中国民航大学学报 ›› 2021, Vol. 39 ›› Issue (5): 16-21.

• 民用航空 • 上一篇    下一篇

基于 NRS-CNN 的民航发动机滑油消耗量预测

瞿红春,高鹏宇,朱伟华,许旺山,郭龙飞    

  1. (中国民航大学航空工程学院,天津 300300)
  • 收稿日期:2020-06-16 修回日期:2020-06-16 接受日期:2020-03-02 出版日期:2021-10-20 发布日期:2021-10-31
  • 作者简介:瞿红春(1971—),男,湖北荆州人,副教授,博士,研究方向为航空发动机状态监控与故障诊断。
  • 基金资助:
    中国民航大学科研基金项目(05yk08m);中央高校基本科研业务费专项(ZXH2010D019)

Lubricating oil consumption prediction of civil aviation engine based on NRS-CNN

QU Hongchun, GAO Pengyu, ZHU Weihua, XU W angshan, GUO Longfei    

  1. (College of Aeronautical Engineering, CAUC, Tianjin 300300, China)
  • Received:2020-06-16 Revised:2020-06-16 Accepted:2020-03-02 Online:2021-10-20 Published:2021-10-31

摘要:

针对民航发动机滑油消耗量受多个飞行阶段的多个参数影响而难以准确预测的问题,提出基于邻域粗糙(NRSneighborhood rough set)和卷积神经网络(CNNconvolutional neural network)的模型来预测滑油消耗量。首先采用 NRS 方法提取对滑油消耗重要度较高的飞行阶段作为特征参数;然后,利用 CNN 对重要度高的飞行阶段参数进行深度特征学习,实现滑油消耗量的预测。 预测结果表明:CNN 能很好地完成对多滑油参数的特征提取,预测结果与实际值的平均绝对误差为 0.129 × 10-3 m3 ,平均相对误差为 3.8%,可满足实际工程应用的需要,为评估民航发动机滑油系统的健康状况提供参考。

关键词: 滑油消耗量, 多参数预测, 邻域粗糙集, 卷积神经网络

Abstract: with regard to the multiple parameters in multiple flight stages and the difficulty of accurately predicting lubricating oil consumption of civil aviation engines, a model based on neighborhood rough set (NRS) and convolutional neural network (CNN) is proposed. First, the NRS method is used to extract the flight phases that are more important to the oil consumption as feature parameters; second, the CNN is used to conduct an in-depth feature study with reference to flight phase parameters so as to make oil consumption prediction. The results show that the CNN can well complete the feature extraction of multiple oil parameters. The average absolute error between the prediction result and the actual value is 0.129 伊10 -3 m3 , and the average relative error is 3.8% , which can meet the needs of practical engineering applications and provide reference for evaluating the health status of civil aviation engine lubricating oil system.

Key words: Oil consumption, multi-parameter prediction, neighborhood rough set (NRS), convolutional neural network(CNN)

中图分类号: