Live · Battery Data Intelligence Platform

research-ready.

BatteryLake unifies the world's fragmented battery cycling datasets into one standardized, quality-assessed foundation — so models are compared on science, not on preprocessing luck.

Explore Datasets View on GitHub

batterylake.org / datasets ?quality=ready

LCO 2007_NASA_PCoE_LCO_18650_1C_1C_25TNASA Prognostics CoE 34 cells 0.96

LFP 2019_Stanford_MIT_TRI_LFP_18650_MultiC_30TStanford · MIT · TRI 124 cells 0.94

NMC 2024_Imperial_Kirkaldy_NMC_21700_MultiC_MultiTImperial College London 21 cells 0.92

LFP 2026_NTU_Ampace-Samsung_LFP-NMC_21700_2C_2C_25TNTU EEE · Singapore 16 cells 0.95

SOH 94.2%cycle 187 · B0005

Quality gate passed26 validation rules

0

Curated datasets

0

Worldwide institutions

0

Years span

2007-2026

0

Total cycles

0

Total cells

0 GB

Data volume

How it works

From raw lab exports to comparable results

Every dataset travels the same auditable path, so downstream numbers mean the same thing everywhere.

Curate & Ingest 41+ public and lab datasets, tracked with DOI-level provenance.

Standardize One ETL pipeline, one schema: metadata, time-series, cycle summaries.

Quality Gate 26 physical-plausibility and schema rules score every dataset.

Benchmark & Share Reproducible SOH / RUL runs with pinned splits and unified metrics.

Platform

One workspace for the full research loop

Browse, benchmark, and audit — each surface is built on the same standardized data core.

Benchmarks

Reproducible experiments, not preprocessing folklore

Random, temporal, and cross-cell split protocols
7 reference models — Ridge to Transformer and PINN
RMSE · MAE · MAPE reported on identical folds

Open the benchmark workbench

Split protocol — cross-cellseed 42

Train

62%

Val

18%

Test

20%

RMSE0.0121

MAE0.0094

MAPE1.08%

Quality

An auditable gate before any model sees the data

Physical plausibility: voltage windows, energy balance
Signal-level QC maps for V(t), I(t), T(t), Qd(n)
JSON reports wired into the benchmark pipeline

See the quality methodology

Quality gate — NASA PCoEready

0.96overall

Voltage range 2.0–4.5 V

Coulombic efficiency 95–105%

Temperature drift — 1 review

APIs

The same platform, programmable

RESTful endpoints with token auth
Parquet, CSV, and JSON exports
Embedded DOI citations in every response

Explore the developer console

GET api.batterylake.org/v1/datasets

import requests

resp = requests.get(
  "https://api.batterylake.org/v1/datasets",
  params={"chemistry": "LFP", "quality": "ready"}
)

for ds in resp.json()["items"]:
  print(ds["ref_name"], ds["quality_score"])

Visitors

Global Reach

See where BatteryLake visitors are exploring from around the world.

Start with research-ready battery data today.

Browse the catalog, download standardized packages, or contribute your lab's datasets to the community.

Browse Catalog Contribute Data

Design by

In partner with

0 datasets

Choose Evaluation Task

SOH Estimation

Step 1: Select Dataset

Step 2: Select Input Signals

Configure Data Partitioning

Train

—

Validation

—

Test

—

Excluded not used in train / validation / test 0 cells

Drag cells between Train / Validation / Test or into Excluded to leave them out

Select Benchmark Models

Execute Local Training Package

Step 1: Download Training Package

Get the configured dataset, model architecture, and runtime environment.

bt_benchmark_soh_lstm.zip

Step 2: Run in Your Terminal

Unzip the package, open your terminal, and execute the following command to start local training.

unzip bt_benchmark_soh_lstm.zip
cd bt_benchmark_soh_lstm
mkdir -p data
# copy the whole processed dataset folder into data/ first
# example: cp -R /path/to/2019_Stanford_MIT_TRI_LFP_18650_MultiC_30T data/
python3 -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
bash run_benchmark.sh

Step 3: Upload Output Folder

When training finishes, upload the generated output folder containing your results.

