首页|基于门控神经网络的新能源汽车故障智能诊断

基于门控神经网络的新能源汽车故障智能诊断

扫码查看
变速箱是汽车发动的主要部件,齿轮作为变速箱中的核心零部件,因其容易受到车况工况的影响而发生故障.因此,针对这一问题,提出了 一种基于门控神经网络的混合神经网络训练算法,用于变速箱齿轮的故障诊断.实验结果表明,所提出的模型能够有效降低梯度消失,减少时间序列变化等问题.相比传统神经网络,所提出的模型在强噪音干扰下的准确率始终保持96%以上.当正常样本数量为40时,所提出的模型的准确率相比于其他传统神经网络模型的准确率更高,达到了 94.86%.综上所述,此次研究所采用的齿轮故障诊断模型具有较好的性能表现,能够用于实际新能源汽车齿轮故障诊断问题中.
Intelligent Fault Diagnosis of New Energy Vehicles Based on Gated Neural Networks
The gearbox is the main component of a car's engine,and as the core component of the gearbox,gears are prone to malfunctions due to the influence of vehicle conditions.Therefore,in response to this issue,this experiment proposes a hybrid neural network training algorithm based on gated neural networks for fault diagnosis of gearbox gears.The experimental results show that the proposed model can effectively reduce gradient vanishing and time series changes.Compared to traditional neural networks,the pro-posed model consistently maintains an accuracy of over 96%under strong noise interference.When the normal sample size is 40,the accuracy of the proposed model is higher than other traditional neural network models,reaching 94.86%.In summary,the gear fault diagnosis model used in this study has good performance and can be used in practical new energy vehicle gear fault diagnosis prob-lems.

gearboxgeargating neural networknoiseaccuracy

陆健

展开 >

杨凌职业技术学院,陕西杨凌 712100

变速箱 齿轮 门控神经网络 噪音 准确率

杨凌职业技术学院自然科学研究基金(2019)

A2019038

2024

自动化与仪器仪表
重庆工业自动化仪表研究所,重庆市自动化与仪器仪表学会

自动化与仪器仪表

CSTPCD
影响因子:0.327
ISSN:1001-9227
年,卷(期):2024.(2)
  • 13