HUFLfev / AutoGluon ↗Electricity Transformer Temperature
Covers 2 electrical transformer series recorded at an hourly cadence across 34,840 timesteps, targeting a 168-hour (7-day) forecast horizon for critical oil temperature and six multi-point electrical load variables to prevent grid asset overheating.
Business Objective: Electricity Transformer Temperature
01Business Context
Electrical power transformers are mission-critical grid nodes subject to thermal stress and insulation degradation during high-demand cycles. Grid operators and asset engineers need multi-day forward visibility into internal oil temperature dynamics and winding load distributions to coordinate dispatch, cooling systems, and maintenance windows.
02Analytical Objective
Forecast transformer oil temperature (OT) and 6 electrical load metrics across high, middle, and low usage states (HUFL, HULL, MUFL, MULL, LUFL, LULL) across a 168-hour (7-day) rolling horizon.
03Economic & Decision Impact
Under-forecasting core oil temperatures risks catastrophic transformer thermal runaway, unplanned substation outages, and accelerated insulation breakdown costing millions in asset replacement. Over-forecasting leads to unnecessary load shedding, premature cooling activation, and uneconomic dispatch rerouting.
On the ETT-1H benchmark, classical statistical baselines like AutoARIMA and Seasonal Naive struggle to capture multi-channel cross-correlation between load states and thermal dissipation over 168-hour horizons. Lag-engineered gradient boosted trees (LightGBM) deliver robust short-range point estimates but degrade over long iterative rollouts. Zero-shot foundation models such as Mistral's Timestral, TimesFM, and MOIRAI excel at capturing long-sequence multi-frequency patterns directly in-context without station-specific hyperparameter retraining.
Target & Feature Column Definitions
Exhaustive business definitions, measurement units, roles, and target variables across all 9 columns.
| Col | Variable / Feature Name | Role | Type | Unit / Scale | Description & Business Meaning |
|---|---|---|---|---|---|
| C | HUFLHigh Use Full Load | TARGET | NUM | kW | High-voltage phase full-load power metric representing maximum continuous operating electrical throughput. |
| D | HULLHigh Use Low Load | TARGET | NUM | kW | High-voltage phase base-load power metric representing baseline minimum operational throughput. |
| E | LUFLLow Use Full Load | TARGET | NUM | kW | Low-voltage distribution circuit full-load electrical power metric. |
| F | LULLLow Use Low Load | TARGET | NUM | kW | Low-voltage distribution circuit baseline idle or minimum operational load level. |
| G | MUFLMiddle Use Full Load | TARGET | NUM | kW | Medium-voltage phase full-load power throughput under peak operational demand. |
| H | MULLMiddle Use Low Load | TARGET | NUM | kW | Medium-voltage phase baseline electrical load level during off-peak intervals. |
| I | OTTransformer Oil Temperature | TARGET | NUM | deg C | Core dielectric insulating oil temperature measured within the transformer tank; primary metric for thermal runaway monitoring. |
| A | idTransformer Identifier | SERIES ID | CAT | — | Unique categorical identifier for the monitored electrical power transformer station (e.g. ETTh1, ETTh2). |
| B | timestampMeasurement Timestamp | TIME | TIME | — | Hourly ISO 8601 timestamp recording the exact local observation interval. |
Interactive Data Table
Explore rows, feature values, and target labels for Electricity Transformer Temperature.
Foundation Models vs Traditional ML & Baselines (41 Methods)
Empirical evaluation directly from the AutoGluon fev benchmark suite evaluating zero-shot foundation models against deep learning, gradient-boosted trees, and statistical baselines across 100 evaluation tasks.
