Public battery field data

Field data from 1 home or building storage system in Japan with lead-acid cells, released with Vink, Ankyu and Koyama, Scientific Data. The release covers 3.3 unit-years of operation and about 0.326 MWh of battery energy (stated).

Download

One command (as of October 2026), from figshare:

python -m fielddata.fetch tsukuba

Load a first slice

import fielddata
frame = fielddata.load('tsukuba', start='2016-06-01', end='2016-06-08')
frame.head()

Field data from 28 passenger electric vehicles in United States, released with Oh, LeBlanc and Peng, IEEE Trans. Intell. Transp. Syst.. The release covers 23.9 unit-years of operation and about 0.072 MWh of battery energy (stated).

Download

One command (as of October 2026), from GitHub:

python -m fielddata.fetch ved

Load a first slice

import fielddata
frame = fielddata.load('ved', unit=10, week='171101')
frame.head()

Field data from 4 home or building storage systems in Brazil, released with Gaura et al., Zenodo. The release covers 2.8 unit-years of operation and about 0.001 MWh of battery energy (computed).

No paper. Related: Trindade et al. 2022, Environ. Sci. Proc. 15, 16

Download

One command (as of October 2026), from Zenodo:

python -m fielddata.fetch brazil_star

Load a first slice

import fielddata
frame = fielddata.load('brazil_star', unit='STAR-A', system='Battery')
frame.head()

Field data from 1,027 home or building storage systems in sub-Saharan Africa with lead-acid cells, released with Aitio and Howey, Joule. The release covers 1,665.0 unit-years of operation and about 0.246 MWh of battery energy (computed).

Download

One command (as of October 2026), from Oxford ORA:

python -m fielddata.fetch aitio

2.9 GB, slow host (Oxford ORA), allow ~15 min. Each file prints when it is done.

Load a first slice

import fielddata
frame = fielddata.load('aitio', unit='0', clean=True)
frame.head()

Field data from 1 passenger electric vehicle in China, released with Wei et al., Journal of Energy Storage. The release covers 1.6 unit-years of operation and about 0.032 MWh of battery energy (computed).

Download

One command (as of October 2026), from Mendeley Data:

python -m fielddata.fetch beijing_ev

Load a first slice

import fielddata
frame = fielddata.load('beijing_ev', unit='2018-01-17')
frame.head()

Field data from 4 units in in orbit (ISS orbit) with NiMH cells, released with Jara et al., Data in Brief. The release covers at least 78 h of operation and about 0 MWh of battery energy (computed).

Download

One command (as of October 2026), from Mendeley Data:

python -m fielddata.fetch birds_cubesat

Load a first slice

import fielddata
frame = fielddata.load('birds_cubesat', unit='TSURU', sheet='March 28')
frame.head()

Field data from 3 second-life battery packs of one building storage system in Belgium with NMC-class cells, released with Gaukema et al., Zenodo. The release covers 1.0 unit-years of operation and about 0.08 MWh of battery energy (computed).

No paper. Related: Voets and Ben-Al-Lal, pv magazine, 21 Jan 2022

Download

One command (as of October 2026), from Zenodo:

python -m fielddata.fetch cloverleaf

Load a first slice

import fielddata
frame = fielddata.load('cloverleaf', unit='SNAM HV - 2100173')
frame.head()

Field data from 464 passenger electric vehicles in China with NMC-class cells, released with He et al., arXiv. The release covers at least 57,265 h of operation.

Download

One command (as of October 2026), from figshare:

python -m fielddata.fetch evbattery

Load a first slice

import fielddata
frame = fielddata.load('evbattery', unit='battery_dataset1:0')
frame.head()

Field data from 1 passenger electric vehicle in Germany, released with Wassiliadis et al., eTransportation. The release covers 0.0 unit-years of operation.

Download

One command (as of October 2026), from mediaTUM:

python -m fielddata.fetch tum_id3_2022

Load a first slice

import fielddata
frame = fielddata.load('tum_id3_2022', track=0)
frame.head()

Field data from 20 passenger electric vehicles in China with NMC cells, released with Deng et al., Applied Energy. The release covers 46.2 unit-years of operation and about 0.953 MWh of battery energy (computed).

