Integrating artificial intelligence into current environmental control systems might cut energy usage for indoor farming by 25%, offering a significant boost to global food production as the population continues to grow, according to engineers from Cornell University.
Incorporating artificial intelligence within existing environmental control systems could lead to a 25% reduction in energy consumption for indoor farming, potentially assisting in feeding a growing global population, researchers at Cornell University have discovered.
This finding was published in the journal Nature Food.
The United Nations predicts that the world population will reach 9.7 billion by the year 2050. This increase, along with the impacts of climate change and urban development, necessitates improvements to the current food production methodologies, according to the research team.
Indoor farming techniques, like plant factories using artificial lighting, face fewer challenges from climate variations, yet they consume a lot of energy and call for meticulous resource management to remain viable.
“Current environmental control systems lack intelligence,” stated Fengqi You, a professor of energy systems engineering at Cornell.
Employing AI strategies, such as deep reinforcement learning and computational optimization, the researchers studied lettuce grown in indoor facilities across various locations in the U.S., including Los Angeles, Chicago, Miami, Seattle, Milwaukee, Phoenix, Fargo, North Dakota, and Ithaca, New York, as well as in ReykjavÃk, Iceland, and Dubai, UAE.
Artificial intelligence helps minimize energy consumption by fine-tuning lighting and climate control systems. With AI, energy use decreased to 6.42 kilowatt hours per kilogram of fresh weight (the amount of energy needed to produce one kilogram of indoor-grown lettuce), down from 9.5 kilowatt hours per kilogram in facilities that do not use AI. In warmer regions like Dubai and the southern U.S., energy usage was reduced to 7.26 kilowatt hours per kilogram fresh weight, a drop from 10.5 kilowatt hours per kilogram.
Employing low ventilation during light phases (16 hours of simulated daylight) and high ventilation during dark phases (eight hours representing night) proved to be an efficient method for maintaining ideal indoor carbon dioxide levels for photosynthesis, providing oxygen for plant growth, and fulfilling various ventilation needs.
“This concept is akin to smart homes,” noted You. “We aim for comfort at home while cutting down on energy usage; crops have similar requirements. This project centers on developing a smart system that enhances food production while being sustainable and reducing carbon footprints. AI excels in this regard, and we can achieve significant savings by using AI to optimize our artificial lighting and energy systems effectively.”
This research received financial backing 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 at Cornell.
