AI weather forecaster helps conserve energy

11 February 2019

Researchers at Cornell University have developed an artificial intelligence (AI) framework that can more accurately predict weather and better inform building control systems, allowing them to conserve significant amounts of energy.
Cassie Sims

The weather has a huge impact on our lives, from what we wear to where we go. We rely on weather forecasts to plan our day, but unfortunately these are often imperfect and inaccurate. For an individual, this might result in minor inconveniences, such as being caught in the rain without an umbrella, but for building engineers it creates a larger problem. Large amounts of energy can be wasted in large buildings with unnecessary cooling and heating, if the weather forecast is wrong.

Newer buildings tend to have sophisticated and adaptable temperature control systems to allow them to conserve energy as best as possible, however, these still rely on weather forecasting tools. By combining these temperature systems with a machine learning model, researchers at Cornell University have developed a smart control system that can reduce energy usage of a building by up to 10 percent.

‘If the building could be ‘smart’ enough to know the weather conditions, or at least somehow understand a little bit more about the weather forecasting information, it could make occupants more comfortable,’ said Fengqi You, Professor in Energy Systems at Cornell.

‘For instance, if I know the sun is going to come up very soon, it's going to be warm, then I probably don't need to heat the house so much. If I know a storm is coming tonight, then I try to heat up a little bit so I can maintain a comfortable level,’ Professor You added. ‘We try to make the energy system smart, so it can predict a little bit of the future and make the optimal decisions.’

The machine learning model has been trained with historical weather forecast data, combined with a mathematical model that considers characteristics of the building such as construction materials, room sizes and positions of windows and sensors. By combining these two methods, a more accurate control system is created.

This framework has potential applications in building control, agriculture and specialist indoor environmental control systems to help adapt them to save significant amounts of energy.

DOI: 10.1016/j.jprocont.2018.12.013

Related Links:

Show me news from
All themes
All categories
All years
search by