摘要
针对基于大样本飞参数据开展航空发动机性能监控时缺乏有效技术手段的问题,提出一种面向大样本飞参数据的航空发动机性能监控方法。首先建立了一种发动机稳定巡航态参数提取模型,用以从飞参数据中提取出监控需要的状态参数;然后提出一种基于支持向量合并的在线支持向量数据描述(SVDD)算法,用以提升发动机性能监控的效率;最后基于稳定巡航态参数提取模型和在线SVDD算法建立了发动机性能监控模型,并使用多台发动机历史飞参数据进行了性能监控实验。结果表明,该方法能够快速准确地判断出发动机的性能状况(异常识别率在80%左右),可以为地勤维护提供辅助决策。
It is lack of effective techniques for monitoring aero-engine performance based on large scale flight data. A method based on large scale flight data is proposed. Firstly, a stable cruise parameters extraction model is formulated to extract the state parameters required for performance monitoring from the flight data. Then, an online support vector data description(SVDD) algorithm based on support vectors merging is proposed to improve the efficiency of engine performance monitoring. Finally, an aero-engine performance monitoring model is achieved by fusing the stable cruise parameters extraction model and the online SVDD algorithm. Performance monitoring experiments are implemented by using the historical flight data of several aero-engines. Experimental results show that this method can quickly and accurately determine the performance condition of the aero-engine. The abnormality recognition rate is about 80%. It can provide the auxiliary decision-making for ground crew.
作者
王小飞
王元鑫
曲建岭
袁涛
Wang Xiaofei;Wang Yuanxin;Qu Jianling;Yuan Tao(Qingdao Branch of Naval Aeronautical University,Qingdao 266041,China)
出处
《仪器仪表学报》
EI
CAS
CSCD
北大核心
2020年第7期175-184,共10页
Chinese Journal of Scientific Instrument
基金
国家自然科学基金(51505491)项目资助
关键词
航空发动机
性能监控
大样本飞参数据
稳定巡航态
支持向量合并
aero-engine
performance monitoring
large scale flight data
stable cruising state
support vector merging
