主办:陕西省汽车工程学会
ISSN 1671-7988  CN 61-1394/TH
创刊:1976年

汽车实用技术 ›› 2023, Vol. 48 ›› Issue (20): 110-116.DOI: 10.16638/j.cnki.1671-7988.2023.020.022

• 测试试验 • 上一篇    

基于自回归移动模型汽车传动系统故障诊断

付小丹   

  1. 江苏信息职业技术学院
  • 出版日期:2023-10-30 发布日期:2023-10-30
  • 通讯作者: 付小丹
  • 作者简介:付小丹(1981-),女,硕士,讲师,研究方向为汽车专业教学,E-mail:2315656991@qq.com

Fault Diagnosis of Automobile Transmission System Based on Autoregressive Moving Model

FU Xiaodan   

  1. Jiangsu Polytechnic of Information Technology
  • Online:2023-10-30 Published:2023-10-30
  • Contact: FU Xiaodan

摘要: 针对复杂汽车传动系统故障集,提出基于自回归移动模型的故障诊断算法研究。以随 机差分理论为基础构建故障信号的时序模型,并确定影响序列值的各种参数,采集原始故障 数据,进行 A/D 转换和数据标准化处理,为保留离散型数据的原始特征并降低系统噪声干扰, 采用数据升维理念形成二维纹理图像,并利用局部二值特征算子提取二维图像的细节。实验 结果显示,提出诊断算法具有更好故障特征分类性能和样本检验一致性,平均诊断精度可以 达到 99.27%。

关键词: 自回归移动模型;变速箱;齿轮组;离散型;升维处理

Abstract: Aiming at the complex fault set of automobile transmission system, a fault diagnosis algorithm based on autoregressive moving model is proposed. Based on the random difference theory, the timing model of the fault signal is constructed, and various parameters affecting the sequence value are determined. The original fault data is collected, and A/D conversion and data standardization are carried out. In order to retain the original features of the discrete data and reduce the interference of system noise, two-dimensional texture images are formed by using the concept of data dimension enhancement. Local binary feature operators are used to extract the details of two- dimensional images. The experimental results show that the proposed algorithm has better fault feature classification performance and sample test consistency, with an average diagnosis accuracy of 99.27%.

Key words: Autoregressive moving model; Transmission case; Gear set; Discrete type; Dimension increasing processing