Skip to main content
CARLA HQ
DATASET PROFILE / TIME SERIES

Solar Photovoltaic Power Output with Weather

Hourly single-asset solar photovoltaic generation benchmark spanning 49,648 observations paired with meteorological covariates for 24-hour day-ahead renewable power forecasting.

Energy & Utilities Operations & Workforce Analytics use casesNACE D35Time Series ForecastingHourly Cadence24 Steps HorizonMarkdown ↗Target: targetfev / AutoGluon ↗
✨ Try in CarlaInstall Chrome Extension ↗100% In-Browser WebGPU • Zero Cloud Upload
49,648Records (Rows)
1Time Series
HourlySampling Cadence
24 HoursForecast Horizon (20 Windows)
24h Diurnal
Dominant Seasonality?Strong diurnal 24-hour solar cycle with seasonal solar zenith angle shifts and astronomical daylight variation across the calendar year.
41 Methodsfev Leaderboard (SQL)
BUSINESS CONTEXT & OBJECTIVERenewable Power & Grid Integration (D35)

Business Objective: Solar Photovoltaic Power Output with Weather

01Business Context

Independent power producers (IPPs), transmission system operators (TSOs), and utility trading desks must submit firm day-ahead generation schedules to wholesale electricity markets (e.g. EPEX SPOT, CAISO, PJM) every 24 hours.

02Analytical Objective

Forecast hourly solar power generation (target) across a rolling 24-hour horizon (H=24) using historical solar telemetry and known numerical weather predictions (NWP).

03Economic & Decision Impact

Over-forecasting incurs severe balancing settlement penalties, real-time dispatch curtailment, and market imbalance fees; under-forecasting forces grid dispatchers to commit costly thermal peaker plants or forgo peak power prices during sunny intervals.

04Temporal Dynamics & Covariate Information Flow

Dominant Seasonality: Strong diurnal 24-hour solar cycle with seasonal solar zenith angle shifts and astronomical daylight variation across the calendar year.

Covariates & Information Flow: Known future covariates include numerical weather predictions (temperature, humidity, atmospheric pressure, rain, snow, day length, wind speed); past dynamic telemetry includes observed cloud cover and global horizontal irradiance (GHI).

ML Benchmark Narrative

Classical univariate statistical methods (AutoARIMA, Seasonal Naive) capture nighttime zero-clipping but fail during intermittent cloud dynamics. Gradient boosted trees (LightGBM with tabular lag features) excel when ingesting NWP weather covariates, while zero-shot pretrained foundation models (e.g. Mistral's Timestral, TimesFM, MOIRAI) demonstrate strong in-context adaptation to irradiance curves without requiring per-asset gradient fine-tuning.

Source Origin:National Renewable Energy Laboratory (NREL)
Creator:National Renewable Energy Laboratory (2024)
License:Open Data / Public Domain
IN-BROWSER FORECAST DEMO

Forecast this dataset in your browser

A representative series is read directly from this dataset’s Parquet file. Move the forecast start, compare held-out observations, and toggle available numeric covariates with Timestral 5M (12 MB) or switch to ci-proto (~346 MB, WebGPU). Benchmark scores above use the stated evaluation setup.

Live sample from dataset Parquet
Timestral 5M · WebGPU
Nord Pool Day-Ahead Electricity PricesHourly · €/MWh
Forecast History
Loading sample data…
Covariates
FEATURE SPECIFICATION & DATA DICTIONARY

Target & Feature Column Definitions

Exhaustive business definitions, measurement units, roles, and target variables across all 12 columns.

INFORMATION AVAILABILITY & TEMPORAL PROTOCOL

Known Future Covariates vs. Historically Observed Covariates

Multi-step forecasting models must respect the arrow of time and real-world information availability. FEV benchmarks and Carla strictly distinguish between features known across the forward prediction window versus realizations available only up to the forecast origin.

FUTURE KNOWN COVARIATESLegitimately available at test time (steps t+1 to t+H)

Scheduled promotions (such as Rossmann's Promo), calendar events (school/state holidays), and operational forecasts (such as Nord Pool's day-ahead grid load and wind power forecasts). These are deterministic or published before market clearing, allowing foundation models to legitimately condition on them across future steps.

