基于小波和Bagging-PNN网络的柴油机轴承故障研究  

A Study of Diesel Engine Bearing Failure Based on Wavelet and Bagging-PNN Networks

作　　者：丁坤岭 王晓峰 舒航 徐可 孙贾梦 Ding Kunling;Wang Xiaofeng;Shu Hang;Xu Ke;Sun Jiameng(School of Electronic and Electrical Engineering;School of Mathematical and Physical Sciences,Chongqing University of Science and Technology,Chongqing 401331,China)

机构地区：[1]重庆科技大学电子与电气工程学院,重庆401331 [2]重庆科技大学数理科学学院,重庆401331

出　　处：《黑龙江工业学院学报（综合版）》2024年第7期97-104,共8页Journal of Heilongjiang University of Technology(Comprehensive Edition)

基　　金：重庆市自然科学基金面上项目(项目编号:CSTB2022NSCQ-MSX0398、CSTB2022NSCQ-MSX1425);重庆科技大学硕士研究生创新计划项目“复杂场景下基于深度学习双目立体视觉和激光雷达车前目标检测跟踪与深度感知算法”(项目编号:YKJCX2220414)。

摘　　要：针对柴油机故障诊断速度慢、诊断模型准确率低等问题。提出一种基于小波和Bagging-PNN网络的柴油机轴承故障诊断法。首先,利用时域、频域对采样后的故障数据进行分析,通过小波分析对数据进行去噪处理;然后,将Bagging算法与概率神经网络(Probabilistic Neural Network, PNN)进行融合,通过多个PNN分类器以相同的方式进行投票建立柴油机轴承故障分类模型,提高诊断准确度;最后,通过对比实验表明基于小波和Bagging-PNN的柴油机轴承故障诊断方法的识别准确性有明显提高。Aiming at the problems of slow speed and low accuracy of diagnostic model for diesel engine fault,a diesel engine bearing fault diagnosis method based on wavelet and Bagging-PNN network is proposed.First,the sampled fault data are analyzed in time and frequency domains,and the data are denoised by wavelet analysis;then,the Bagging algorithm is fused with Probabilistic Neural Network(PNN),and the data obtained by multiple PNN classifiers voting in the same way are used as the final classification results.The output of the denoised data is used to establish a diesel engine bearing fault classification model to improve the diagnostic accuracy;finally,the comparison experiments show that the recognition accuracy of the wavelet and Bagging-PNN based diesel engine bearing fault diagnosis method is significantly improved.

关 键 词：柴油机轴承 故障诊断 BAGGING PNN 小波分析 

分 类 号：TH133.3[机械工程—机械制造及自动化] TP18[自动化与计算机技术—控制理论与控制工程]