Drag & drop the output folder here, or click to browse Expected: metrics.json, predictions.csv, and optional per_cell_metrics.csv

Waiting for local output folder

Benchmark Evaluation Results

Performance Highlights

Best ModelLSTM-Battery-v2

Test RMSE0.0124

Ground Truth Predicted Value Retirement Line 80%

Step 1 of 6 - Benchmark Task

— models

0

Completed

0

In Progress

0

Pending

0

Skipped

ID	Dataset	Status	Meta	Series	Summary	QC

Format pattern · Connected

BatteryLake reference name sequencer

Inspecting Field 1 of 7

Assembled reference name


          2007_NASA_PCoE_LCO_18650_1C_1C_25T

Interaction: automatic field walkthrough; hover or focus to inspect one field.7 fields · underscore-delimited

Examples

Dataset	ref_name
NASA PCoE	2007_NASA_PCoE_LCO_18650_1C_1C_25T
Stanford-MIT-TRI	2019_Stanford_MIT_TRI_LFP_18650_MultiC_30T
NTU EEE Internal	2026_NTU_Ampace-Samsung_LFP-NMC_21700_2C_2C_25T
Imperial 21700	2024_Imperial_Kirkaldy_NMC_21700_MultiC_MultiT

Field Vocabulary

Sources (25+)

NASA_PCoE · CALCE_UMD · Stanford_MIT_TRI · Oxford_Howey · RWTH_Aachen · NTUEEE · SNL · HNEI · UL_Purdue · XJTU · KIT · Tsinghua · CMU_Bills · Stanford_Onori · TUM · Beihang · Imperial · ISU_ILCC · EVERLASTING_4TU · Mendeley · Figshare · Zenodo ...

Chemistries

LFP · NMC · NMC811 · LCO · NCA · LiIon (generic) · MultiChem

Form Factors

18650 · 21700 · Pouch · Prismatic · Cyl · Auto · EV-BMS

Getting Started

Clone the repository and install the conda environment.

git clone https://github.com/tianwen1209/BatteryLake-Benchmark-DataPrep.git
cd BatteryLake-Benchmark-DataPrep
conda create -n batterylake python=3.10
conda activate batterylake
pip install -r requirements.txt

Dataset Registry

All datasets are tracked in dataset_registry.csv with columns for dataset identity, DOI, source URL, assigned owner, ref_name, processing status, QC status, last update, and notes.

Evaluation

The benchmark evaluation framework uses evaluate.py and dataset_interface.py to run baseline models across selected datasets and split protocols.

BatteryLake Benchmark

Open benchmark infrastructure for battery PHM research.

BatteryLake Benchmark curates, standardizes, and evaluates lithium-ion battery cycling datasets from research labs worldwide, enabling fair comparison of machine learning models for state-of-health estimation and remaining useful life prediction.

View repository Nanyang Technological University, Singapore

Curated datasets41

Prediction tasksSOH / RUL

Benchmark scopeETL + Models

Team

Research and engineering contributors

BatteryLake combines battery-domain supervision, benchmark research, and frontend engineering into one reproducible platform.

Project Supervisors

YW

Prof. Yonggang Wen

Principal Investigator

WH

Dr Wang Hao

Research Fellow

Core Researcher

ZT

Zhu Tianwen

PhD Candidate

Web Engineering

LK

Liu Kefan

Frontend Engineer

CH

Cao Han

Frontend Engineer

CY

Cai Yezi

Frontend Engineer

Affiliation

Nanyang Technological University

Singapore

Repository

BatteryLake-Benchmark-DataPrep

github.com/tianwen1209/BatteryLake-Benchmark-DataPrep

Licence

BatteryLake provides access to battery datasets, metadata, benchmarking resources, and related research materials collected from multiple sources.