HISTORICALLY OBSERVED COVARIATESStrictly restricted to history (time ≤ t)

Stochastic operational observations (such as Rossmann's actual customer footfall Customers, Rohlik's actual precipitation and snowfall, or sudden warehouse shutdowns). Because these are only realized in real-time, future values must never enter a prediction to prevent lookahead target leakage.

Observed Dataset Dynamics: Known future covariates include numerical weather predictions (temperature, humidity, atmospheric pressure, rain, snow, day length, wind speed); past dynamic telemetry includes observed cloud cover and global horizontal irradiance (GHI).
ColVariable / Feature NameRoleTypeUnit / ScaleDescription & Business Meaning
CtargetSolar Power OutputTARGETNUMMW
Actual active solar power generation measured at the photovoltaic inverter
AidSeries IdentifierSERIES IDCAT—
Unique categorical identifier indexing the photovoltaic asset or tracking installation
BtimestampObservation TimestampTIMETIME—
ISO 8601 hourly timestamp indicating the start of each observation interval
Dday_lengthAstronomical Day LengthFUTURE KNOWNNUMminutes
Total daylight duration calculated for the geographic coordinates
• Known future driver: Legitimate lookahead across forecast horizon (steps t+1 to t+H, e.g. calendar/promotion/forecast).
EhumidityRelative HumidityFUTURE KNOWNNUM%
Ambient relative humidity percentage measured or forecast at surface level
• Known future driver: Legitimate lookahead across forecast horizon (steps t+1 to t+H, e.g. calendar/promotion/forecast).
FpressureAtmospheric PressureFUTURE KNOWNNUMhPa
Barometric surface pressure influencing air density and cloud formation
• Known future driver: Legitimate lookahead across forecast horizon (steps t+1 to t+H, e.g. calendar/promotion/forecast).
Grain_1hHourly RainfallFUTURE KNOWNNUMmm
Liquid precipitation accumulated over the preceding 1-hour interval
• Known future driver: Legitimate lookahead across forecast horizon (steps t+1 to t+H, e.g. calendar/promotion/forecast).
Hsnow_1hHourly SnowfallFUTURE KNOWNNUMmm
Snowfall depth or liquid equivalent accumulated over the preceding 1-hour interval
• Known future driver: Legitimate lookahead across forecast horizon (steps t+1 to t+H, e.g. calendar/promotion/forecast).
ItempAmbient TemperatureFUTURE KNOWNNUMdeg C
Outside air temperature affecting PV panel semiconductor operating efficiency
• Known future driver: Legitimate lookahead across forecast horizon (steps t+1 to t+H, e.g. calendar/promotion/forecast).
Jwind_speedSurface Wind SpeedFUTURE KNOWNNUMm/s
Horizontal wind speed contributing to convective cooling of solar panels
• Known future driver: Legitimate lookahead across forecast horizon (steps t+1 to t+H, e.g. calendar/promotion/forecast).
Kclouds_allTotal Cloud CoverPAST ONLYNUM%
Observed percentage of total cloud cover attenuating direct solar irradiance
• Historically observed: Only available up to forecast origin t. Future values masked to prevent lookahead target leakage.
Lglobal_horizontal_irradianceGlobal Horizontal IrradiancePAST ONLYNUMW/m2
Total solar radiation incident on a horizontal surface at ground level
• Historically observed: Only available up to forecast origin t. Future values masked to prevent lookahead target leakage.
DATASET PREVIEW

Interactive Data Table

Explore rows, feature values, and target labels for Solar Photovoltaic Power Output with Weather.

Loading dataset...
GLOBAL BENCHMARK PROVENANCE (FEV-BENCH)

Foundation Models vs Traditional ML & Baselines (35 Methods)

Empirical evaluation directly from the AutoGluon fev benchmark suite. Global aggregate ranking: The win rates and skill scores below represent cross-dataset performance across all 100 standardized tasks in fev-bench, rather than results isolated to Solar Photovoltaic Power Output with Weather.

