Weather confidence for district heating
Reduce the uncertainty in your heat demand predictions with a hyperlocal temperature forecast that is up to 50% more accurate than traditional forecasts.
A more accurate forecast where it matters most
Weather is the single biggest factor affecting the daily demand for district heating. Even small forecast errors can greatly impact operational costs and efficiency. Unfortunately, traditional weather forecasts lack local precision and leave district heating companies exposed to sudden shifts in conditions.
A more accurate forecast—powered by the latest advances in weather sensing and machine learning—enables district heating companies to improve operational efficiency, lower costs, and reduce waste. Xweather offers a unique hyperlocal forecasting solution that includes sensors, software, and advanced AI-powered forecasting as a complete service.
Improve your heat demand forecast with local weather observations
Temperatures in metropolitan areas can vary greatly depending on the topography, variations in land use, proximity to water, and the unique built environment. The first step to improving your weather forecast is to account for these variations by measuring the local environment.
Wireless Xcast sensors are easily deployed at the locations that matter to you. Our experts will assist with site selection and manage the sensor network on your behalf. The service includes remote monitoring, quality-controlled data, and a continuous hardware warranty.
Enhanced forecasts, powered by observations from your local sensors
Our enhanced forecasts are powered by Xcast, a technology that uses machine learning, trained on your local sensor observations, to deliver a more accurate forecast for the locations that matter to you. Xcast improves accuracy at all forecast horizons achieving an average 50% improvement across all deployed sensors. Forecasts are updated every 15 minutes so that you can make decisions with the latest information.
Success story
59% fewer large errors in the 24-hour forecast
Fortum is a Nordic energy company operating in areas where district heating is in high demand during the cold winters. Fortum worked with Vaisala Xweather to implement hyperlocal forecasting for its district heating network in Helsinki and its surrounding cities.
Challenge
There are significant variations in local weather conditions across Fortum’s large district heating network. Large differences in topography, vegetation, and the built environment, make accurate forecasting challenging.
Solution
Vaisala Xweather delivered a hyperlocal forecasting solution using a machine-learning model trained on real-time observations from a managed network of 26 weather sensors installed at key locations across Fortum’s network.
Results
The Xweather solution improved the accuracy of Fortum's 6-hour forecast by up to 36% and reduced the number of errors greater than 2.5 °C in the 24-hour forecast by 59%.
"Our observation network of Vaisala weather stations provides us with accurate local data of critical weather parameters, enabling us to optimize heating supply temperature more precisely than before."
Viki Kaasinen
Head of Asset Digitalization
Fortum
Overcoming weather challenges in district heating
The essential guide to hyperlocal forecasting
When even small errors in the forecast can greatly impact your operational efficiency, you need weather confidence to make the right decisions. Read our 23-page guide to learn how to improve your operational efficiency, lower costs, and reduce waste with hyperlocal forecasts.
Vaisala AtmoCast sensor network
Vaisala-managed network of wireless weather sensors with excellent accuracy, hassle-free installation, and reliable year-round operation.
Xweather Optimize portal
Cloud-based UI for configuration and visualization and an API for integrating observations and forecasts into your daily operations.
Xcast technology
Xcast technology combines your sensor observations with machine learning to improve the forecast at each location.