基于深度学习的柴油机气门健康状态评估
DOI:
作者:
作者单位:

作者简介:

通讯作者:

中图分类号:

TH212

基金项目:

军内科研项目(LJ20191C050518)


Evaluation of Diesel Engine Valve Health Status Based on Deep Learning
Author:
Affiliation:

Fund Project:

Military scientific research project (LJ20191C050518)

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    柴油发动机在运行过程中,其气门间隙会随其性能状态退化发生改变,为了解决传统的健康状态评估方法健康指标确定困难、权重人为经验依赖性大的问题,本文提出一种基于深度学习的柴油机气门健康状态评估方法。首先通过小波包分解算法对柴油机缸盖振动信号进行分解,对分解得到的节点信号分别提取常见的14个时域特征和小波包分解信号能量比向量,构建多维综合健康评估指标向量。然后基于一维卷积神经网络(One-Dimensional Convolutional Neural Network, 1DCNN)构建健康状态评估模型,将得到的健康评估指标向量输入模型中进行训练与健康状态评估。本文通过柴油机实验台开展气门退化模拟实验验证了该方法的有效性,与传统方法相比解决了健康指标的筛选问题以及人为主观经验带来的影响,并具有更好的健康状态评估效果。

    Abstract:

    During the operation of a diesel engine, its valve clearance will change with the degradation of its performance state. In order to solve the problems of the traditional health assessment method, the determination of health indicators is difficult and the weight is artificially dependent on human experience. This paper proposes a diesel engine based on deep learning. Evaluation method of valve health status. Firstly, the diesel engine cylinder head vibration signal is decomposed by wavelet packet decomposition algorithm, and 14 common time-domain features and wavelet packet decomposition signal energy ratio vector are extracted from the decomposed node signals, and a multi-dimensional comprehensive health assessment index vector is constructed. Then build a health evaluation model based on One-Dimensional Convolutional Neural Network (1DCNN), and input the obtained health evaluation index vector into the model for training and health evaluation. This paper verifies the effectiveness of the method through the valve degradation simulation experiment carried out on the diesel engine test bench. Compared with the traditional method, it solves the problem of screening health indicators and the influence of subjective experience of people, and has a better health evaluation effect.

    参考文献
    相似文献
    引证文献
引用本文

白雲杰,贾希胜,梁庆海,等. 基于深度学习的柴油机气门健康状态评估[J]. 科学技术与工程, 2022, 22(10): 3941-3950.
Bai Yunjie, Jia Xisheng, Liang Qinghai, et al. Evaluation of Diesel Engine Valve Health Status Based on Deep Learning[J]. Science Technology and Engineering,2022,22(10):3941-3950.

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2021-07-07
  • 最后修改日期:2022-01-26
  • 录用日期:2021-11-30
  • 在线发布日期: 2022-04-14
  • 出版日期:
×
律回春渐,新元肇启|《科学技术与工程》编辑部恭祝新岁!
亟待确认版面费归属稿件,敬请作者关注