Thermal image, or thermogram, becomes a new type of signal for machine condition monitoring and fault diagnosis due to the capability to display real-time temperature distribution and possibility to indicate the mach...Thermal image, or thermogram, becomes a new type of signal for machine condition monitoring and fault diagnosis due to the capability to display real-time temperature distribution and possibility to indicate the machine’s operating condition through its temperature. In this paper, an investigation of using the second-order statistical features of thermogram in association with minimum redundancy maximum relevance (mRMR) feature selection and simplified fuzzy ARTMAP (SFAM) classification is conducted for rotating machinery fault diagnosis. The thermograms of different machine conditions are firstly preprocessed for improving the image contrast, removing noise, and cropping to obtain the regions of interest (ROIs). Then, an enhanced algorithm based on bi-dimensional empirical mode decomposition is implemented to further increase the quality of ROIs before the second-order statistical features are extracted from their gray-level co-occurrence matrix (GLCM). The highly relevant features to the machine condition are selected from the total feature set by mRMR and are fed into SFAM to accomplish the fault diagnosis. In order to verify this investigation, the thermograms acquired from different conditions of a fault simulator including normal, misalignment, faulty bearing, and mass unbalance are used. This investigation also provides a comparative study of SFAM and other traditional methods such as back-propagation and probabilistic neural networks. The results show that the second-order statistical features used in this framework can provide a plausible accuracy in fault diagnosis of rotating machinery.展开更多
针对滚动轴承原始振动信号重要特征信息被较强背景噪声淹没以及提取的时域特征冗余度较高、相关性较强的缺点,提出一种基于最大相关-最小冗余(max-relevance and min-redundancy,mRMR)特征筛选和随机森林的滚动轴承故障诊断研究方法。...针对滚动轴承原始振动信号重要特征信息被较强背景噪声淹没以及提取的时域特征冗余度较高、相关性较强的缺点,提出一种基于最大相关-最小冗余(max-relevance and min-redundancy,mRMR)特征筛选和随机森林的滚动轴承故障诊断研究方法。首先将原始信号进行自适应噪声完整集成经验模态分解(CEEMDAN)得到一系列固有模态分量(IMFs),分析IMF并去掉高频噪声和一部分虚假分量,再将信号进行重构并提取其时域特征,通过mRMR去除冗余性和相关性较高的特征向量,使筛选出的特征子集与标签有最大的依赖性,最后将该特征子集输入到随机森林分类器进行分类。实验表明,mRMR具有优良的特征搜索策略,重要特征均靠前得到选取,仅需3个特征便能达到较高的分类准确率,效率高于其余特征选择算法。展开更多
提出一种基于改进最大相关最小冗余判据(maximal relevance and minimal redundancy,mRMR)的暂态稳定评估特征选择方法。首先对标准mRMR方法进行改进,在最大相关、最小冗余判据中引入一个权重因子以细化对特征相关性和冗余性的度量。然...提出一种基于改进最大相关最小冗余判据(maximal relevance and minimal redundancy,mRMR)的暂态稳定评估特征选择方法。首先对标准mRMR方法进行改进,在最大相关、最小冗余判据中引入一个权重因子以细化对特征相关性和冗余性的度量。然后,考虑相量测量单元可以提供的故障后实测信息,构造由系统特征构成的原始特征集,将改进的mRMR应用于特征选择。通过增量搜索算法得到一组嵌套的候选特征子集,并使用支持向量机分类器验证各候选特征子集的分类性能,选择得到具有最大分类正确率的特征子集。基于新英格兰39节点系统和IEEE 50机测试系统的算例结果验证了所提特征选择方法的有效性。展开更多
文摘Thermal image, or thermogram, becomes a new type of signal for machine condition monitoring and fault diagnosis due to the capability to display real-time temperature distribution and possibility to indicate the machine’s operating condition through its temperature. In this paper, an investigation of using the second-order statistical features of thermogram in association with minimum redundancy maximum relevance (mRMR) feature selection and simplified fuzzy ARTMAP (SFAM) classification is conducted for rotating machinery fault diagnosis. The thermograms of different machine conditions are firstly preprocessed for improving the image contrast, removing noise, and cropping to obtain the regions of interest (ROIs). Then, an enhanced algorithm based on bi-dimensional empirical mode decomposition is implemented to further increase the quality of ROIs before the second-order statistical features are extracted from their gray-level co-occurrence matrix (GLCM). The highly relevant features to the machine condition are selected from the total feature set by mRMR and are fed into SFAM to accomplish the fault diagnosis. In order to verify this investigation, the thermograms acquired from different conditions of a fault simulator including normal, misalignment, faulty bearing, and mass unbalance are used. This investigation also provides a comparative study of SFAM and other traditional methods such as back-propagation and probabilistic neural networks. The results show that the second-order statistical features used in this framework can provide a plausible accuracy in fault diagnosis of rotating machinery.
文摘针对滚动轴承原始振动信号重要特征信息被较强背景噪声淹没以及提取的时域特征冗余度较高、相关性较强的缺点,提出一种基于最大相关-最小冗余(max-relevance and min-redundancy,mRMR)特征筛选和随机森林的滚动轴承故障诊断研究方法。首先将原始信号进行自适应噪声完整集成经验模态分解(CEEMDAN)得到一系列固有模态分量(IMFs),分析IMF并去掉高频噪声和一部分虚假分量,再将信号进行重构并提取其时域特征,通过mRMR去除冗余性和相关性较高的特征向量,使筛选出的特征子集与标签有最大的依赖性,最后将该特征子集输入到随机森林分类器进行分类。实验表明,mRMR具有优良的特征搜索策略,重要特征均靠前得到选取,仅需3个特征便能达到较高的分类准确率,效率高于其余特征选择算法。
