Our solar power prediction system is based on our modular prediction platform Anemos, a commercial spin-off of 25 years of research and development activities that is today implemented on a world-wide scale.
The high accuracy of our wind forecasts rests on a consistent multi-model approach. Each prediction involves combining multiple weather models as well as different physical and statistical power prediction models in such a way that an optimal accuracy is achieved at every point in time.
In addition to our own developments, we can draw on the models of several partners who are among the top providers worldwide.
Utility-scale solar farms as well as increasing amounts of distributed rooftop solar PV installations pose new challenges to electricity grid and market operation. We are able to provide accurate forecasting for all types of solar power, large-scale farms and distributed, fixed and tracking.
Our advanced solar forecasting models have been further developed in order to precisely describe the path of the sun over the sky, support the use of satellite images, support global meteorological models with relatively low temporal resolution, and include optimal combination of several meteorological models, statistical downscaling, shadow detection, and more.
Any prediction needs to include a specification of the related statistical uncertainty. Through an advanced statistical analysis of past behavior, we can specify the current accuracy of our predictions on a reliable basis.
Due to the modular structure of our forecasting platform, it is open to the excellent forecasting models of our commercial and research partners. The fast implementation of results from current research leads to the continuous improvement of our predictions.
Our long-standing experience in wind and solar power predictions is utilized for other prediction services as well: Direct marketing of renewable energies, load predictions and predictions of thermal load for district heating networks.
For our customers, predictions are an essential component of their business processes. For this reason, we do our utmost to ensure a high availability of our systems. With our years of experience, redundant server systems, quality management and a support team available 24/7, we maintained 100% availability in the last 10 years. This applies to server solutions operated by us for our customers as well as to systems integrated on-site into the IT infrastructure of our clients. Our prediction system was intensively tested by customers both for stability and maintaining No-Single-Point-of-Failure criteria.
As part of the interplay between research institutions and commercial companies, we continuously improve state-of-the-art solutions along with the commercial application side of wind and solar power predictions. We have been working in the area of power predictions for over 25 years and, together with our Anemos partners, are currently responsible for predicting more than 120 GW of wind and solar energy worldwide.
For the majority of our applications we integrate the advanced statistical and mathematical models of our long-term partner ENFOR. ENFOR was established in 2006 as a spin-off from the Technical University of Denmark, and is one of the world-leading companies in the field of statistical modelling.