[TIM] A High-Precision Aeroengine Bearing Fault Diagnosis Based on Spatial Enhancement Convolution and Vision Transformer🥳

This is the official PyTorch codes for the paper:

Bin Wang, Yongcheng Xiong, Liguo Tan*. A high-precision aeroengine bearing fault diagnosis based on spatial enhancement convolution and vision transformer[J]. IEEE Transactions on Instrumentation and Measurement, 2025, 74: 1-15.

Network Architecture 💐

News 🚀

Dec 18, 2024: We release training code.

Getting started

Install

We test the code on Pytorch 2.2.1 + CUDA 11.8

Create a new conda environment, or take advantage of an existing conda environment.

[!NOTE]
Make sure python version 3.9 is available.

1 2	conda create -n CSSTNet python=3.9 conda activate CSSTNet

Install dependencies

1 2	conda install pytorch==2.2.1 torchvision==0.17.1 torchaudio==2.2.1 pytorch-cuda=11.8 -c pytorch -c nvidia pip install -r requirements.txt

Training and Evaluation

Create New folder

|-CSST_Net
  |-data
  |-Loss_Acc
  |-predata
  |-results

Prepare dataset

You can download the datasets on Google Drive。

Then arrange the data in the following format：

|-data
  |-data1.npy
  |-data2.npy
  |-data3.npy
  |-data4.npy
  |-data5.npy

Once you have downloaded the data set and settled it according to the above requirements, run the following statement to make the data you need for training and testing.

1	python dataset.py

[!CAUTION]
The current code can only be used to make noisy data, if you make noise-free data, please manually change the code.

After production, the predata folder will contain the following files:

|-predata
  |-data.npy
  |-label.npy
  |-data_0.npy
  |-label_0.npy
  |-data_2.npy
  |-label_2.npy
  |-data_6.npy
  |-label_6.npy
  |-data_10.npy
  |-label_10.npy
  |-data_-2.npy
  |-label_-2.npy

Train

Once the data set is ready, training can be performed

1	python train.py

Test

Please match the pre-trained weights with the dataset for testing, then:

1	python test.py

Citation 💓

If you find our work useful for your research, please cite us:

@ARTICLE{10758748,
  author={Wang, Bin and Xiong, Yongcheng and Tan, Liguo},
  journal={IEEE Transactions on Instrumentation and Measurement}, 
  title={A High-Precision Aeroengine Bearing Fault Diagnosis Based on Spatial Enhancement Convolution and Vision Transformer}, 
  year={2025},
  volume={74},
  number={},
  pages={1-15},
  keywords={Feature extraction;Fault diagnosis;Aircraft propulsion;Accuracy;Noise;Convolution;Time-frequency analysis;Transforms;Transformers;Computer vision;Aeroengine intershaft bearing;convolutional neural network (CNN);multisensor information fusion;spatial enhancement;vision transformer (VIT)},
  doi={10.1109/TIM.2024.3502884}}