摘要
针对采集的供输弹系统测试信号成分复杂,故障难以识别问题,提出一种基于固有时间尺度分解(ITD)与粒予群优化支持向量机(PSO-SVM)的供输弹系统故障诊断方法。首先在时域内使用ITD方法对信号进行分解,对分解产生的分量进行相关系数计算,然后选取与原始信号相关系数大的前5层分量进一步验证在频域内ITD方法的有效性,在频域内提取前5层分量的样本熵值,最后将提取的样本爛值用PSO-SVM对供输弹系统故障进行故障诊断,并与支持向量机(SVM)的诊断结果进行对比,结果表明:PSO-SVM相对于SVM可以提高故障诊断的正确率,正确率高达92.31%。
Aiming at the complex signal components of the collected missile system and the difficulty of identifying the fault,a fault diagnosis method for the missile system based on inherent time scale decomposition(ITD)and particle swarm optimization support vector machine(PSO-SVM)is proposed.First,the ITD method is used to decompose the signal in the time domain,and the correlation coefficient is calculated for the components generated by the decomposition.Then select the first 5 layers of the correlation coefficient with the original signal to further verify the effectiveness of the ITD method in the frequency domain.Extracting sample entropy values of the first 5 layers of components in the frequency domain,Finally,the extracted sample entropy value is diagnosed by the PSO-SVM for the fault of the transport system,and compared with the diagnosis result of the support vector machine(SVM).The results show that PSO-SVM can improve the accuracy of fault diagnosis compared with SVM,and the correct rate is as high as 92.31%.
作者
张航
许昕
潘宏侠
梁海英
ZHANG Hang;XU XIN;PAN Hongxia;LIANG Haiying(Mechanical Engineering Institute,North University of China,Taiyuan 030051,China;Institute of System Identification and Diagnosis Technology,China,Taiyuan 030051,China)
出处
《机械设计与研究》
CSCD
北大核心
2019年第6期197-200,共4页
Machine Design And Research
基金
国家肖然科学基金资助项目(51675491,51175480)。
