%0 Journal Article
%T Adaboost_SVM集成模型的滚动轴承早期故障诊断<br>Early Fault Diagnosis of Rolling Bearing based on Ensemble Model of Adaboost SVM
%A 陈法法
%A 杨晶晶
%A 肖文荣
%A 程珩
%A 张发军
%J 机械科学与技术
%D 2018
%X 针对滚动轴承早期故障诊断中故障特征微弱难以有效检测的问题，提出一种基于Adaboost提升支持向量机（Support vector machines，SVM）集成学习模型的滚动轴承早期故障诊断方法。首先以Cincinnati大学实测的滚动轴承全寿命振动数据为基础，采用特征参数跟踪法，建立特征参数的趋势分析，并据此选择用于滚动轴承早期故障诊断的敏感特征参量，然后通过构造Adaboost提升SVM集成学习模型，并将其应用于滚动轴承的早期故障检测中。AdaBoost能够自适应地提升单一SVM的分类性能，相对于传统的单一SVM分类器Adaboost_SVM稳定性最好，早期故障诊断准确率最高。实验结果表明，结合优选的敏感特征参量，Adaboost_SVM能有效地诊断滚动轴承的早期故障模式。<br>Aiming at the early fault features of roller bearings are too weak so that it is difficult to get effective identification, an early fault diagnosis method based on Adaboost_SVM integrated learning model for rolling bearing early fault diagnosis is proposed in this paper. Firstly, those sensitive feature parameters are selected through analyzing the feature parameters developing trend based on the rolling bearing vibration data in whole life process acquired by the university of Cincinnati. Then, the ensemble learning model with Adaboost_SVM is constructed, and applied to the rolling bearing early fault identification. Adaboost can adaptively improve the classification performance of conventional SVM. Compared with the traditional single SVM classifier, Adaboost_SVM has the best stability and the highest early fault diagnosis accuracy. The experimental results show that the Adaboost_SVM can effectively diagnose rolling bearing early failure modes with those sensitive feature parameters
%K 集成学习模型
%K 支持向量机
%K Adaboost
%K 滚动轴承
%K 故障检测&lt
%K br&gt
%K Ensemble learning model
%K support vector machine (SVM)
%K Adaboost
%K roller bearing
%K fault detection
%U http://journals.nwpu.edu.cn/jxkxyjs/CN/abstract/abstract6940.shtml