典型快速谱峭图算法的研究及应用

Typical fast kurtogram algorithm and its application

针对工程实际中状态监测下的具体机械部件在特定运行环境中发生故障的原因及类型具有典型性,在分析快速谱峭图算法的基础上,提出了典型快速谱峭图算法。该算法借鉴快速谱峭图算法分层和选取中心频率的思想,以典型故障的三倍特征频率为带宽,快速定位谱峭度最大的频率区间,并将该算法应用于共振解调技术滤波器参数的确定。为了验证该算法的有效性,在滚动轴承故障诊断试验台上进行了振动测试,采用基于典型快速谱峭图算法的共振解调技术进行故障诊断和类型识别。结果表明,基于典型快速谱峭图算法的共振解调技术能够较好的诊断典型故障,具有一定的工程应用价值。

In practical engineering,the causes and types of occurring faults on specific mechanical components are typical.On the basis of analyzing fast kurtograms,the typical fast kurtogram algorithm was presented here.With the fast kurtogram ideas of hierarchy and selecting central frequency,three times of characteristic frequencies of typical faults were taken as a bandwidth to quickly find the frequency interval of the maximum spectral kurtosis.This algorithm was used to automatically determine filter＆#39;s parameters for resonance demodulation technique.In order to verify the validity of this algorithm,vibration tests were conducted on a rolling bearing fault diagnosis test platform.The fault diagnosis and type recognition were implemented using the reonance dempdulation technique based on the typical fast kurtogram alororithm. The results demonstrated that the proposed algorithm is reliable and can meet the requirements of engineering application.