turbofan

Unified PHM framework for Remaining Useful Life (RUL) prediction, fault diagnosis, fault detection, and anomaly detection for bearings, turbofan engines, and other industrial systems.

Gas Turbine / Engine Deck in Python for use in flight test.

End-to-end predictive maintenance pipeline on NASA C-MAPSS with LSTM, GRU, Transformer, benchmarking, and automated reporting.

The project aims to study the parametrical parameters of a two-spool turbofan with convergent nozzles

⚡ Aerospace Fleet Telemetry & Predictive Maintenance Platform. Java 21 Loom (>48k msgs/s), Dual ML (LightGBM + ONNX 0.014ms), and Deterministic CMMS Triage.

Deep learning-based Remaining Useful Life (RUL) prediction for turbofan engines using multivariate sensor time-series data and LSTM networks.

Small environment for debugging JS code.

基于 NASA C-MAPSS 涡轮发动机退化仿真数据，构建剩余有用寿命（Remaining Useful Life, RUL）预测系统

ML-based remaining useful life estimation and maintenance alerting for aircraft turbofan engines (C-MAPSS, TE7483)

Predictive Maintenance MLOps — RUL Prediction for Turbofan Engines with MLflow, DagsHub, SHAP, Evidently, Streamlit, and Prometheus

This project builds a machine learning framework for predictive maintenance of turbofan engines, estimating Remaining Useful Life (RUL) from the NASA C-MAPSS sensor dataset. Methods included anomaly detection with CUSUM and autoencoders, and LSTM models, achieving significant RMSE reduction over baselines.

GE F404 turbofan cycle deck — design-point sizing and altitude/Mach/dTs off-design sweeps, built on OpenMDAO's pyCycle.

A turbofan engine and its four-engine airliner, modelled entirely in code with Three.js: cutaways, a working wing and landing gear, a cabin, and one landing. An interactive lesson on how a jet engine works.

Interactive 3D WebGL aero engine visualizer featuring NACA airfoil blades, dual-spool (N1/N2) kinematics, station cutaways, flow particles, and real-time telemetry. Built with Three.js.

Engine test-bed logs in, requirement verdicts out: limits, overshoot, rise/settling time, and sensor-fault vs control-problem diagnosis on synthetic turbofan data.

End-to-end predictive maintenance pipeline on the NASA CMAPSS turbofan dataset (Python, scikit-learn, XGBoost, Plotly Dash).

Production-grade turbofan predictive maintenance: piecewise RUL regression, sensor anomaly isolation, capacity-constrained worklist ranking, and TreeSHAP on NASA C-MAPSS FD001.

Python-based turbofan cycle, performance, off-design and emissions modelling with ICAO reference data comparisons and statistical analysis.

Remaining-life baseline for turbofan engines with an engine-level train/test split (synthetic data, C-MAPSS format)

Parametric trade-study comparing high OPR vs. high T4 design approaches for a two-spool turbofan engine using pyCycle and OpenMDAO.