Download

One command (as of October 2026), from GitHub:

python -m fielddata.fetch deng

Load a first slice

import fielddata
frame = fielddata.load('deng', pack='#1')
frame.head()

Field data from 10 automated guided vehicle packs in Italy with Li-ion cells, released with Zarfati and Bedogni, IEEE CCNC. The release covers 2.1 unit-years of operation.

Download

One command (as of October 2026), from Zenodo:

python -m fielddata.fetch flashbattery-agv

Load a first slice

import fielddata
frame = fielddata.load('flashbattery-agv', unit='FB-0')
frame.head()

Field data from 347 passenger electric vehicles in China with NMC-class cells, released with Zhang et al., Nature Communications. The release covers at least 45,030 h of operation.

Download

One command (as of October 2026), from figshare:

python -m fielddata.fetch zhang2023

Load a first slice

import fielddata
frame = fielddata.load('zhang2023', unit='battery_brand1:1')
frame.head()

Field data from 2 passenger electric vehicles in Germany with NMC, LFP cells, released with Bilfinger et al., eTransportation. The release covers 3.2 unit-years of operation and about 0.111 MWh of battery energy (stated).

Download

Manual, in a web browser (as of October 2026), from mediaTUM. The fetch command prints these steps:

bilfinger2024 requires a manual download.
1. Open https://mediatum.ub.tum.de/1737452
2. Download the data files listed below. Do not download papers.
   - data.zip (about 0.3 GB)
   Note: data.zip is the "Download all" zip of the mediaTUM data-server share linked from the record.
3. Save the files under data/Battery pack diagnostics for electric vehicles - Transfer of differential voltage and incremental capacity analysis from cell to vehicle level/data
4. Run python -m fielddata.verify bilfinger2024 to check the saved files.

Load a first slice

import fielddata
frame = fielddata.load('bilfinger2024', unit='Tesla_JB_8A_CEE7_C57_2021')
frame.head()

Field data from 4 electric buses (country not stated), released with Fei, Mendeley Data. The release has no timestamps, so the time it covers is not known.

Download

One command (as of October 2026), from Mendeley Data:

python -m fielddata.fetch fei_bus

Load a first slice

import fielddata
frame = fielddata.load('fei_bus', unit='vin1')
frame.head()

Field data from 10 grid-scale storage units in Germany with LMO/NMC, LFP, LTO, lead-acid cells, released with Evaluation operation report, RWTH Publications. The release covers 0.7 unit-years of operation and about 6.714 MWh of battery energy (stated).

Download

Manual, in a web browser (as of October 2026), from RWTH Publications. The fetch command prints these steps:

m5bat-2023-04 requires a manual download.
1. Open https://publications.rwth-aachen.de/record/985923
2. Download the data files listed below. Do not download papers.
   - M5BAT_04-2023_RAW.zip (about 0.1 GB)
   Note: RWTH Publications needs a web browser; its download links do not work from a script.
3. Save the files under data/M5BAT Large-Scale Battery Storage System Dataset - Evaluation Operation Report 04-2023/data
4. Run python -m fielddata.verify m5bat-2023-04 to check the saved files.

Load a first slice

import fielddata
frame = fielddata.load('m5bat-2023-04', unit='Batt2')
frame.head()

Field data from 21 home or building storage systems in Germany with LFP, LMO/NMC, NMC cells, released with Figgener et al., Nature Energy. The release covers 120.1 unit-years of operation and about 0.153 MWh of battery energy (stated).

Download

One command (as of October 2026), from Zenodo:

python -m fielddata.fetch rwth-home

Load a first slice

import fielddata
frame = fielddata.load('rwth-home', unit='01', month='2019-06')
frame.head()

Field data from 28 home or building storage systems (country not stated) with LFP cells, released with Schaeffer et al., Cell Reports Physical Science. The release covers 56.9 unit-years of operation and about 0.108 MWh of battery energy (computed).

Download

One command (as of October 2026), from Zenodo:

python -m fielddata.fetch schaeffer

Load a first slice

import fielddata
frame = next(fielddata.load('schaeffer', system=1, chunksize=500000))
frame.head()

Field data from 12 passenger electric vehicles in Germany with NMC, LFP cells, released with Bilfinger et al., npj Clean Energy. The release covers 6.9 unit-years of operation and about 0.373 MWh of battery energy (stated).

