基于JRD和CUSUM的滚动轴承性能退化状态识别与评估

Performance degradation status identification and assessment of rolling element bearing based on the JRD and CUSUM

退化特征提取是滚动轴承性能退化状态识别和评估的关键,JRD克服了传统特征无法准确反应轴承当前技术状态的不足,但在全寿命阶段上稳定性、单调性差,应用累积和(CUSUM)对其进行改进,从而准确识别和评估轴承性能退化状态。应用小波包变换对原始信号进行降噪;计算不同技术状态下信号的Renyi熵,并对比与标准状态的相似程度得出JRD值,作为滚动轴承退化状态特征;应用CUSUM增强JRD值对于寿命微弱变化的敏感性及轴承全寿命的单调性。通过试验验证,滚动轴承性能退化状态的识别率能达到100%,同时能够分阶段、单调性地评估轴承性能退化状态。

Degradation feature extraction is the key of the identification and assessment of bearings’degradation status.The JRD overcomes the shortcoming of traditionally used features that cannot accurately reflect the current status of bearing,but its stability and monotonicity are poor in the whole life period.So,the CUSUM was introduced to accurately identify and evaluate the status of bearing performance degradation.The wavelet packet transform was used to denoise the original signal,the Renyi entropy of different status of bearings was calculated,and the modified JRD value,adopted as a new bearings’degradation feature,was calculated by comparing the similarity to the standard status.The CUSUM can enhance the sensitivity to weak changes and the monotonicity in the whole life.According to the experimental data,the recognition rate of the performance degradation of rolling bearings can reach 100%and the degradation status evaluation can be implemented by stages and in monotonicity.