(Photo: Freepik)
The latest research, conducted by academics at the University of Surrey in the UK, suggests that AI may help revolutionize carbon capture technology.
The climate solution is one that industry and other heavy emitters worldwide are relying on to decarbonize their operations and mitigate CO2 emissions. Therefore, the prospect of radically reducing energy consumption would significantly expedite its implementation.
The new research shows that AI models could help optimize the performance of carbon capture, by both reducing energy use and increasing the amount of captured CO2 at the same time, according to Professor Jin Xuan from the University of Surrey.
When power plants burn fuel, they produce CO2. But it can be captured by bubbling the flue gas through water containing limestone. CO2 reacts with the calcium carbonate in the limestone and produces bicarbonate, in a process called “enhanced weathering”.
It takes energy to pump water and CO2. The CO2 capture plant had its own wind turbine but in calmer weather, it took energy from the National Grid.
Researchers used AI to teach a model system to predict what would happen so it can pump less water when there was less CO2 to capture, or when renewable energy availability was lower.
They claim that with the assistance of AI models, they can potentially reduce the energy used by a carbon capture unit at a coal-fired power plant by over a third. It could also capture 16.7% more carbon dioxide emissions.
The team hope their findings can be applied more widely across the industry, contributing towards UN Sustainability Goals (SDGs) 7, 9, 12 and 13.
The new research is released just in time, as the UK government prepares to develop carbon capture clusters in the Humber and Scotland, as announced last year.
These clusters are to be formed by major industrial sites, equipped with carbon capture technology that is to be predominantly powered by renewable energy available on site.
However, due to weather changes, switching to power from the grid will be inevitable from time to time. This is where Surrey university’s AI models come into play, as these can take available energy, plant output, and other crucial factors into account to help mitigate this issue.
