Portfolio

Predictive Maintenance: Turbofan Engine RUL Prediction (NASA CMAPSS)

Developed an end-to-end LSTM deep learning model to predict Remaining Useful Life (RUL) for turbofan engines using the NASA CMAPSS FD001 dataset.

Key Results:

• Test MAE: 36.26 cycles

• Test RMSE: 41.16 cycles

• Evaluated on 93 valid test engines (standard protocol)

This project demonstrates time-series forecasting, model stability techniques (gradient clipping, L2 regularization), and real-world ROI for predictive maintenance in manufacturing and defense sectors.

View Full Code on GitHub

Page updated

Report abuse