MoNuTeR project: Better forecasting solar production to reduce costs of variability

null

Innovation

23/12/2021

by

Guillaume Tremoy

5 min

null

null

null

The predictability of solar resource variability presents several limitations, as meteorological models are imperfect and the state of the atmosphere is never fully known. Therefore, the improvement of forecasting systems represents a major challenge for integrating more PV at lower cost, especially in poorly and non interconnected regions.

Funded by the French agency for ecological transition (ADEME, PIA 3), the MoNuTeR project consisted in developing, optimizing and industrializing a new generation of weather forecasting technologies and services specially adapted to the solar energy sector. The forecasting chain, which relies on the use of a high-resolution regional numerical weather prediction model, makes it possible to locally improve the accuracy of forecasts up to 48 hours ahead and to better anticipate the intraday solar resource variability.

The project consists in developing a forecasting chain based on the WRF meteorological model, with a highly flexible configuration perfectly suited for solar applications.

null

null

null

null

During the last three years, we have:

- Developed an advanced weather forecasting chain, based on the WRF model (Weather Research and Forecasting)
- Tested the model on more than ten climatic regions (insular, desertic, tropical, etc.)
- Realized more than 4,000 simulations at spatial resolution reaching 1 km (> 10,000 h of calculation in total)
- Predetermined ideal model configurations depending the region of interest and the associated local weather phenomena
- Industrialized the solution in order to easily and quickly deploy the model over new areas, while ensuring the best possible results from commissioning
- Reduced our forecast errors by around 5-10% on average in insular regions

null

Examples of downward surface solar radiation forecast from the WRF model over Tahiti island compared to satellite observations.

Examples of downward surface solar radiation forecast from the WRF model over Tahiti island compared to satellite observations.

MoNuTeR project: Better forecasting solar production to reduce costs of variability

null

null

null

null

Validated in complex environments, the solution is now integrated into Steadysun’s new operational platform and used for decision-making.

Several projects have already been set up, as for example over the island of Tahiti on behalf of the local grid operator (EDT ENGIE). In the future, other key players will also benefit from this operational system in this region. For instance, independent producers will soon be operating hybrid PV-storage power plants that will require the use of the most reliable forecasts possible to comply with the applicable grid code.

Following on from MoNuTeR, new R&D work has already been initiated, aiming at improving in-situ/satellite data assimilation techniques as well as the modelling of cloud-aerosol-radiation interactions, especially in desertic regions.

null

null

null

null

null

null

null

null

null

null

null

null

null

null

null

null

null

null

null

null

null

null

null

null

null

null

null

null

null

null

null

null

null

null

null

null

null

null

null

null

null

null

null

null

null

null

null

null

null

null

null

null

null

null

null

null

null

Get In Touch  

Any questions? Please contact us.   

FOLLOW US

© Copyright 2024 Steadysun sas. All Rights Reserved.