| 液体火箭发动机基于神经网络的实时故障检测算法实现 |
| |
| 引用本文： | 黄强,吴建军,刘洪刚,谢廷峰. 液体火箭发动机基于神经网络的实时故障检测算法实现[J]. 国防科技大学学报, 2007, 29(5): 10-13 |
| |
| 作者姓名： | 黄强 吴建军 刘洪刚 谢廷峰 |
| |
| 作者单位： | 国防科技大学,航天与材料工程学院,湖南,长沙,410073;国防科技大学,航天与材料工程学院,湖南,长沙,410073;国防科技大学,航天与材料工程学院,湖南,长沙,410073;国防科技大学,航天与材料工程学院,湖南,长沙,410073 |
| |
| 基金项目： | 国家863高技术资助项目(2005AA7XX070)，教育部“新世纪优秀人才支持计划”项目(NCET-06-09278) |
| |
| 摘 要： | 以某大型液体火箭发动机为研究对象,针对其启动和稳态工作过程,利用Matlab和Lab Windows/CVI等编程工具,基于神经网络技术,开发实现了其地面试车过程实时故障检测的BP(Back Propagation)和RBF(Radial Basis Function)算法。多次试车数据离线检验和实时在线考核结果均表明该方法能够及时、有效地检测出发动机工作过程中的故障,没有出现误报警和漏报警,并能够很好地满足现场试车的实时性和鲁棒性等要求。
|
| 关 键 词： | 液体火箭发动机 故障检测 神经网络 BP网络 RBF网络 |
| 文章编号： | 1001-2486(2007)05-0010-04 |
| 收稿时间： | 2007-01-09 |
| 修稿时间： | 2007-01-09 |
Implementation of Real-time Fault Detection Algorithms Based on Neural Network for Liquid Propellant Rocket Engines |
| |
| HUANG Qiang,WU Jianjun,LIU Honggang and XIE Tingfeng. Implementation of Real-time Fault Detection Algorithms Based on Neural Network for Liquid Propellant Rocket Engines[J]. Journal of National University of Defense Technology, 2007, 29(5): 10-13 |
| |
| Authors: | HUANG Qiang WU Jianjun LIU Honggang XIE Tingfeng |
| |
| Affiliation: | College of Aerospace and Materials Engineering, National Univ. of Defense Technology, Changsha 410073, China;College of Aerospace and Materials Engineering, National Univ. of Defense Technology, Changsha 410073, China;College of Aerospace and Materials Engineering, National Univ. of Defense Technology, Changsha 410073, China;College of Aerospace and Materials Engineering, National Univ. of Defense Technology, Changsha 410073, China |
| |
| Abstract: | Based on the back propagation and radial basis function neural network,and using the tool of Matlab and Lab Windows/CVI,the real-time fault detection algorithms for the start-up and main-stage process of a certain liquid-propellant rocket engine in ground tests are developed in this paper.The algorithms realized were verified with a great deal of historical test-data and also validated in the practical ground tests of the engine.The results show that the algorithms not only can detect the fault of the engine in time and efficiently without false alarm and missing alarm,but also can meet the real-time ability and robustness requirement. |
| |
| Keywords: | liquid-propellant rocket engine fault detection neural network back propagation network radial basis function network |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
| 点击此处可从《国防科技大学学报》浏览原始摘要信息 |
|
点击此处可从《国防科技大学学报》下载免费的PDF全文 |
|
