首页|基于胶囊神经网络的涡扇发动机故障诊断方法

基于胶囊神经网络的涡扇发动机故障诊断方法

扫码查看
针对航空涡扇发动机数据集故障分类准确率较低的问题,提出一种基于胶囊神经网络的涡扇发动机故障诊断方法.首先确定故障类型和关键变量,然后构建卷积胶囊神经网络模型,将分割的训练集数据输入模型进行训练,最后利用诊断模型诊断测试集数据并计算分类识别准确率.将所提算法在NASA涡扇发动机数据集上进行测试,证明了该模型的分类识别准确率有所提高,可为涡扇发动机的故障诊断提供帮助.
Fault Diagnosis Method of Turbofan Engine Based on Capsule Neural Network
In order to solve the problem of low fault classification accuracy of aviation turbofan engine dataset,a fault diagnosis method of turbofan engine based on capsule neural network is proposed.First,the fault type and key variables are determined,then the convolutional capsule neural network model is constructed,and the segmented training set data is input into the model for training.Finally,the diagnostic model is used to diagnose the test set data and calculate the classification recognition accuracy.The proposed algorithm is tested on the NASA turbofan engine dataset,which proves that the classification and recognition accuracy of the model is improved,and can help the development of fault diagnosis of turbofan engine.

turbofan engineconvolutional neural networkcapsule networkfault diagnosis

胡鑫、谭曾盛

展开 >

湖南工业大学计算机学院,湖南株洲 412007

涡扇发动机 卷积神经网络 胶囊网络 故障诊断

2024

信息与电脑
北京电子控股有限责任公司

信息与电脑

影响因子:1.143
ISSN:1003-9767
年,卷(期):2024.36(3)
  • 10