The data for this project comes from the NASA battery prognostics dataset ([link](https://ti.arc.nasa.gov/tech/dash/pcoe/battery-prognostics/)). The data consists of charge/discharge curves for batteries which have been cycled under different conditions as well as intermittent impedance measurements. The raw data from NASA is available as a set of `.mat` (matlab data storage type) files. Code which was used to turn these into Python dictionaries as well as examples can be found in the `nasa-battery-data-analysis.ipynb` notebook. ---------- #### Project ideas - Visualization - How do Re and Rct change over cycling under the different conditions? Are they correlated? - Use bokeh to create an interactive visualization for searching through the large amounts of data or enable the comparison between cells. - Tool building - Write a function to integrate the current during charge/discharge to make a plot of state-of-health/capacity vs cycle number - Download the remaining datasets and write script for reading in all of the data - Machine learning - Use the cycling/impedance data to predict the capacity of the cells. - Predict the end of life for the cells (published work from the original data set, [pdf](http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.154.8090&rep=rep1&type=pdf))