Predict Energy Production Using Data-Driven Models
Utilizing historical solar radiation data, this example predicts energy production from solar panels through data-driven modeling. By incorporating radiance data from external databases into a photovoltaic (PV) plant, it simulates energy output and addresses uncertainty in power generation.
To run this example, you'll need
The latest versions of System Modeler and Mathematica.
Please make a selection:
Get afree trial Continue
with download
Connect with External Databases
Collect radiation data for a location from PVGIS. Define location and other attributes required by the API. Execute the URL and import the result.
Radiation data is collected for 10 years for Illinois, USA.
Process the raw data and check the irradiance plots.
During the summer, the irradiance is at maximum, peaking at around 1000 
, whereas during the winter, the irradiance is at minimum, around 500 
.
, whereas during the winter, the irradiance is at minimum, around 500 .Plot the Daily Irradiance Variation
Split the time series into yearly data. Overlay the yearly time series and observe the daily variation.
A wide variation in irradiance data can be observed.
PV Plant
Create a model of the PV plant that has 20 modules attached to a constant load.
A PV plant with a maximum power rating of 4 kW consisting of 20 modules.
Predict Energy Production with Uncertainty
Use the historical irradiation data for 10 years and predict the monthly power production by the solar plant.
Maximum amount of energy is produced in the month of June, whereas minimum energy is produced in December.
Wolfram System Modeler
Try
Buy
System Modeler is available in English
and Japanese
on Windows, macOS & Linux »
Questions? Comments? Contact a Wolfram expert »