Benchmark Provenance & Latency Measurement:
100-Task Global Aggregation:The rankings evaluate zero-shot foundation models across 100 heterogeneous benchmark tasks in fev-bench, measuring aggregate generalization across diverse frequencies and prediction horizons.
FEV Python vs. In-Browser WebGPU Timing:The "FEV Latency" column reports server-side Python execution times from the official AutoGluon FEV benchmark. Independently, Carla measures and reports true client-side WebGPU execution inside Google Sheets (284ms on this dataset) with zero server reliance.
#1 Official Ranked Model?ci-proto from Continuous Intelligence ranks #1 on the 35-method official fev-bench view (90.0% SQL win rate; #3 overall at 84.37% across all 52 evaluated models) and can also run locally in our browser WebGPU demo.
ci-proto90.0% Win Rate • 0.53s FEV latency
Evaluated Methods?35 official leaderboard methods (52 total evaluated including closed APIs and checkpoint variants across foundation models, deep neural architectures, trees, and statistical baselines).
35 Official52 total • Continuous Intelligence, H2O.ai, Mistral AI, Google, AWS
Primary Evaluation Metric?Scaled Quantile Loss (SQL) penalizes both point prediction errors and quantile miscalibration across 9 probabilistic levels (p10 to p90).
SQL & MASEProbabilistic calibration
Independent Browser WebGPU?Carla Engine executes Timestral-123M locally in Google Sheets via WebGPU shaders, while ci-proto can also be tested directly in-browser via ONNX Runtime WebGPU and Rust/WASM above.
284ms / windowTimestral-123M (Carla Engine) • Zero server upload
RankMethod / ModelFamilyOrganization
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.
FEV Latency?FEV benchmark reference latency measured in server Python (median seconds per 100 series). Independent in-browser WebGPU execution for Carla is reported in the telemetry card above.
Link
#1ci-protoFoundation ModelContinuous Intelligence90.0%+51.5%88.4%0.53sRepo ↗
#2TimeH2O-v1Foundation ModelH2O.ai89.7%+50.8%88.2%2.82sRepo ↗
#3Timestral-209MFoundation ModelMistral AI87.1%+50.4%85.4%0.51sRepo ↗
#4t0-gammaFoundation ModelThe Forecasting Company84.7%+49.4%82.0%4.38sRepo ↗
#5Simons-297MFoundation ModelAbel AI Lab82.0%+48.4%80.9%0.94sRepo ↗
#6timesfm-3Foundation ModelGoogle79.3%+49.0%77.6%3.41sRepo ↗
#7iAmTime-base-forecastFoundation ModelWalmart Labs73.7%+46.8%72.5%0.56sRepo ↗
#8Chronos-2Foundation ModelAWS72.2%+47.3%69.9%0.84sRepo ↗
#9t0-betaFoundation ModelThe Forecasting Company69.9%+46.6%66.9%2.29sRepo ↗
#10Toto-2.0-2.5BBaselineCommunity69.4%+44.4%67.8%5.58sRepo ↗
#11TiRex-2Foundation ModelNX-AI69.2%+45.5%66.3%0.27sRepo ↗
#12TiRexFoundation ModelNX-AI62.2%+42.6%58.0%0.24sRepo ↗
#13Toto-2.0-22mBaselineCommunity61.6%+42.9%59.4%0.25sRepo ↗
#14TimesFM-2.5BaselineCommunity58.4%+42.2%56.7%1.87sRepo ↗
#15TabPFN-TS-3Foundation ModelPrior Labs55.5%+43.1%53.3%235sRepo ↗
#16TS-ICLFoundation ModelEDF-Lab55.1%+43.1%53.4%2.49sRepo ↗
#17citras-fmFoundation ModelHitachi52.9%+41.2%50.6%0.60sRepo ↗
#18FlowStateFoundation ModelIBM52.7%+39.9%53.0%2.29sRepo ↗
#19Toto-1.0BaselineCommunity51.6%+40.8%48.8%22.35sRepo ↗
#20TabPFN-TSFoundation ModelPrior Labs50.5%+41.5%48.0%109sRepo ↗
#21Moirai-2.0BaselineCommunity46.2%+39.3%44.2%0.35sRepo ↗
#22Chronos-BoltFoundation ModelAWS45.4%+38.9%42.9%0.26sRepo ↗
#23TFTDeep Learning—34.7%+32.2%31.6%829sRepo ↗
#24Sundial-BaseFoundation ModelTsinghua University30.8%+33.4%37.0%8.01sRepo ↗
#25Stat. EnsembleBaselineCommunity30.6%+20.7%33.6%147sRepo ↗
#26PatchTSTDeep Learning—30.0%+30.1%27.8%664sRepo ↗
#27DeepARDeep Learning—28.9%+29.1%25.9%1113sRepo ↗
#28AutoARIMAStatistical Baseline—26.6%+21.3%24.2%20.14sRepo ↗
#29AutoETSStatistical Baseline—22.5%-26.8%24.1%3.48sRepo ↗
#30CatBoostTree Ensemble—21.8%+23.0%37.5%29.48sRepo ↗
#31LightGBMTree Ensemble—20.3%+21.0%35.7%3.03sRepo ↗
#32AutoThetaStatistical Baseline—17.9%+5.5%22.2%3.28sRepo ↗
#33Seasonal NaiveBaselineCommunity12.0%0.0%13.2%0.48sRepo ↗
#34NaiveStatistical Baseline—7.8%-45.4%12.2%0.47sRepo ↗
#35DriftStatistical Baseline—7.0%-45.8%11.1%0.45sRepo ↗
STATISTICAL DISTRIBUTIONS