Download

Manual, in a web browser (as of October 2026), from mediaTUM. The fetch command prints these steps:

bilfinger2025 requires a manual download.
1. Open https://mediatum.ub.tum.de/1765567
2. Download the data files listed below. Do not download papers.
   - data.zip (size not stated)
   Note: data.zip is the "Download all" zip of the mediaTUM data-server share linked from the record (https://dataserv.ub.tum.de/index.php/s/m1765567, about 306 MB).
   Note: From a script the share's download needs its session cookie; without it the server returns a small stub, so check the size.
3. Save the files under data/Why we need a standardized state of health measurement procedure for electric vehicle battery packs - A proposal for energy- and capacity-based metrics/data
4. Run python -m fielddata.verify bilfinger2025 to check the saved files.

Load a first slice

import fielddata
frame = fielddata.load('bilfinger2025', unit='Tesla_Model_Y_JB_8A_CEE7_MI_2025')
frame.head()

Field data from 514 passenger electric vehicles in China with NMC-class, LFP cells, released with Cao et al., Nature Communications. The release covers at least 151,900 h of operation and about 15.107 MWh of battery energy (computed).

Download

One command (as of October 2026), from Zenodo:

python -m fielddata.fetch cao

Load a first slice

import fielddata
frame = fielddata.load('cao', brand='QAS', vehicle='0', names='inferred', clean=True)
frame.head()

In-depth notebook: a worked example of a release published without column names, units or timestamps.

Field data from 300 passenger electric vehicles in China with NMC cells, released with Liu et al., Nature Communications. The release covers 975.4 unit-years of operation and about 16.294 MWh of battery energy (computed).

Download

Manual, in a web browser (as of October 2026), from Changan portal. The fetch command prints these steps:

changan requires a manual download.
1. Open http://ivstskl.changan.com.cn/?p=2697
2. Download the data files listed below. Do not download papers.
   - raw/RAW_DATA.z01 (about 51.5 GB)
   - raw/RAW_DATA.zip (about 22.8 GB)
   Note: the two files are one split zip of about 74 GB (RAW_DATA.z01 51.5 GB, RAW_DATA.zip 22.8 GB); check you have the space first.
   Note: save both in a subfolder named raw inside the folder in step 3 (create it), and do not unzip them.
3. Save the files under data/Multi-modal framework for battery state of health evaluation using open-source electric vehicle data/data
4. Run python -m fielddata.verify changan to check the saved files.

Load a first slice

import fielddata
frame = fielddata.load('changan', unit='vin1', nrows=200000, clean=True)
frame.head()

Field data from 2 passenger electric vehicles in Belgium with LFP, NCA cells, released with Yasko et al., Scientific Data. The release covers 1.8 unit-years of operation and about 0.139 MWh of battery energy (stated).

Download

Manual, in a web browser (as of October 2026), from KU Leuven RDR. The fetch command prints these steps:

ku_leuven_bev requires a manual download.
1. Open https://doi.org/10.48804/8KPDTW
2. Download the data files listed below. Do not download papers.
   - doi-10.48804-8kpdtw.zip (about 0.2 GB)
   Note: the zip is Dataverse's "Access dataset > Download ZIP" on the KU Leuven RDR page.
3. Save the files under data/Unveiling Energy Dynamics of Battery Electric Vehicle Using High-Resolution Data/data
4. Run python -m fielddata.verify ku_leuven_bev to check the saved files.

Load a first slice

import fielddata
frame = fielddata.load('ku_leuven_bev', unit='BEV1', session_type='driving')
frame.head()

Field data from 88 grid-scale storage units in Sweden, released with Fast et al., Zenodo. The release covers 56.5 unit-years of operation and about 0.115 MWh of battery energy (stated).

No paper. Related: project README (RISE, Swedish Energy Agency)

Download

One command (as of October 2026), from Zenodo:

python -m fielddata.fetch rise_conbat

Load a first slice

import fielddata
frame = fielddata.load('rise_conbat', window='9. 240210-240215', level='module', unit='H400788-1')
frame.head()

Field data from 31,449 charging sessions in China with LFP, NMC, LMO, LCO, lithium polymer (cell format, chemistry not stated) cells, released with Yang et al., Nature Communications. The release covers at least 21,479 h of operation.