| Rank | Method / Model | Family | Organization | SQL Win Rate?Average win rate (%) against all competing models across the benchmark tasks under Scaled Quantile Loss. | SQL Skill?Normalized skill score relative to the Seasonal Naive baseline (0% = baseline, higher is better). | MASE Win Rate?Average win rate (%) under Mean Absolute Scaled Error. | Latency?Median end-to-end execution time per 100 time series in seconds. | Link |
|---|---|---|---|---|---|---|---|---|
| #1 | Timestral-209M | Foundation Model | Mistral AI | 87.7% | +50.4% | 86.0% | 0.51s | Repo ↗ |
| #2 | Timestral-123M (Carla Engine) | Foundation Model | Mistral AI | 86.8% | +50.2% | 85.0% | 0.32s | Repo ↗ |
| #3 | TimeRSI | Closed-Source API | GaTech & SamayLabs | 84.8% | +51.2% | 83.8% | 2512s | Repo ↗ |
| #4 | t0-gamma | Foundation Model | The Forecasting Company | 84.4% | +49.4% | 81.8% | 4.38s | Repo ↗ |
| #5 | Timestral-28M | Foundation Model | Mistral AI | 79.2% | +48.4% | 77.0% | 0.11s | Repo ↗ |
| #6 | timesfm-3 | Foundation Model | 78.5% | +49.0% | 77.0% | 3.41s | Repo ↗ | |
| #7 | ChakraTS | Baseline | Community | 78.2% | +48.1% | 79.5% | 14.22s | Repo ↗ |
| #8 | Lingjiang2.0 | Baseline | Community | 73.9% | +47.0% | 71.3% | 8.89s | Repo ↗ |
| #9 | Chronos-2 | Foundation Model | AWS | 70.4% | +47.3% | 68.4% | 0.84s | Repo ↗ |
| #10 | timeh2o-v1-tiny | Foundation Model | H2O.ai | 69.9% | +45.5% | 67.0% | 0.19s | Repo ↗ |
| #11 | Timestral-5M | Foundation Model | Mistral AI | 69.0% | +46.2% | 67.7% | 40ms | Repo ↗ |
| #12 | Toto-2.0-2.5B | Baseline | Community | 67.5% | +44.4% | 65.7% | 5.58s | Repo ↗ |
| #13 | t0-beta | Foundation Model | The Forecasting Company | 67.4% | +46.6% | 64.5% | 2.29s | Repo ↗ |
| #14 | Toto-2.0-1B | Baseline | Community | 66.4% | +44.4% | 65.5% | 2.48s | Repo ↗ |
| #15 | TiRex-2 | Foundation Model | NX-AI | 65.8% | +45.5% | 63.4% | 0.27s | Repo ↗ |
| #16 | Toto-2.0-313m | Baseline | Community | 65.1% | +44.2% | 63.5% | 1.06s | Repo ↗ |
| #17 | TiRex | Foundation Model | NX-AI | 58.0% | +42.6% | 54.1% | 0.24s | Repo ↗ |
| #18 | Toto-2.0-22m | Baseline | Community | 57.6% | +42.9% | 56.0% | 0.25s | Repo ↗ |
| #19 | TimesFM-2.5 | Baseline | Community | 54.6% | +42.2% | 53.5% | 1.87s | Repo ↗ |
| #20 | TabPFN-TS-3 | Foundation Model | Prior Labs | 52.4% | +43.1% | 51.3% | 235s | Repo ↗ |
| #21 | TS-ICL | Foundation Model | EDF-Lab | 51.2% | +43.1% | 49.8% | 2.49s | Repo ↗ |
| #22 | FlowState | Foundation Model | IBM | 49.3% | +39.9% | 49.9% | 2.29s | Repo ↗ |
| #23 | citras-fm | Foundation Model | Hitachi | 48.8% | +41.2% | 46.9% | 0.60s | Repo ↗ |
| #24 | Toto-1.0 | Baseline | Community | 47.5% | +40.8% | 45.9% | 22.35s | Repo ↗ |
| #25 | TabPFN-TS | Foundation Model | Prior Labs | 47.3% | +41.5% | 45.4% | 109s | Repo ↗ |
| #26 | Toto-2.0-4m | Baseline | Community | 46.2% | +40.7% | 43.6% | 0.20s | Repo ↗ |
| #27 | Moirai-2.0 | Baseline | Community | 41.9% | +39.3% | 40.5% | 0.35s | Repo ↗ |
| #28 | Chronos-Bolt | Foundation Model | AWS | 41.4% | +38.9% | 39.3% | 0.26s | Repo ↗ |
| #29 | TFT | Deep Learning | — | 31.7% | +32.2% | 29.2% | 829s | Repo ↗ |
| #30 | Stat. Ensemble | Baseline | Community | 27.5% | +20.7% | 30.6% | 147s | Repo ↗ |
| #31 | Sundial-Base | Foundation Model | Tsinghua University | 27.4% | +33.4% | 33.6% | 8.01s | Repo ↗ |
| #32 | PatchTST | Deep Learning | — | 26.8% | +30.1% | 25.1% | 664s | Repo ↗ |
| #33 | DeepAR | Deep Learning | — | 26.1% | +29.1% | 23.6% | 1113s | Repo ↗ |
| #34 | AutoARIMA | Statistical Baseline | — | 23.3% | +21.3% | 21.5% | 20.14s | Repo ↗ |
| #35 | AutoETS | Statistical Baseline | — | 20.3% | -26.8% | 22.5% | 3.48s | Repo ↗ |
| #36 | CatBoost | Tree Ensemble | — | 19.1% | +23.0% | 35.1% | 29.48s | Repo ↗ |
| #37 | LightGBM | Tree Ensemble | — | 17.8% | +21.0% | 33.1% | 3.03s | Repo ↗ |
| #38 | AutoTheta | Statistical Baseline | — | 15.6% | +5.5% | 19.8% | 3.28s | Repo ↗ |
| #39 | Seasonal Naive | Baseline | Community | 10.3% | 0.0% | 11.8% | 0.48s | Repo ↗ |
| #40 | Naive | Statistical Baseline | — | 6.7% | -45.4% | 10.9% | 0.47s | Repo ↗ |
| #41 | Drift | Statistical Baseline | — | 6.1% | -45.8% | 10.0% | 0.45s | Repo ↗ |
Variable Schema & Summary Distributions
Observed numerical ranges, category cardinalities, missing rates, and sample values across 34,840 rows.