Individual datasets, algorithms, and other resources may come with their own licences and citation requests. Please honour these requirements. BatteryLake will display the applicable licence and citation information whenever it is known.

Before downloading, redistributing, modifying, or using a resource, please review the licence and usage terms provided on its dataset or resource page. When source-specific terms are available, those terms take precedence over the general information provided on this page.

BatteryLake does not grant additional rights to third-party content beyond those provided by the original authors, institutions, or data owners.

Citation

If you use BatteryLake, its curated datasets, standardized outputs, benchmarking resources, or platform tools in academic work, please cite the BatteryLake paper below.

BatteryLake: Agentic, Physics-Grounded Curation of Heterogeneous Battery Aging Data and Benchmarking

Tianwen Zhu, Hao Wang, and Yonggang Wen, 2026.

BibTeX

@misc{zhu2026batterylakeagenticphysicsgroundedcuration,
      title         = {BatteryLake: Agentic, Physics-Grounded Curation of Heterogeneous Battery Aging Data and Benchmarking},
      author        = {Tianwen Zhu and Hao Wang and Yonggang Wen},
      year          = {2026},
      eprint        = {2607.09762},
      archivePrefix = {arXiv},
      primaryClass  = {cs.AI},
      url           = {https://arxiv.org/abs/2607.09762}
}

When using an individual dataset, algorithm, model, or other contributed resource, please also cite its original authors and follow any citation instructions shown on the corresponding resource page.

Citing both BatteryLake and the original resource helps ensure that the platform contributors, dataset creators, and research institutions receive appropriate credit.

Upload a dataset to assess

No file selected.

Example Result dataset_03_quality_assesment.json · Assessed 14:32 Ready 1 warning

0.97

Completeness

Missing channels2.1%

0.95

Consistency

Sequence issues3 flags

0.92

Accuracy

Physical checks5 / 6 pass

1.00

Validity

Schema errors0

0.96

Overall Score

Ready 1 warning

Physical Plausibility Checks

High-signal checks that determine whether a dataset can safely enter preprocessing and benchmark training.

5 passed - 1 warning

Voltage Range Validation
All cell voltages within 2.0V-4.5V nominal operating range for the stated chemistry.
Energy Balance Check
Charge/discharge energy integral consistency; coulombic efficiency remains within 95-105% per cycle.
Capacity Monotonicity
Degradation trajectory follows expected non-increasing trend with allowable recovery windows.
Temperature Consistency
Cell surface temperature must remain within 5°C of stated test condition; one dataset needs review.
Timestamp Integrity
Monotonically increasing timestamps with no negative intervals or unreasonable gaps above 24h.
Current Direction Consistency
Charge and discharge current signs follow one convention throughout the dataset.

Diagnostic Output

quality_report.json

{
  "dataset_id": "dataset_03",
  "quality_score": {
    "completeness": 0.97,
    "consistency": 0.95,
    "accuracy": 0.92,
    "validity": 1.00
  },
  "overall": 0.96,
  "gate": "ready_with_warning",
  "checks": [
    { "name": "voltage_range", "passed": true },
    { "name": "temperature_consistency", "status": "review" },
    { "name": "capacity_mono", "passed": true }
  ],
  "generated_at": "2026-04-28T12:00:00Z"
}

1Dataset context

2AI results

3Confirm mapping

4Download skill

5Verification

Task 1: Upload raw measurement data

Input: Raw measurement filesOutput: *_timeseries.csv + *_cycle_summary.csv

Choose or drop almost any raw file or folder

No files queued.

or provide a path accessible to your AI agent

Task 2: Metadata extraction

Input: Source URLs, GitHub repos, DOI pages and papersOutput: dataset_metadata.csv + source evidence

Start with the dataset source URL. The agent checks what the source page can support, reports missing BatteryLake fields, then uses an optional paper URL to fill the gaps before you manually confirm the final metadata.

Step A - dataset source URL

Use the official dataset page first. Missing fields stay marked as "Source page not stated".