Download

One command (as of October 2026), from Mendeley Data:

python -m fielddata.fetch shenzhen_charging

Load a first slice

import fielddata
frame = fielddata.load('shenzhen_charging', unit='44020100400000012010151800001210', file='processed_data_longer_than_30')
frame.head()

Field data from 4 passenger electric vehicles in China with LFP, NMC cells, released with Tao et al., eTransportation. The release covers 9.9 unit-years of operation and about 0.189 MWh of battery energy (computed).

Download

One command (as of October 2026), from Zenodo:

python -m fielddata.fetch tongji_rest

Load a first slice

import fielddata
frame = fielddata.load('tongji_rest', unit='NMC02', kind='rest')
frame.head()

Field data from 7 passenger electric vehicles in Germany with NMC cells, released with Schreiber et al., eTransportation. The release covers 6.7 unit-years of operation and about 0.406 MWh of battery energy (stated).

Download

One command (as of October 2026), from GitHub:

python -m fielddata.fetch tumftm

Load a first slice

import fielddata
frame = fielddata.load('tumftm', unit='CUP1', value_id=1200, clean=True)
frame.head()

Field data from 7 passenger electric vehicles in Germany with NMC, LFP cells, released with Bilfinger et al., eTransportation. The release covers 0.3 unit-years of operation and about 0.401 MWh of battery energy (stated).

Download

Manual, in a web browser (as of October 2026), from mediaTUM. The fetch command prints these steps:

bilfinger2026 requires a manual download.
1. Open https://mediatum.ub.tum.de/1840112
2. Download the data files listed below. Do not download papers.
   - data.zip (about 0.3 GB)
   Note: data.zip is the "Download all" zip of the mediaTUM data-server share linked from the record.
3. Save the files under data/Battery pack diagnostics for electric vehicles - Robustness of the state of health measurement and differential voltage analysis at the vehicle level/data
4. Run python -m fielddata.verify bilfinger2026 to check the saved files.

Load a first slice

import fielddata
frame = fielddata.load('bilfinger2026', unit='Cupra_204_JB_8A_CEE7_C45')
frame.head()

Field data from 16 grid-scale storage units (country not stated) with LFP, NMC cells, released with He et al., Zenodo. The release covers at least 54,750 h of operation and about 6.5 MWh of battery energy (computed).

No paper. Related: manuscript unpublished, authors' code record, Zenodo 21105057 (2026-07)

Download

One command (as of October 2026), from Zenodo:

python -m fielddata.fetch cell_to_pack

Load a first slice

import fielddata
frame = fielddata.load('cell_to_pack', unit='Pack_field#1', columns=['time(s)', 'current(A)', 'Vpack', 'SOC'])
frame.head()

Field data from 1 grid-scale storage unit (country not stated) with LFP cells, released with Li et al., Cell Reports Physical Science. The release covers 1,203 h of operation and about 0.156 MWh of battery energy (computed).

Download

One command (as of October 2026), from Zenodo:

python -m fielddata.fetch li2026

Load a first slice

import fielddata
frame = fielddata.load('li2026', unit='real_world_02-03/battery_01')
frame.head()

Field data from 1 grid-scale storage unit in Germany with lead-acid cells, released with Zurmühlen, Koltermann and Sauer, Energies. The release covers 8.1 unit-years of operation and about 1.066 MWh of battery energy (stated).

Download

Manual, in a web browser (as of October 2026), from RWTH Publications. The fetch command prints these steps:

m5bat-pbacid requires a manual download.
1. Open https://publications.rwth-aachen.de/record/1038622
2. Download the data files listed below. Do not download papers.
   - Full_Dataset_M5BAT_Battery_Unit_Pb1.zip (4.33 GB (page display))
   Note: RWTH Publications needs a web browser; its download links do not work from a script.
3. Save the files under data/Operational Degradation of Flooded Lead-Acid Storage Under Frequency Containment Reserve/data
4. Run python -m fielddata.verify m5bat-pbacid to check the saved files.

Load a first slice

from fielddata.loaders import m5bat_pbacid
frame = m5bat_pbacid.preview(source='bms', year=2023, rows=200000)
frame.head()

Field data from 12 grid-scale storage units in United States with LFP, NMC, LMO/NMC, vanadium flow cells, released with Crawford and Thompson, Zenodo. The release covers 17.2 unit-years of operation.

