Around 350 BC, Aristotle made the first meteorological predictions by studying the similarities between air and water. As it turns out, his approach to atmospheric science was not far off. In the last decades, the art of weather forecasting got finetuned with large-scale technological innovation. Currently, Artificial Intelligence (AI) is writing the next chapter of meteorological innovation. In this article, we will explore how Peltarion, Sweden’s leading AI innovator and a Google Cloud customer, leverages AI-driven meteorology to stimulate innovation in the renewable energy sector.
Forecasting services are always on the lookout for change and innovation, and AI offers the perfect opportunity to do so. By driving innovation in atmospheric prediction models and forecasting systems, AI enables meteorological institutes to provide the most precise weather predictions and alerts.
Innovation with AI impacts power supply services and energy companies as well. For example, weather predictions based on digital databases and state-of-the-art workstations offer the opportunity to anticipate meteorological phenomena that could cause issues in power operations. As such, utility companies can improve their current services based on improved AI-powered forecasting methods.
Adequate energy storage and power production from renewable sources are directly affected by weather conditions. Therefore, Tekniska Verken, a Swedish energy company, relies heavily on weather forecasts to predict the amount of electricity its wind turbines must produce. Moreover, the electricity from wind turbines must be put into the energy grid 24 hours in advance – rendering imperfect weather forecasts exceptionally damaging. In sum, Tekniska Verken requires enhanced forecasting to schedule power resources and manage its power production efficiently.
Sweden’s leading AI technology lab, Peltarion, designed an AI-powered model to improve weather forecasting and monitoring for Tekniska Verken. The model, named Deep Weather, can process large datasets of past weather and operational electricity requirements in specific locations. The model enables Tekniska Verken to create all sorts of technical solutions, such as offering updated forecasts every 10 seconds and automatically adjusting its weather systems, saving billions in energy waste.
Deep Weather has already demonstrated its added value in improving efficiency and reducing operational costs. As compared to earlier systems, Tekniska Verken can now make the same predictions in 100 milliseconds, at a fraction of its earlier computing costs. Apart from optimizing wind electricity production, the system has the potential to spark all kinds of innovations. For example, office buildings could be provided with updated forecasts every 10 seconds and automatically adjust their climate systems accordingly. Off-piste skiers and mountain climbers could enter the wilderness with full confidence that adverse weather will not leave them stranded. Or, as Erik Olsson, Business Developer at Tekniska Verken, puts it:
“We have an obligation to always challenge the boundaries of what is currently possible. Deep Weather has demonstrated that it is possible to rethink how to solve complex problems with AI, and that it’s possible to leverage the downsides facing wind power into advantages that will ultimately lead to more renewable energy.”
— Erik Olsson, Business Developer at Tekniska Verken