Seoul National University Markov-Chain User-Profile Cycling Data: 130 VTC6 and Samsung 25R Cells for RUL Prediction (bsebench copy)

Provider
Yong Jin Jeong et al. (Seoul National University)
License
CC-BY-4.0 (source)
Size
5 files
Format
JSON (in sampled files)
Stored at
Hugging Face (https://huggingface.co/datasets/bsebench-org/snu-markov-rul-2025-raw)
Access
Free and public. No account with the data provider is needed.
DOI
10.17632/npjy7vdgky.1
Category
Energy
Source
https://doi.org/10.17632/npjy7vdgky.1

About this dataset

Cycling data from Seoul National University for A unified deep learning framework for high-performance remaining useful life (RUL) prediction, published on Mendeley Data. Three datasets cover 130 cells: datasets 1 and 2 use Sony/Murata 18650 VTC6 cells (NCA/graphite-silicon, 3 Ah) with 90 and 12 cells, and dataset 3 uses 28 Samsung SDI INR18650-25R cells (NCA+NCM/graphite-silicon, 2.5 Ah). Each cell receives its own operating profile of up to 1000 cycles generated from Markov matrices representing user groups, alternating 3C fast charging and 0.5C slow charging (CC-CV) with a 1C constant-current discharge; dataset 2 restricts charging to the mid-to-high SOC range (3.2 to 4.2 V). Cells were tested on a Wonik PNE CRC 05-10 cycler at 25 °C, with a 0.2C diagnostic cycle every 30 cycles and voltage, current and capacity logged every 10 s. The BSEBench copy holds the original archive, Battery raw data.zip (about 2.9 GB). Licensed CC BY 4.0; cite Jeong, Mendeley Data, doi:10.17632/npjy7vdgky.1.

How to load this dataset in Python

from huggingface_hub import hf_hub_download
path = hf_hub_download(repo_id="bsebench-org/snu-markov-rul-2025-raw", filename="source/Battery raw data.zip", repo_type="dataset")
print(path)

Data structure

File types in sampled files: JSON

Files

Example file: source/Battery raw data.zip

Analyze this dataset without downloading it

On Scigantic, Seoul National University Markov-Chain User-Profile Cycling Data: 130 VTC6 and Samsung 25R Cells for RUL Prediction (bsebench copy) opens in a hosted JupyterLab notebook with the data already mounted, so you can read files where they are stored instead of copying them first. Each notebook includes an AI assistant that can see the mounted files and write and run analysis code. Open the dataset page to launch a notebook, or browse the full catalog.

This page is also available as Markdown and as JSON.

Related Energy datasets