Step B - paper URL for missing fields

Only fields that are still missing or lower-confidence will be replaced by paper-supported values.

No source URL has been checked yet.

dataset_metadata.csv draft

Review every field and confirm before downloading.

Confirm	Field	Value	Evidence	Source	Confidence

I reviewed the values, evidence and missing fields. Generate dataset_metadata.csv from the confirmed rows only.

AI inspection results - review and confirm

Confirm only what is correct. Anything unconfirmed is written as Unknown and excluded from the generated ref_name.

✓Inspection complete24 files · 6 sampled · read-only

0 / 9fields inferred

9need confirmation

-overall confidence

No files classified yetUpload raw data for Task 1 or provide a source URL for Task 2.

Review each AI-detected field

The AI agent can suggest values, but the skill package only trusts fields you explicitly confirm.

0 / 9 fields confirmed

Generated ref_name from confirmed fields only

UNKNOWN_UNKNOWN_UNKNOWN_UNKNOWN

Confirm fields above to replace Unknown slots. Unconfirmed fields stay Unknown in the downloaded skill.

Field	AI pre-fill	Evidence	Confidence
Dataset type	Calendar aging	Folders named `storage_25C`; records contain periodic RPT blocks separated by multi-day rests.	96%
Data format	XLSX + CSV	18 `.xlsx` workbooks and 6 exported `.csv` files detected.	100%
Year	2023	Workbook creation metadata and source README both state 2023.	94%
Source / lab	Zenodo / Example Lab	DOI and author affiliation found in `README.md`.	91%
Chemistry	NMC811 / graphite	Cell specification sheet names NMC811 cathode and graphite anode.	93%
Form factor	21700	`INR21700` appears in filenames and cell specification.	98%
C-rate	Needs confirmation	RPT current is measurable, but nominal capacity is inconsistent across two source documents.	42%
Temperature	25 °C, 45 °C	Folder names, chamber setpoint column and median sample temperatures agree.	97%
Cell count	24	24 unique IDs after normalizing workbook names and `cell_id` values.	99%
Cell ID rule	Source filename	Stable pattern `NMC_25C_C01`; no conflicting in-file identifier.	89%
License	Needs confirmation	DOI resolves, but no explicit license text was found in the downloaded package.	18%

Scientific mapping plan

The required outputs adapt to the experiment type. Fields that do not scientifically exist remain empty — they are never fabricated.

voltage_V

Mapped · Voltage [V]

current_A

Mapped · Current [mA] → A

cycle_index

Optional · RPT only

capacity_Ah

Derived from RPT blocks

SOH / RUL

Not required for schema

temperature_C

Mapped · Chamber temp

dT_dt_C_per_s

Preserve as extension

storage_time_h

Preserve as extension

Important distinction: this dataset can be BatteryTwin schema-compliant while still being unsuitable for an SOH/RUL benchmark. Benchmark eligibility is assessed separately and reports the missing scientific prerequisites.

Download the dataset-specific BatteryTwin skill.

The package is generated from the inspected evidence and your confirmed corrections, not from an unverified form.

Skill package

bt_skill_pending_inspection

Dataset ref_name

Pending confirmed context

Included files

SKILL.md + schema + context + validator

The skill instructs the local agent to re-check source evidence, preserve extended physical channels, archive existing outputs and record all fallbacks before writing processed data.

Validation, fallbacks and output audit

The generated skill verifies that its supporting files actually work and records every fallback or destructive-risk decision.

BatteryTwin schemaReadable

schema.json parsed successfully. Required fields will be selected according to dataset type.

Dataset contextDraft needed

No context file found. The skill will generate a draft from DOI, README and workbook metadata for user confirmation.

Bundled validatorFallback ready

Executable check failed in the sampled environment. Built-in validation will run and the failure reason will be logged.

Output directory12 old files

Existing outputs will be archived to a timestamped directory before new files are written.

