电子测量与仪器学报

梁晓雯,蒋爱平,王国涛,李响,薛永越.参数优化决策树算法的密封继电器多余物信号识别技术[J].电子测量与仪器学报,2020,34(1):178-185

参数优化决策树算法的密封继电器多余物信号识别技术

Sealed relay loose particle signal recognition technology based on decision tree algorithm of parameter optimization

DOI：

中文关键词: 继电器多余物 PIND 决策树

英文关键词:relay loose particles PIND decision tree

基金项目:国家自然科学基金（51607059，51077022）、黑龙江省自然科学基金（QC2017059）、黑龙江省博士后基金（LBH Z16169）、黑龙江省高校基本科研业务费（HDRCCX-201604）、黑龙江省教育厅科技成果培育（TSTAU C2018016）、黑龙江大学校内项目（HDJMRH201912，2012TD007，QL2015）资助

作者	单位
梁晓雯	1.黑龙江大学电子工程学院
蒋爱平	1.黑龙江大学电子工程学院
王国涛	1.黑龙江大学电子工程学院,2.哈尔滨工业大学军用电器研究所
李响	1.黑龙江大学电子工程学院
薛永越	1.黑龙江大学电子工程学院

Author	Institution
Liang Xiaowen	1. Electronic Engineering College of Heilongjiang University of Technology
Jiang Aiping	1. Electronic Engineering College of Heilongjiang University of Technology
Wang Guotao	1. Electronic Engineering College of Heilongjiang University of Technology,2. Military Apparatus Research Institute of Harbin Institute of Technology
Li Xiang	1. Electronic Engineering College of Heilongjiang University of Technology
Xue Yongyue	1. Electronic Engineering College of Heilongjiang University of Technology

摘要点击次数: 603

全文下载次数: 0

中文摘要:

在航天密封继电器的生产过程中，多余物检测是一个必不可少的过程。微粒碰撞噪声检测（PIND）是我国军标规定的航天电子元器件多余物检测方法。针对传统检测方法中多余物信号和组件信号的误判问题，使用参数优化的决策树算法对检测信号进行分类。通过对比组件信号和多余物信号时域频域波形，提取出最具有代表性的特征作为决策树的分裂属性。采用网格搜索法寻找决策树最优分裂准则和分裂深度，然后采用参数优化的决策树建立分类模型。实验结果表明，采用参数优化的决策树算法进行多余物检测信号分类有效提高了分类准确率、G means值和F measure值等各项分类指标值。

英文摘要:

Detection of the loose particles is urgently required in the Aerospace seal relay production processes. Particle impact noise detection (PIND) is a national aerospace electronic component loose particles detection method. Aiming at the misjudgment of the loose particles signal and component signal in the traditional detection method, this paper uses the parameter optimized decision tree algorithm to classify the detection signal. After comparing the waveforms of the component signal and the loose particle signal in the time domain and the frequency domain, select the most representative feature as the split attribute of the decision tree. The grid search method is used to find the optimal splitting criterion and splitting depth of the decision tree, then use the parameter optimization decision tree to establish the classification model. The experimental results show that using the parameter optimized decision tree algorithm to classify the loose particles detection signals can effectively improve the accuracy of the classification results, G means value and F measure value.

查看全文查看/发表评论下载PDF阅读器