摘要
针对民航发动机寿命预测研究中单参数监控不全面而多监测参数利用率低的问题,提出了一种基于多参数融合相似的寿命预测方法。针对参数敏感性修正,利用ReliefF-PCA算法对发动机多种监控参数进行属性筛选,并融合为表征发动机健康状态的参数——健康指数;针对发动机不同衰退阶段,对相似度度量算法进行趋势敏感度修正,增大变化趋势对预测结果的影响;通过对样本轨迹进行平移处理,减小正常阶段时域对算法的影响;最后通过实际数据的对比来验证该方法的有效性,结果表明,改进的预测方法有较好的预测精度。
Aiming at the problems that single parameter monitoring was not comprehensive and multi-monitoring parameter utilization rate was low in the research of civil aviation engine life prediction,a life prediction method was proposed based on multi-parameter fusion.In order to correct the parameter sensitivity,the ReliefF-PCA algorithm was used to screen the attributes of various monitoring parameters of the engine and to fuse them into the parameter-health index representing the engine health states.According to the different recession stages of the engine,the trend sensitivity of the similarity measurement algorithm was modified to increase the influence of the change trend on the prediction results.The effects of time domain on the algorithm were reduced by the translation of sample trajectory.Finally,the validity of the proposed method was verified by comparing the actual data.The results show that the improved method has better prediction accuracy.
作者
曹惠玲
崔科璐
梁佳旺
CAO Huiling;CUI Kelu;LIANG Jiawang(School of Aeronautical Engineering,Civil Aviation University of China, Tianjin,300300;Shanghai Aircraft Customer Service Co.,Ltd.,Shanghai,200241)
出处
《中国机械工程》
EI
CAS
CSCD
北大核心
2020年第7期781-787,共7页
China Mechanical Engineering
基金
中国民航大学博士启动基金资助项目(QD02s04)
PHM机载系统预诊断与告警模型研究项目(20190102010200)。
