BatteryLife_Processed

News

• [June 3rd]: Released v6, adding the complete Stanford dataset as "Stanford_2" (now including both releases of the Stanford dataset). We also corrected data statistics in the readme documents.
• [Apr. 29th 18:00 pm]: We found that some "SOC_interval" records in the ISU_ILCC dataset were wrong. We have fixed the wrong "SOC_interval" records in version v5.
• [Mar. 5th]: Due to some network issues, only the data since Mar. 5th, 2025 are complete. Please download the data of version v4 from Zenodo or download the datasets from Huggingface since Mar. 5th, 2025.

Note: Time is (Anywhere on Earth time zone).

Here is the processed dataset for our publication, "BatteryLife: A Comprehensive Dataset and Benchmark for Battery Life Prediction."

The purpose of proposing this data is to facilitate battery life prediction research. It is also possible to use the processed datasets for other battery informatics tasks. This dataset consists of four sub-datasets: Li-ion, Zn-ion, Na-ion, and CALB datasets. It includes 990 batteries, 8 different battery formats, 59 different chemical systems, 9 different operation temperatures, and 421 different charge/discharge protocols in a uniform data format. Other details can be found in README.md in each sub-dataset. The life labels of the processed batteries are in "Life Labels.zip".

The official repository for preprocessing and benchmarking is available at: Ruifeng-Tan/BatteryLife: The official BatteryLife repository (github.com). If you use the data, you should cite the BatteryLife: A Comprehensive Dataset and Benchmark for Battery Life Prediction and the original papers that produced the data. The updated code of conduct is shown in the Github repository.

If you are interested in contributing, please contact us via email at rtan474@connect.hkust-gz.edu.cn and whong719@connect.hkust-gz.edu.cn.