Variable Schema & Summary Distributions

Observed numerical ranges, category cardinalities, missing rates, and sample values across 49,648 rows.

ColVariable NameTypeMissingDistinctSummary Stats / DistributionSample Values
CtargetTARGETNUM010,349[0, 19008] μ=2271.1 σ=4066.2000
AidCAT01—000
BtimestampCAT049,648—2017-01-01T01:00:002017-01-01T02:00:002017-01-01T03:00:00
Dday_lengthFUTURE KNOWNNUM456 (0.92%)39[450, 1020] μ=748.6 σ=194.9450450450
EhumidityFUTURE KNOWNNUM456 (0.92%)79[22, 100] μ=79.8 σ=15.6100100100
FpressureFUTURE KNOWNNUM456 (0.92%)71[977, 1047] μ=1015.3 σ=9.6102010201019
Grain_1hFUTURE KNOWNNUM456 (0.92%)311[0, 8.09000015258789] μ=0.1 σ=0.3000
Hsnow_1hFUTURE KNOWNNUM456 (0.92%)129[0, 2.819999933242798] μ=0.0 σ=0.1000
ItempFUTURE KNOWNNUM456 (0.92%)503[-16.600000381469727, 35.79999923706055] μ=9.8 σ=8.01.70000004768371581.8999999761581422
Jwind_speedFUTURE KNOWNNUM456 (0.92%)136[0, 14.300000190734863] μ=3.9 σ=1.85.1999998092651375.55.699999809265137
Kclouds_allPAST ONLYNUM456 (0.92%)101[0, 100] μ=66.0 σ=36.6100100100
Lglobal_horizontal_irradiancePAST ONLYNUM456 (0.92%)8,553[0, 227.8249969482422] μ=32.6 σ=51.7000
FREQUENTLY ASKED QUESTIONS

Frequently Asked Questions: Solar Photovoltaic Power Output with Weather

Common questions regarding the Solar Photovoltaic Power Output with Weather dataset, machine learning task formulations, and in-browser tabular inference.

What is the forecasting horizon for Solar Photovoltaic Power Output with Weather?

The canonical horizon is 24 hours (H=24) at an hourly cadence, designed to mirror day-ahead electricity market clearing and transmission scheduling requirements.

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 moving data to external servers.

Which models perform best on Solar Photovoltaic Power Output with Weather?

Mistral's Timestral zero-shot foundation model and LightGBM with numerical weather covariates outperform classical univariate baselines (AutoARIMA, Seasonal Naive) by capturing non-linear weather dependencies.

LIVE EVALUATION IN GOOGLE SHEETS

Forecast with Carla: Solar Photovoltaic Power Output with Weather

Open the pre-loaded Google Sheet and let Carla generate multi-step probabilistic forecasts using Mistral's Timestral directly inside your browser with zero server uploads.

Step 1

Launch Carla

Click once to open the spreadsheet and Carla side panel together.

Step 2

Review the Setup

Carla auto-detects the target sequence, frequency cadence, and temporal covariates.

Step 3

Forecast & Calibrate

Generate multi-step quantile forecasts with zero server uploads.

Provenance & Attribution

NREL Solar Power and Meteorological Telemetry Benchmark.

License: Open AccessData Source: autogluon/fev_datasets