The Aircraft Engine Remaining Useful Life (RUL) Prediction System is an AI-powered predictive maintenance platform designed to estimate the remaining operational life of aircraft turbofan engines using multivariate sensor data.
The system combines Transformer and GRU deep learning architectures with Digital Twin Technology to provide:
- ✅ Accurate RUL predictions
- ✅ Real-time engine health monitoring
- ✅ Maintenance planning support
The project utilizes the NASA C-MAPSS FD001 dataset and provides an interactive Streamlit dashboard for engineers and administrators to monitor engine health, analyze degradation trends, and manage maintenance activities.
- Secure login system
- Role-based access control
- Engineer and Admin dashboards
- Transformer-GRU hybrid architecture
- Predicts Remaining Useful Life (RUL)
- Generates prediction confidence scores
Monitors critical engine parameters:
- T2 – Total temperature at fan inlet
- T24 – Total temperature at LPC outlet
- T30 – Total temperature at HPC outlet
- T50 – Total temperature at LPT outlet
- P2 – Pressure at fan inlet
- P15 – Pressure in bypass duct
- P30 – Pressure at HPC outlet
- Nf – Physical fan speed
- Multi-Head Self Attention
- Positional Encoding
- Layer Normalization
- Captures temporal dependencies
- Learns engine degradation patterns
- Predicts Remaining Useful Life
This project uses the NASA C-MAPSS FD001 dataset.
Files Used
train_FD001.txttest_FD001.txtRUL_FD001.txt
The dataset contains:
- Engine operational cycles
- Sensor measurements
- Degradation patterns
- Remaining Useful Life targets
| Metric | Score |
|---|---|
| Validation R² | 0.91 |
| Last-Cycle Test R² | 0.89 |
git clone https://github.com/Bhoomi002/Aircraft_Engine_Predictive_Maintenance.git
cd Aircraft_Engine_Predictive_Maintenance
pip install -r requirements.txt- Copy
config_example.yamland rename it toconfig.yaml. - Update usernames, passwords, and the secret key.
- Run the application:
streamlit run app.py
Watch a short demonstration of the Aircraft Engine RUL Prediction System.
Bhoomika M
MCA Student, JSS Academy of Technical Education, Bengaluru
📧 Email: mbhoomika00@gmail.com
💼 LinkedIn: https://www.linkedin.com/in/bhoomika-m-80834a327/
This project is licensed under the MIT License.