This project focuses on predictive maintenance using machine learning techniques for bearing fault detection. The goal is to develop models that can accurately predict the occurrence of faults in the CWRU ball bearing dataset.
The dataset used in this project is the CWRU ball bearing dataset, which contains vibration signals from different types of bearing faults. The dataset is available in Dataset/
The following machine learning algorithms have been implemented and evaluated in this project:
- RandomForestClassification
- KNeighborsClassification
- ArtificialNeuralNetworks
- GradientBoostingClassification
- Naive Bayes
For each algorithm, nine models have been trained and tested for different fault types:
- B007
- B014
- B021
- IR007
- IR014
- IR021
- OR007
- OR014
- OR021
The performance of each algorithm and model combination has been evaluated using various metrics, such as accuracy, precision, recall, F1-score , confusion matricies and ROC cuves. The results can be found in the results/ directory.
Contributions to this project are welcome. If you find any issues or have suggestions for improvement, feel free to open an issue or submit a pull request.
- This project is based on the CWRU ball bearing dataset provided by Case Westren Reserve University.