摘要
ZTZ96A坦克是我国的主战装备,其智能故障诊断一直是部队亟待解决的问题,因此本文研究的目的在于解决其智能故障诊断难的问题。在ZTZ96A坦克炮控系统故障诊断中,可使用的样本有限,故障特征又都呈现出非线性的特点,并且故障都很难被定位到具体位置'基于以上问题,本文提出一种基于改进支持向量机的ZTZ96A坦克炮控系统故障诊断的方法。该方法首先对ZTZ96A坦克炮控系统历史运行数据进行预处理,然后使用PCA方法对ZTZ96A坦克炮控系统参数数据进行特征提取,其次根据提取的特征数量和故障模式数量建立决策模型,利用处理好的历史运行数据对其进行训练,最后使用两种模型性能评价指标对训练好的模型进行性能评价。实验结果表明该方法的诊断精度高、误差小、速度快。实车应用表明该方法能够获得非常好的学习和扩展能力、置信度高、可行性强。
ZTZ96A Tank is the main battle equipment in China,and intelligent fault diagnosis of its has always been an urgent problem for troops to be solved.Therefore,the purpose of the research of this paper is to solve the problem of its intelligent fault diagnosis difficult.In the fault diagnosis of the gun-control-system of the ZTZ96A Tank,the available samples are limited,and the fault characteristics are nonlinear,and the faults are difficult to be located at specific locations.Based on the above problems,a method for fault diagnosis of the gun-control-system of the ZTZ96A Tank is proposed based on the improved support vector machine.Firstly the historical running-data of the gun-control-system of ZTZ96A Tank is preprocessed.Secondly,PCA method is used to extract the feature data of the gun-control-system of ZTZ96A Tank.Thirdly,according to the quantity of the extracted feature and the number of failure modes,the decision model is extablished,and the processed historical running-data is used to train it,And finally the performance evaluation of the trained model is performed using two model performance evaluation indicators.Experimental results show that the method has high diagnostic accuracy and small error.The real vehicle application shows that the method can obtain very good learning and expansion ability,high confidence and strong feasibility.
作者
李英顺
张童鑫
伊枭剑
LI Yingshun;ZHANG Tongxin;YIN Xlaojian(Beijing Institute of Petrochemical Technology School of Information Engineering,Beijing 102617;Shenyang University of Technology Chemical Process Automation Institute,Liaoning ShenYang 110000;Beijing university of technology college of mechatronics,Bejing 100081,China;China North Vehicle Research Institute General Technology Department,Beijing 100072,China)
出处
《自动化与仪器仪表》
2019年第7期155-160,共6页
Automation & Instrumentation
基金
青年科学基金项目(No.71801196)