| Col | Variable Name | Type | Missing | Distinct | Summary Stats / Distribution | Sample Values |
|---|---|---|---|---|---|---|
| C | HUFLTARGET | NUM | 0 | 1,295 | [-22.70599937438965, 107.89299774169922] μ=22.3 σ=17.3 | 5.8270001411437995.6929998397827155.1570000648498535 |
| D | HULLTARGET | NUM | 0 | 524 | [-18.68000030517578, 36.43899917602539] μ=5.4 σ=5.5 | 2.0090000629425052.07599997520446781.7410000562667847 |
| E | LUFLTARGET | NUM | 0 | 1,897 | [-14.350000381469727, 17.218000411987305] μ=-0.2 σ=5.5 | 4.2030000686645514.1420001983642583.7769999504089355 |
| F | LULLTARGET | NUM | 0 | 486 | [-31.461999893188477, 3.0460000038146973] μ=-0.6 σ=4.5 | 1.3400000333786011.3710000514984131.218000054359436 |
| G | MUFLTARGET | NUM | 0 | 2,550 | [-25.08799934387207, 93.2300033569336] μ=24.1 σ=22.3 | 1.59899997711181641.49199998378753661.2790000438690186 |
| H | MULLTARGET | NUM | 0 | 1,109 | [-6.593999862670898, 28.736000061035156] μ=4.6 σ=5.0 | 0.46200001239776610.42599999904632570.35499998927116394 |
| I | OTTARGET | NUM | 0 | 2,287 | [-4.079999923706055, 58.87699890136719] μ=20.0 σ=12.3 | 30.531000137329127.7870006561279327.78700065612793 |
| A | id | CAT | 0 | 2 | — | ETTh1ETTh1ETTh1 |
| B | timestamp | CAT | 0 | 17,420 | — | 2016-07-01T00:00:002016-07-01T01:00:002016-07-01T02:00:00 |
Frequently Asked Questions: Electricity Transformer Temperature
Common questions regarding the Electricity Transformer Temperature dataset, machine learning task formulations, and in-browser tabular inference.
What is the forecasting horizon for Electricity Transformer Temperature (ETT_1H)?
The standard forecasting horizon for ETT_1H is 168 hours (7 full calendar days) at an hourly cadence, providing grid operators with a week-ahead view of transformer thermal and electrical load dynamics.
Can I run zero-server forecasts on this dataset in Google Sheets?
Yes, Carla enables zero-server in-context time series forecasting directly in spreadsheets using Mistral's Timestral foundation model without configuring external Python or cloud environments.
Which models perform best on Electricity Transformer Temperature?
Zero-shot foundation models like Mistral's Timestral and TimesFM, alongside specialized neural architectures (PatchTST, Autoformer) and tuned LightGBM multi-lag regressors, deliver top predictive accuracy across long horizons on ETT.
Test TabICLv2 on Electricity Transformer Temperature Yourself
Open the pre-loaded Google Sheet and let Carla configure the target and task for local, zero-cloud tabular machine learning.
Launch Carla
Click once to open the spreadsheet and Carla side panel together.
Review the Setup
Carla selects the dataset target and task from this page automatically.
Evaluate & Predict
Run predictions and compute metrics with zero server uploads.
Provenance & Attribution
Zhou, H., Zhang, S., Peng, J., Zhang, S., Li, J., Xiong, H., & Zhang, W. (2021). Informer: Beyond Efficient Transformer for Long Sequence Time-Series Forecasting. Proceedings of the AAAI Conference on Artificial Intelligence, 35(12), 11106-11115.