同步压缩小波与极限梯度提升树融合的柴油机失火故障诊断

A Diagnostic Method for Diesel Engine Misfire Based on Integrating of Synchro-Squeezed Wavelet Transform and XGBoost

针对柴油机失火故障诊断特征提取分辨率较低和分类评估容易出现过拟合的问题,提出了一种同步压缩小波变换和极限梯度提升树融合的诊断方法。在不同转速下进行柴油机失火性能试验,采集缸盖振动信号,对信号利用时域统计、同步压缩小波提取特征,再采用局部线性嵌入方法进行特征降维,最后利用极限梯度提升树进行失火评估分类。不同工况与评估方法下的对比实验结果表明,所提方法的分类准确率最高可达99.828%,相比小波包特征提取的评估方法提升至少10%。在低模型复杂度下,所提方法具有最小的模型预测均方根误差,证明了方法的鲁棒性和抑制模型过拟合的能力。

A diagnostic method based on combining synchro-squeezed wavelet transform and XGBoost is proposed to solve the problem of low resolution in feature extraction and overfitting in classification of diesel engine misfire fault diagnosis.Vibrational signals of a cylinder head are acquired through experiments for testing the performance of misfire in diesel engine at different speeds.Features of signals are extracted by using a time-domain statistics method and the synchro-squeezed wavelet transform.Then,the local linear embedding method is adopted to reduce the dimension of features.The XGBoost is finally applied to evaluate misfire situations.Comparative experimental results on different working conditions and evaluation methods show that the classification accuracy of the proposed method is up to 99.828%and increases at least 10%than that of the method based on wavelet packet feature extraction.The proposed method achieves the minimum root mean squared error under low model complexity,and its robustness and ability for restraining model overfitting are verified.