Smartgrid simulator

An EV smartgrid python simulation based on real PV, and user profiles.

---

Installation

This project uses Python 3.9 along with venv. To set up virtual environment :

Linux :

python3 -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt

Windows :

python -m venv .venv
.\.venv\Scripts\activate
pip install -r requirements.txt

---

Download DATA

Weather data

The weather data file should be downloaded from PV-GIS using the hourly data tool with the two-axis option. Remove the first 8 lines from the downloaded CSV so that it changes from this:

"Latitude (decimal degrees):	49.102"
"Longitude (decimal degrees):	2.091"
"Elevation (m):	89"
"Radiation database:	PVGIS-SARAH3"


Slope: - deg. 
Azimuth: - deg. 
time,G(i),H_sun,T2m,WS10m,Int
20210101:0010,0.0,0.0,-1.8,1.66,0.0

to this :

time,G(i),H_sun,T2m,WS10m,Int
20210101:0010,0.0,0.0,-1.8,1.66,0.0

The file can then be renamed (e.g., WEATHER_2020.csv) and placed inside the data/ folder.

Traffic data

The traffic data file should be downloaded from AVATAR, simply select a sensor and download the flow rate ($veh/h$) with an hourly rate :

Horodate,Mrm50.31,Mrm50.32,Moyenne,Somme
2025-07-01T00:00:00+02:00,103,94,98.5,197

The file can then be renamed (e.g., TRAFFIC_2020.csv) and placed inside the data/ folder.

AI training

It is recommended to have at least three years of data at an hourly resolution for both traffic and weather.

---

Usage

Via Jupyter notebook

Ensure you have a local jupyter notebook setup and running. Use the file located in notebooks to play with the simulator. It is recommended to use the local virtualenv created before in order to run without any bugs the notebooks. You can try your installation with the jupyter/visualizing/EditTraffic.ipynb one.