BG-CNN: A Hybrid Fault Diagnosis Method for Improved Fault Isolation. This repository presents the BG-CNN method, a novel approach that combines the Bond-Graph technique with Convolutional Neural Networks (CNNs) for efficient fault isolation.
The data used in this project is from the ALL_DC_motor_Data folder, which contains 12 CSV files of DC motor signals under different fault conditions.
The code for this project is divided into five Jupyter notebooks:
- Part_1_Data_pre_prcessing.ipynb: This notebook contains the code for data pre-processing, such as loading, splitting, normalizing, and in depth visualization various fault data.
- Part_2_CNN_Training_ARR_github.ipynb: This notebook contains the code for training and testing a convolutional neural network (CNN) model using the ARR signals as input.
- Part_3_HyperParam_Tuning_github.ipynb: This notebook contains the code for hyperparameter tuning of the CNN model using Grid Serach.
- Part_4_NumberOfLabels.ipynb: This notebook contains the code for analyzing the effect of the number of labels on the performance of the CNN model.
- Part_5_Comparision.ipynb: This notebook contains the code for comparing the CNN model with other baseline models, such as support vector machine (SVM), random forest (RF), and multilayer perceptron (MLP).
The code is written in Python and uses libraries such as pandas, numpy, matplotlib, seaborn, sklearn, tensorflow, keras, and bayesian_optimization.
This project is licensed under the MIT License - see the LICENSE file for details.