# Electricity consumption and PV production and forecast on 4 residential buildings

## Project Description

This file contains - Measurements of electricity consumption in the 4 buildings of the E4C Smart Grid Thermal Electric demonstrator - Estimated (or real) PV production based on SIRTA measurements - Predicted PV from MéteoFrance forecasts (ARPEGE model) day ahead. All with a step of 5 minutes. The PV capacity considered (62 kWp) is that which is planned for these buildings, and which is currently being installed. So these are realistic values.

---

## Standardized Metadata

### File Information

- **File Name:** SmartGTE_pv_conso_forecast_20240801-20260211.csv
- **Creation Date:** 2026-02-15
- **Data Date Range:** 2024-08-01 to 2026-02-11
- **Contact:** [e4c_datahub@ip-paris.fr](mailto:e4c_datahub@ip-paris.fr)
- **DOI:** https://doi.org/10.14768/094a7485-b312-482c-8cb6-59d1f5a72acf
- **License:** [CC BY-NC 4.0 ](https://creativecommons.org/licenses/by-nc/4.0/) | The use of data and code for AI training is forbidden without explicit authorization.

### Data Collection Methodology

This file combines electric consumption measuerements from 4 buildings of Ecole polytechnique with photovoltaic (PV) production estimation using SIRTA's radiative, temperature and wind speed measurements (https://sirta.ipsl.fr/) and PV forecast from ARPEGE MeteoFrance Numerical Weather Predictions (of solar irradiance, air temperature and wind speed) 

PV is computed (whether from SIRTA measurements or ARPEGE forecasts) from solar irradiance, wind speed and air temperature, using python pvlib v0.13 functions, considering a PV installed capacity of 62 kWp with PV modules tilted 10º towards South. 
In particular, the methods used are : 
- plane of array irradiance :  pvlib.pvsystem.sapm_effective_irradiance
- PV cell temperature : pvlib.temperature.sapm_cell
- PV DC power : pvlib.pvsystem.pvwatts_dc
- PV AC power : pvlib.inverter.pvwatts

Time step : 5 minutes 
- For Consumption : native resolution is 5 minutes (instantaneaus power samples)
- For ARPEGE : 5-minute interpolation computed from the 1-hour native time step
- For SIRTA measurements : 5-minute averages computed from the 1-min native time step  

Missing Data :
The file has two main periods of missing data:
- For consumption: Approximately between January 6 and 18, 2025
- For the AROME-PEARP forecast: Approximately from November 20, 2024, to January 9, 2025

### Acknowledgement request

Users of these data are kindly requested to acknowledge the E4C DataHub service when publishing results derived from this dataset. Proper acknowledgment helps us demonstrate the scientific impact of the infrastructure and supports the continued operation and development of the service.

When using these data in a publication, presentation, or any public dissemination of results, please include one of the following acknowledgments:

#### Recommended acknowledgment
Data used in this work were accessed through the E4C DataHub, a data service of the Energy4Climate Interdisciplinary Center (E4C) at Institut Polytechnique de Paris. The Energy4Climate Interdisciplinary Center is partly supported by the 3rd Programme d’Investissements d’Avenir [ANR-18-EUR-0006-02] and by the Foundation of Ecole polytechnique with the private sponsor Fonds Ifker pour le Climat, financed by Stéphane and Agnès Ifker.

#### Recommended acknowledgment (short version)
This work uses data provided by the E4C DataHub of the Energy4Climate Interdisciplinary Center (E4C), Institut Polytechnique de Paris, partly supported by the 3rd Programme d’Investissements d’Avenir [ANR-18-EUR-0006-02] and by the Foundation of Ecole polytechnique with the private sponsor Fonds Ifker pour le Climat.

#### Dataset citation
Badosa, J., Vidal, B., (2026). Electricity consumption and PV production and forecast on 4 residential buildings. E4C DataHub, Energy4Climate Interdisciplinary Center (E4C), Institut Polytechnique de Paris. https://doi.org/10.14768/094a7485-b312-482c-8cb6-59d1f5a72acf

### Authors and contributors

Jordi Badosa
Bastien Vidal

---

## Standardized Variables Section

### Variable Definitions (Machine-Readable Format)

```YAML
variables:
  - raw_name: "Date and time (UTC)"
    descriptive_name: "Timestamp"
    description: "Date and time in UTC. Each timestamp corresponds to the starting time for the measurement period (5 min)."
    unit: "YYYY-MM-DD HH:MM:SS"
    
  - raw_name: "power_conso_bat_a"
    descriptive_name: "Electric power consumption from building A"
    description: "Electric power consumption from building A (building 43)"
    unit: "kW"
    
  - raw_name: "power_conso_bat_b"
    descriptive_name: "Electric power consumption from building B"
    description: "Electric power consumption from building B (building 44)"
    unit: "kW"
    
  - raw_name: "power_conso_bat_c"
    descriptive_name: "Electric power consumption from building C"
    description: "Electric power consumption from building C (building 45)"
    unit: "kW"
    
  - raw_name: "power_conso_bat_d"
    descriptive_name: "Electric power consumption from building D"
    description: "Electric power consumption from building D (building 66)"
    unit: "kW"

  - raw_name: "pv_real"
    descriptive_name: "Estimated total PV power"
    description: "Total PV power estimations using solar irradiance, air temperature and wind speed measured at SIRTA observatory"
    unit: "kW"

  - raw_name: "pv_arpege"
    descriptive_name: "Forecasted total PV power"
    description: "Total PV power forecast using ARPEGE numerical weather prediction's outputs: solar irradiance, air temperature and wind speed"
    unit: "kW"

  - raw_name: "airtemp_arpege"
    descriptive_name: "Air temperature (2 m above the ground) forecasted by ARPEGE"
    description: "Air temperature (2 m above the ground) forecast from ARPEGE"
    unit: "A"

```