If we adopt artificial intelligence in agricultural plant factories (large indoor farms with full lighting and climate control) around the world, we can promote photosynthesis, transpiration and respiration of crops in these buildings,...we can expect to see significant reductions in energy while increasing efficiency and saving valuable resources‘ said Benjamin Decardi-Nelson, a postdoctoral researcher in the laboratory of Fengqi You, the Roxanne E. and Michael J. Zak Professor of Energy Systems Engineering at Cornell's College of Engineering.
The research ‘Artificial intelligence can regulate light and climate systems to reduce energy consumption in plant factories and support sustainable food production’ was published on 9 September in the journal Nature Food. ‘Existing environmental control systems are not smart enough,’ said Fengqi You, co-director of Cornell University's Institute for Digital Agriculture and co-director of Cornell's Institute for Artificial Intelligence Science.
Ventilation reduces energy consumption but can complicate plant growth by affecting CO2 levels and water balance. Artificial intelligence tools can help regulation methods take this criterion into account. ‘Artificial intelligence offers a promising solution by managing multiple complicating factors,’ says Benjamin Decardi-Nelson.
Byusing AI techniques such as deep reinforcement learning and computational optimisation, scientists analysed lettuces grown in eight different regions of the United States, including Los Angeles, Chicago, Miami, Seattle, Milwaukee, Phoenix, Fargo, North Dakota, and Ithaca, New York, as well as indoor agricultural facilities in Reykjavik, Iceland, and Dubai, UAE.
Artificial intelligence reduces energy consumption by optimising lighting and climate control systems. With the application of artificial intelligence technology, thefresh weight per kilogram (the amount of energy required or used to produce one kilogram of indoor-grown lettuce) was reduced to 6.42 kilowatt-hours from 9.5 kilowatt-hours without the application of AI technologies. The researchers found that in warmer climates, such as Dubai or the southern United States, the AI technology could reduce energy consumption from 10.5 kWh to 7.26 kWh per kilogram of fresh weight.
Low ventilation during periods of light (16 hours of simulated daylight) and high ventilation during periods of darkness (8 hours of simulated night) provide an energy-efficient solution that optimises indoor carbon dioxide levels for photosynthesis, respiration, and oxygen levels for plant growth, and balances other ventilation requirements.
This is very similar to the concept of a smart home, where we want to be comfortable at home while reducing energy consumption; the same is true for crops. The focus of this work isBuilding an intelligent system to optimise food production, achieve sustainability and reduce carbon footprints. That's what AI is good at. If we use AI to carefully optimise artificial lighting and other energy systems, we can save a lot of money.
By using AI to streamline operations to reduce energy consumption, indoor farms are becoming viable even in areas with limited energy-saving opportunities.
“By strategically integrating environmental control system technologies with plant biology,’ says Benjamin Decardi-Nelson, ’ventilation can be used to save energy while minimising CO2 waste and maintaining ideal growing conditions.’
Financial support for this research came from the U.S. Department of Agriculture (National Institute of Food and Agriculture), the Natural Sciences and Engineering Research Council of Canada, and the Eric and Wendy Schmidt AI in Science Postdoctoral Fellowship.
Article content from the ‘verticalfarmdaily’ official website
English 
Chinese