RUL Estimation for EV Batteries

A flow chart of the XGBoost Regression Model

This was a course project as part of the course DS203: Programming for Data Science taught by Prof. Amit Sethi. We chose the problem statement of predicting the remaining useful life of Li-ion EV batteries using indirect discharge cycle parameters. We started by performing exploratory data analysis on the charging, discharging and impedance cycles for Li-ion batteries using NASA’s PCoE datasets. We moved on to understanding, implementing and testing various regression models such as SVR, Multilayer Perceptron, LSTM, and various Boosting algorithms. We were able to achieve an R2 score of 98.09 for estimating the battery capacity using the XGBoost regression model.

Report GitHub

References

2016

XGBoost: A Scalable Tree Boosting System

Tianqi Chen, and Carlos Guestrin

In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Aug 2016

Abs

Tree boosting is a highly effective and widely used machine learning method. In this paper, we describe a scalable end-to-end tree boosting system called XGBoost, which is used widely by data scientists to achieve state-of-the-art results on many machine learning challenges. We propose a novel sparsity-aware algorithm for sparse data and weighted quantile sketch for approximate tree learning. More importantly, we provide insights on cache access patterns, data compression and sharding to build a scalable tree boosting system. By combining these insights, XGBoost scales beyond billions of examples using far fewer resources than existing systems.