【摘要】 针对滚动直线导轨副退化状态难以识别、缺少自适应识别的评估指标等问题，提出了一种基于局部均值变分模态分解(local mean-variational mode decomposition, LM-VMD)和Jensen-Rényi divergence(JRD)指标的退化状态识别方法。首先对采集到的全寿命振动信号进行有效信息的提取；其次将信号进行局部均值分解得到模态分解数量K值；然后根据K值进行变分模态分解，通过奇异值分解提取信号中的本征模态函数-奇异值谱概率分布特征；最后计算JRD指标，根据JRD指标的拐点，实现滚动直线导轨副退化状态的自适应识别，并与预紧拖动力和滚道形貌特征进行对比分析。结果显示，所提方法能够准确识别导轨副的退化状态。更多还原

【Abstract】 Here, aiming at problems of difficulty in recognizing degradation state of rolling linear guide pair and lack of evaluation indexes for adaptive recognition, a degradation state recognition method based on local mean variational mode decomposition(LM-VMD) and Jensen-Rényi divergence(JRD) index was proposed. Firstly, effective information was extracted from the collected full-life vibration signals. Then, local mean decomposition was performed for signal to obtain the mode decomposition number K value. Variational mode decomposition was performed for signal according to K value, and the probability distribution characteristics of intrinsic mode function-singular in signal were extracted with singular value decomposition. Finally, JRD index was calculated. According to inflection point of JRD index, adaptive recognition of degradation state of rolling linear guide pair was realized. Compared with pre-tightening traction force and track morphology characteristics, the results showed that the proposed method can correctly recognize degradation state of rolling linear guide pair.更多还原