorithm is putted forward to diagnose the crankshaft bearing of a diesel engine. Through the testing data is processed with wavelet packet filter, the frequency is separated, so the characteristic parameters can be extracted by frequency band energy cumulation. And then, the optimized classified matrix and clustering centers, through using the Fuzzy C Mean algorithm to cluster the indexs, formed the standard samples for fault diagnosis. By calculating the close degree between the testing sample and the standard one, the fault pattern is identified. This method is applied in crankshaft bearing fault diagnosis and gets a good result.
出处
《军事交通学院学报》
2009年第2期44-47,68,共5页
Journal of Military Transportation University
