CASIA OpenIR  > 毕业生  > 硕士学位论文
基于智能分析的航空发动机故障诊断专家系统研究
姚琦
学位类型工程硕士
导师王健
2016-05
学位授予单位中国科学院研究生院
学位授予地点北京
关键词航空发动机 故障诊断 故障树分析 专家系统
摘要对飞机故障进行快速的诊断并给出维修建议,对于提高飞行安全和降低运营成本具有重要意义。传统航空发动机故障诊断主要采用事后人工判读方式,每飞行1个小时,需要耗费半个小时进行数据判读,故障诊断效率低。为此,本文以某型发动机地面故障诊断项目为背景,通过对比分析基于模型、基于数据驱动和基于知识的故障诊断方法,选取工程实用性好、可扩展性强的专家系统为重点,研究建立基于智能分析的航空发动机故障诊断专家系统,以实现准确、快速的自动化故障诊断,提高发动机故障诊断效率。
本文的主要工作和成果如下:
(1)分析研究了国内外发动机故障诊断及专家系统的研究现状,明确了论文的研究内容。
(2)对比分析了发动机故障诊断中常用的基于模型、基于数据驱动和基于知识的故障诊断方法,指出了每种方法的优缺点和适用范围,从工程实用性的角度,选择专家系统解决发动机故障诊断问题。
(3)针对专家系统中的知识获取和故障自动推理问题,研究了基于故障树和规则的智能分析方法,通过对故障树的定性分析求解出最小割集,然后利用最小割集生成一系列简化的故障推理规则,避免了直接利用故障树建立复杂逻辑规则而给推理诊断带来不便的问题,提高了故障推理诊断速度。
(4)研究设计了基于智能分析的航空发动机故障诊断专家系统软件框架,并基于Windows平台、VC++编程环境和MySQL数据库,开发实现了基于智能分析的航空发动机故障诊断专家系统软件。
论文研究成果对提高航空发动机故障诊断效率,实现准确、快速的自动化诊断具有重要意义和实用价值,开发完成的基于智能分析的航空发动机故障诊断专家系统软件已在中航工业集团某公司得到成功应用,有效实现了对真实发动机数据的快速判读和故障诊断,诊断准确度和运行速度满足预定要求。
其他摘要It is significant to make a rapid diagnosis and offer maintenance suggestions when the aircraft fault happened, which is important to improve the flight safety and reduce the operation cost. The traditional diagnosis method of aero engine mainly uses the off-line manual diagnosis and the diagnosis efficiency is low. According to the statistics, it needs to spend half an hour for data interpretation and comparison for one hour running data. This paper takes a certain type of engine ground fault diagnosis project as the background. Comparison and analyses of the fault diagnosis methods based on the module, the fault diagnosis methods based on data driven and the fault diagnosis methods based on knowledge was given, the expert system was selected due to its good practicability and expansibility. This paper mainly focused on developing a fault diagnosis expert system of aero engine based on intelligent analysis to realize accurate, rapid diagnosis.
The main content of the paper are as follows:
(1) The current research status of engine diagnosis and expert system at home and
abroad were analyzed in order to determine the studying direction of this paper.
(2) Comparison and analyses of the fault diagnosis methods based on the module, the fault diagnosis methods based on data driven and the fault diagnosis methods based on knowledge was given, the advantages and disadvantages of each method and its application range were analyzed, eventually the expert system was selected to resolve the issues for the point of view of engineering practicability.
(3) For solving the problem of knowledge acquisition and fault automatic reasoning in expert system,a intelligent analysis method based on fault tree analasis and rule reasoning is proposed. In  particular, use the fault tree qualitative analysis to solve the minimal cut sets, and then use the minimum cut sets to generate a series of simplified fault reasoning rules. It can improve the speed of fault diagnosis through avoiding complex logical reasoning based on fault tree.
(4) A software framework of fault diagnosis expert system of aero engine based on intelligent analysis was designed. Then, a practical aero engine fault diagnosis expert system was developed based on the windows platform with VC++ programming environment and MySQL.
The research results are of important practical value to improve the efficiency of aero engine fault diagnosis and achieve accurate, fast automated diagnosis. The aero engine fault diagnosis expert system based on intelligent analysis has been successfully implied in a company of Aviation Industry Corporation of China, it has been proved that the system is effective to make fast fault diagnosis on the real engine data, as the diagnosis accuracy and speed meet the predefined requirements.
文献类型学位论文
条目标识符http://ir.ia.ac.cn/handle/173211/11810
专题毕业生_硕士学位论文
推荐引用方式
GB/T 7714
姚琦. 基于智能分析的航空发动机故障诊断专家系统研究[D]. 北京. 中国科学院研究生院,2016.
条目包含的文件
文件名称/大小 文献类型 版本类型 开放类型 使用许可
论文终稿.pdf(3730KB)学位论文 暂不开放CC BY-NC-SA请求全文
个性服务
推荐该条目
保存到收藏夹
查看访问统计
导出为Endnote文件
谷歌学术
谷歌学术中相似的文章
[姚琦]的文章
百度学术
百度学术中相似的文章
[姚琦]的文章
必应学术
必应学术中相似的文章
[姚琦]的文章
相关权益政策
暂无数据
收藏/分享
所有评论 (0)
暂无评论
 

除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。