No paper. Related: Crawford 2026, DOE OE peer-review poster

Download

One command (as of October 2026), from Zenodo:

python -m fielddata.fetch pnnl_epri

Load a first slice

import fielddata
frame = fielddata.load('pnnl_epri', unit='ENW', meter='PDC', columns=['Voltage', 'Current', 'SOC', 'Power'])
frame.head()

Field data from 1 grid-scale storage unit in United States with Li-ion cells, released with Kyeremeh et al., IEEE Access. The release covers 8.3 unit-years of operation and about 2 MWh of battery energy (stated).

Download

Partly manual (as of October 2026). One command downloads the files on GitHub:

python -m fielddata.fetch ppl

The command then prints the manual step for the rest:

ppl is only partly downloaded by this command. One more step is manual:
   The cell-level data (about 66.6 GB, one zip per month from December 2016) is not in the GitHub repository.
   The repository's file BESS-Analysis/Cell_Level_Data links to it on a Box share:
   https://app.box.com/s/s0ns2o1i4q0f7yxcnn96df7e5px5k0s6
   Download the monthly zips in a web browser and save them in a folder named cell_level next to BESS-Analysis.
   The ppl loader reads only the GitHub files, so this step is needed only to use the cell-level data yourself.

Load a first slice

import fielddata
frame = fielddata.load('ppl', years=[2019])
frame.head()

Field data from 33 mobile phones in Germany with Li-ion cells, released with Laufen et al., RWTH Publications. The release covers 3.5 unit-years of operation and about 0.001 MWh of battery energy (computed).

Download

Manual, in a web browser (as of October 2026), from RWTH Publications. The fetch command prints these steps:

rwth-android requires a manual download.
1. Open https://publications.rwth-aachen.de/record/1002905
2. Download the data files listed below. Do not download papers.
   - Dataset_from_Mobile_Battery_Data_Explorer_2026-04.zip (about 76.1 MB)
   Note: RWTH Publications needs a web browser; its download links do not work from a script.
3. Save the files under data/Battery Cell Field Data of Multiple Mobile Android Devices/data
4. Run python -m fielddata.verify rwth-android to check the saved files.

Load a first slice

import fielddata
frame = fielddata.load('rwth-android', unit='03b3c5a4a5d9a7a3d9829afa52b734c1')
frame.head()

Field data from 1 passenger electric vehicle in United States, released with Lanubile et al., Stanford Digital Repository. The release covers 3.5 unit-years of operation and about 0.091 MWh of battery energy (computed).

No paper. Related: companion paper on the same vehicle, arXiv 2603.29107 (2026)

Download

One command (as of October 2026), from Stanford Digital Repository:

python -m fielddata.fetch stanford_ev

Load a first slice

import fielddata
frame = fielddata.load('stanford_ev', session='20210921_234539', signal='V_pack_bms')
frame.head()

No preview: the release's license (CC BY-NC-ND 4.0) does not allow derived files.

Field data from 271 light electric vehicle packs (country not stated) with LFP cells, released with Xie et al., Cell Reports Physical Science. The release covers 4.8 unit-years of operation.

Download

One command (as of October 2026), from Zenodo:

python -m fielddata.fetch xie

Load a first slice

import fielddata
frame = fielddata.load('xie', unit='0022234cc5964eae8c383cc9032cd51f')
frame.head()

Field data from 3 units in China with NMC, LFP cells, released with Zhou et al., Nature Energy. The release covers 10.4 unit-years of operation and about 0.209 MWh of battery energy (stated).

Download

Manual, in a web browser (as of October 2026), from OneDrive. The fetch command prints these steps:

zhou2026 requires a manual download.
1. Open https://1drv.ms/u/c/909aeacce295a32a/IQCsGxZgHovYTqJEXhfqRIIZAWASy12k6Urytil7VPa0LJY?e=ISLuvO
2. Download the data files listed below. Do not download papers.
   - Vehicle data.zip (about 2.2 GB)
3. Save the files under data/Quantifying the impact of cell-to-cell inconsistency on electric vehicle battery degradation and utilization/data
4. Run python -m fielddata.verify zhou2026 to check the saved files.

Load a first slice

import fielddata
frame = fielddata.load('zhou2026', unit='LFP01', part=0, columns=['Timestamp', 'TotalVoltage', 'TotalCurrent', 'SOC', 'MaxCellVoltage', 'MinCellVoltage'])
frame.head()