遗传规划提取优化特征在轴承寿命预测中的应用  

Application of optimization feature extraction from bearing based on genetic programming for life prediction

作　　者：王豪 董广明[1] 陈进[1] WANG Hao;DONG Guang-ming;CHEN Jin(School of Mechanical Engineering,Shanghai Jiao Tong University,Shanghai 200240,China)

机构地区：[1]上海交通大学机械与动力工程学院,上海200240

出　　处：《振动工程学报》2021年第3期626-632,共7页Journal of Vibration Engineering

基　　金：国家重点研发计划资助项目(2019YFB2004600);国家自然科学基金资助项目(51575339,51775330)。

摘　　要：在滚动轴承故障诊断领域中,针对轴承剩余寿命预测这一关键问题,提出了一种基于GP(遗传规划)提取特征的方法,该方法将多个特征组合为一个特征树,实现多维输入到一维输入的转换,并用改良的适应度评价特征树的优良性,经过反复迭代,最后输出适应度最大的特征树,该特征树对应的特征值曲线在时域上最接近线性变化,将其作为一个独立的特征,称为优化特征。最后利用轴承全寿命振动信号,以优化特征为模型预测轴承剩余使用寿命,验证了算法预测的准确性。In the field of the rolling bearing fault diagnosis,the remaining useful life prediction is very important.This paper proposes an approach based on genetic programming for features extraction,and multiple features are combined into a feature tree,so multi-dimensional input transfers to single-dimensional input.Furthermore,using the improved fitness to estimate the quality of the feature tree.After repeated iterations,the final output is the feature tree,whose fitness is the maximal.Besides,the curve of this feature tree is the closest to the linear trend in the time domain,hence,it is regarded as an independent feature named the optimization feature.This paper uses the vibration signal of the entire bearing life to predict the remaining useful life of the bearing with the optimization feature as the prediction model,and verifies the accuracy of prediction.

关 键 词：故障诊断 滚动轴承 特征提取 遗传规划 寿命预测 

分 类 号：TH165.3[机械工程—机械制造及自动化] TH133.33