1.8%missing values

37anomalies to flag

3unit conversions

4unmapped fields retained in audit

Raw Cycles

→

Cleaning

→

Resampling

→

Harmonization

→

Cycle Analytics

→

Quality Check

→

Export

Cleaning & Standardization

Remove noise, outliers, and corrupted records. Standardize column names, units (V, A, Ah, °C), and sign conventions across heterogeneous source formats.

Outlier detection Unit conversion Column mapping NaN handling

Resampling & Interpolation

Uniformly resample time-series to configurable target intervals. Apply linear or spline interpolation to align measurements from different instruments and sampling rates.

Temporal alignment Target-point resampling Interpolation Anti-aliasing

Metadata Harmonization

Unify metadata schemas across datasets: cell chemistry, capacity, form factor, test protocol, lab of origin, DOI, and environmental conditions into a common taxonomy.

Schema alignment Taxonomy mapping Protocol tagging Provenance tracking

Cycle Analytics & Feature Extraction

Compute per-cycle summary statistics: discharge capacity, charge capacity, coulombic efficiency, internal resistance, dQ/dV peak positions, and health indicators.

Capacity extraction CE calculation IR estimation dQ/dV analysis

Preprocessing Capabilities Reference

Capability	Description	Status
Noise filtering	Savitzky–Golay, median, and rolling-window filters for voltage/current smoothing	Active
Outlier detection	IQR-based and domain-aware anomaly flagging (e.g. voltage spikes, current transients)	Active
Cycle segmentation	Automatic charge/discharge/rest detection from current polarity and voltage patterns	Active
Capacity integration	Trapezoidal and Simpson's rule integration for Ah capacity per half-cycle	Active
Multi-format ingestion	Support for CSV, Parquet, HDF5, MATLAB .mat, Excel, MACCOR, Arbin, Neware	Active
dQ/dV computation	Incremental capacity analysis with configurable voltage binning and smoothing	In dev
EIS preprocessing	Impedance spectra parsing, Nyquist/Bode extraction, ECM fitting	Planned

Developer console

Explore endpoints as a working platform surface, not a brochure.

Choose a route from the library, inspect its parameters, preview the request, then run a simulated call to see the exact response shape the platform returns.

Live routes0

Formats0

Auth modeToken

Dataset catalog endpoint

Designed for search, filtering, and benchmark preparation. The response returns only datasets that pass the required ETL and quality gates.

Available

GET/v1/datasets?chemistry=LCO&quality=ready

chemistryLCO | LFP | NMC

formatCSV | Parquet

split_readytrue

dataset_id

cell_count

cycle_count

quality_score

response · application/json200 OK

Request preview

import requests

resp = requests.get(
  "https://api.batterylake.org/v1/datasets",
  headers={"Authorization": "Bearer BT_TOKEN"},
  params={"chemistry": "LCO", "quality": "ready"}
)

for dataset in resp.json()["items"]:
  print(dataset["name"], dataset["quality_score"])

Application workbench

Applications are grouped by what the user is trying to do, not by abstract product names.

3 surfaces

BenchmarkExperiment Builder

Pick datasets, split cells, select models, and run comparable SOH/RUL benchmarks.

FeaturesFormula Studio

Upload custom signal formulas and compute cycle-level model inputs from raw curves.

ExplainInterpretation Lab

Inspect feature importance, attention maps, and degradation signals after training.

Open API

Data Download API

Access harmonized, quality-assessed datasets via RESTful endpoints with embedded citations, metadata, and provenance tracking.

RESTful endpoints for dataset listing and download
Parquet, CSV, and JSON export formats
Embedded DOI citations and provenance
Selective cell/cycle range queries

Open API

Benchmark API

Run standardized ML benchmarks across curated datasets with reproducible train/val/test splits and unified evaluation metrics.

