collection of predictive maintenance solutions for NASAs turbofan (CMAPSS) dataset
This repo contains the notebooks accompanying a small series of blog posts [1] on the NASA turbofan degradation dataset [2]. The turbofan dataset consists of 4 separate challenges of increasing difficulty. The engines operate normally in the beginning but develop a fault over time. For each challenge, the engines in the train set are run to failure. The timeseries in the test set end 'sometime' before failure. The goal is to predict the Remaining Useful Life (RUL) of each turbofan engine in the test set. See the table below for a short overview of the challenges.
Dataset | Operating conditions | Fault modes | Train size (nr. of engines) | Test size (nr. of engines) |
---|---|---|---|---|
FD001 | 1 | 1 | 100 | 100 |
FD002 | 6 | 1 | 260 | 259 |
FD003 | 1 | 2 | 100 | 100 |
FD004 | 6 | 2 | 248 | 249 |
The notebooks are used to explore the dataset and try various modeling techniques (both Machine Learning and Neural Networks). For the full explanation of the techniques and choices made during model development I recommend reading the blog posts [1].
cd exploring-nasas-turbofan-data-set
mkdir data
.zip
file to exploring-nasas-turbofan-data-set/data/
exploring-nasas-turbofan-data-set/data/CMAPSSData/
exploring-nasas-turbofan-data-set
and start your local notebook server by typing jupyter notebook
[1] blog post series
[2] A. Saxena, K. Goebel, D. Simon, and N. Eklund, “Damage Propagation Modeling for Aircraft Engine Run-to-Failure Simulation”, in the Proceedings of the Ist International Conference on Prognostics and Health Management (PHM08), Denver CO, Oct 2008.
[3] data repository: https://ti.arc.nasa.gov/tech/dash/groups/pcoe/prognostic-data-repository/#turbofan