Discussions of energy and climate change often focus on the need to reduce or modify energy end-use to cut down on emissions, which have the potential to accelerate average global temperature increases and lead to changing weather patterns. But researchers at IBM are focused on the inverse of that relationship – looking more closely at the weather, in part to help reduce energy consumption.
IBM’s Deep Thunder program is designed to provide clients with accurate, short-term forecasts of weather impacts to enable more efficient management of resources. It has wide-ranging practical applications, from water allocation for agricultural irrigation to mitigating floods to electricity use.
IBM is working with the University of Brunei Darussalam to analyze the country’s energy data in a bid to move towards more efficient allocation and distribution of electricity, such as through the application of energy ratings to domestic and commercial appliances. Their collaboration builds on a three-year partnership to improve weather and climate models, with a focus on key national interests, such as emergency management and sustainability.
“We started to focus on short-term weather and how that has impacts on various sectors within Brunei, with public safety being an initial driver with severe storms,” IBM Distinguished Engineer and Chief Scientist of Deep Thunder Lloyd Treinish told Breaking Energy.
These efforts to improve the accuracy of short-term weather forecasting, in combination with the Brunei 2035 plan, which seeks to enhance the country’s security and sustainability, among other goals, have created a fertile environment for forward-thinking approaches to using better data on both weather and energy use to cut down on consumption. “From an energy perspective and a utility perspective, most executives that run those companies will probably tell you that weather is the single most important external influence,” Treinish said.
“[Weather] connects directly to energy demand, to energy load on their systems, whether it’s the distribution or transmission of generation,” Treinish said. “If you’re a company that uses renewable energy, that’s driven directly by the weather.”
Better weather forecasting offers opportunities for pre-emptive energy management that can capitalize on external weather conditions, in the same way that advance knowledge of peak energy demand periods can allow energy consumers to capitalize on price differentials between peak and off-peak demand periods. “With regard to buildings, we want to understand what the weather will be to determine how to control the energy usage within the building,” Treinish said.
“This is not so much an issue for individual homes, but for larger buildings where the systems do not respond instantaneously,” he said. “There’s latency in the thermal systems within large buildings – you can’t do things in real time. For certain types of buildings with specific features, you can take advantage of predictive weather to make adjustments.”
One such option is the use of fans to bring in cool, dry air from outside, rather than maintaining cool internal temperatures entirely through air-conditioning, provided weather conditions are right, Treinish said. “It’s much cheaper.”
But he noted that quality of information is critical to make this cost-effective. “You need to know [weather conditions] very precisely many hours ahead of time, and it has to be for a long enough period that you end up saving money, not spending more,” he said.
And certain options are only applicable to specific buildings and specific climates. Using cool air from outdoors might not be feasible very often in Brunei, in the same way that incorporating solar heat-loading into a building may not offer year-round opportunities in Finland.
Accurate short-term weather forecasting can also allow large buildings and facilities to more effectively cut back on consumption during peak demand periods, when energy is often much more expensive. “If you can predict a day or two ahead of time the likely maximum energy that you might be consuming, you can make adjustments to reduce peak load for that day, and that may reduce the maximum for the month, keeping the overall [energy] costs down,” Treinish said.
Looking ahead, as understanding of weather patterns improves along with data collection and analysis, new buildings and retrofits can be designed to better account for local climate conditions, Treinish said.
