中文摘要：

发动机结构日益复杂,其故障具有多样性和频发性的特点,收集大量故障样本存在很多实施障碍。为了提高车辆发动机的故障识别的效率和准确性,提出了一种新的结合故障树（FTA）和支持向量机（SVM）各自特点,从故障模式分析到故障类型识别的FTA-SVM故障识别方法。首先利用故障树在复杂系统故障模式分析中的优势,找出系统的故障模式,建立故障树模型,通过对故障树模型中各故障事件的分析,采集与故障事件状态相关的数据,建立数据与故障树底事件的映射模型,最后利用支持向量机在小样本数据处理中的优势,进行故障类型的识别。以发动机的失火故障为例建立了发动机失火故障树模型及故障数据与故障模式映射模型,验证了FTA-SVM方法的有效性和适用性。

英文摘要：

Engine structure has become increasingly complex and its faults have the characteristics of diversity and frequency,so there exist difficulties in collecting a large quantity of fault datasets.In order to improve the fault recognition efficiency and accuracy of vehicles engine,in this paper,a new fault recognition method（FTA-SVM） has proposed combined the respective features of fault tree analysis（FTA） with support vector machine（SVM）,which can give us a method of fault mode analysis and fault type recognition.Firstly,FTA is utilized in finding the fault model and establishing the fault tree model of the engine system,taking the advantage of FTA in complex system fault mode analysis, Secondly,according to analyzing each fault event on fault tree model and collecting data related to fault event state,a mapping model has established between data and bottom event of fault tree, Lastly,SVM is used in fault type recognition,taking the advantage of SVM in small sample data processing; Then,taking the miss fire fault of engine as an example,a miss fire fault tree model of engine and a mapping model between fault data and fault mode are established to verify the validity and applicability of FTA-SVM method.