基于MED与小波包熵的矿用排水泵轴承故障特征提取方法研究

Study on Fault Feature Extraction Method of Mine Drainage Pump Bearing Based on MED and Wavelet Packet Entropy

煤矿排水泵运行的可靠性关系到煤矿工人生命和生产的安全。为了提高煤矿排水泵轴承故障诊断的精度,提出一种基于最小熵反褶积(MED)和小波包熵相结合的故障特征提取方法。该方法在MED降噪基础上利用小波包进行分解重构,再结合信息熵理论求取小波包熵值,突出了信号中有效冲击成分,克服了小波能量特征提取的局限性。通过对水泵轴承振动200组数据进行分析验证,该方法准确率达到100%,具有较高的应用价值。

The reliability of coal mine drainage pump operation is related to the life and production safety of coal mine workers. A fault feature extraction method based on minimum entropy deconvolution(MED) and wavelet packet entropy was proposed to improve the accuracy of bearing fault diagnosis in coal mine drainage pump. Based on the MED noise reduction, the wavelet packet was decomposed and reconstructed, and the wavelet packet entropy was obtained by combining the information entropy theory, which highlights the effective impact components in the signal and overcomes the limitation of wavelet energy feature extraction. Through the analysis of 200 sets of data of pump bearing vibration, it was verified that the accuracy of this method is 100% and has high application value.