SOH estimation and RUL prediction tasks
5+ baseline models (LR, RF, XGBoost, LSTM, Transformer)
Random, temporal, and cross-cell split protocols
RMSE, MAE, MAPE unified metric suite

Coming Soon

Feature Engineering

Build custom feature extraction pipelines with interactive notebooks. Extract dQ/dV features, health indicators, and domain-specific signals for advanced analytics.

dQ/dV peak detection and tracking
Health indicator extraction (IC, DV)
Statistical feature pipeline
EIS parameter fitting

Coming Soon

Model Interpretation

Understand model predictions with explainability tools. Visualize attention maps, feature importance, and identify key battery aging patterns driving predictions.

SHAP feature importance analysis
Attention map visualization
Gradient-based attribution
Integrated gradients for deep models

Platform Roadmap

Data Download API

Open API — harmonized dataset access with provenance and citations.

Available

Benchmark API

Open API — reproducible SOH / RUL experiment execution.

In Progress

Feature Engineering

End-to-end application — dQ/dV, health indicators, formula features.

Planned

Model Interpretation

End-to-end application — SHAP, attention maps, attribution.

Planned

Submission cockpit

Package a dataset for review with the same discipline as a benchmark run.

The contribution page now shows what must be prepared, how the validation lane works, and which files are required before a pull request or upload can be accepted.

Required files4

Review gates5

AttributionDOI

Readiness checklist

A submission should not move forward until these minimum conditions are visible.

Time-series channelsVoltage, current, timestamp, and cell identifier are present. Required

Experiment metadataChemistry, nominal capacity, protocol, lab, and temperature are documented. Required

! Publication sourceDOI, license, or repository URL is attached for attribution. Preferred

Protocol notesRest periods, C-rates, formation cycles, and test cutoffs are explained. Required

Dataset intake package

The user sees the exact bundle expected before opening GitHub or uploading files.

01raw_cycles/Native cycler exports or normalized CSV/Parquet time-series.

02metadata.jsonCell metadata, protocol, institution, DOI, and license fields.

03protocol.mdCharge/discharge details, rest periods, test equipment, and notes.

04manifest.yamlFile map, checksums, expected channels, and reviewer comments.

submission/NASA_PCoE_2010/manifest.yaml Open repository

Validation lane

Each contribution passes through the same checks used by the platform.

1Schema parseColumn mapping, unit detection, and required metadata validation.

2Quality reportPhysical plausibility, missingness, and consistency scores are generated.

3Catalog publishDataset is added with provenance, citation, and benchmark readiness.

Share Your Battery Data with the Community

Every contributed dataset is processed through our standardized pipeline, quality-assessed, and made available to researchers worldwide - with full attribution and DOI citation to your original work.

Contribution Workflow

1

Prepare Data

Organize your raw data files, metadata, and protocol documentation according to our schema.

2

Submit via PR

Open a GitHub Pull Request to the BatteryLake repository with your dataset and metadata.

3

Validation

Our team runs the ETL pipeline and quality assessment to verify data integrity.

4

Publication

Your dataset is added to the catalog with full provenance, DOI, and your attribution.

What to Submit

Raw Data Files

Time-series measurements in CSV, Parquet, HDF5, MATLAB, or native cycler format (Arbin, MACCOR, Neware).

Metadata

Cell chemistry, nominal capacity, form factor, temperature, C-rate protocol, number of cells, and cell identifiers.

Protocol Description

Test protocol document detailing charge/discharge procedures, rest periods, characterization tests, and equipment used.

License / DOI / Source

Publication DOI, data license (CC-BY preferred), original source URL, and any attribution requirements.

Validation Template & Schema Checklist

Use our validation template to ensure your submission meets all requirements before opening a PR.

Check Item	Requirement	Required?
Time-series columns	timestamp, voltage_V, current_A, capacity_Ah (minimum)	Required
Cell identifier	Unique cell_id for each battery unit	Required
Chemistry & form factor	Stated in metadata.json or README	Required
Nominal capacity	Manufacturer-rated capacity in Ah	Required
Temperature channel	Cell/ambient temperature per measurement	Preferred
Publication DOI	Link to associated paper or data repository	Preferred
EIS data	Impedance spectroscopy if available	Optional

Open GitHub Repository to Contribute