Today, we are proud to be launching ClimateHack.AI 2023-24: an international grand challenge with a ยฃ20k prize pool bringing together students at 18 universities across the UK, the US and Canada to help tackle climate change with AI. ๐
Students will be competing to develop state-of-the-art machine learning methods for more accurately forecasting near-term site-level solar photovoltaic power generation using 600 gigabytes of satellite imagery, weather forecasts and air quality forecasts.
The competition has been organised in collaboration with Open Climate Fix, a non-profit lab focused on rapidly cutting greenhouse gas emissions, and is supported this year by Newcross Healthcare, University College London, PGIM Real Estate via their RealAssetX innovation lab, and Climate X.
Open Climate Fix estimate that more accurate solar power forecasting techniques could cut carbon emissions by up to 100,000 tonnes per year in the UK alone โ and by up to 100 million tonnes per year if deployed worldwide โ by allowing grid operators to reduce their use of non-renewable sources to account for the variability of solar power generation.
Students have until Friday 8th March 2024 to make their final submission in the qualifying around, and the top three students from each university will be invited to develop a joint solution and present their work to a panel of expert judges at in-person finals simultaneously taking place at UCL and Harvard University on Friday 5th and Saturday 6th April 2024.
The winning teams will be awarded ยฃ10,000 (~$12.7k US), ยฃ6,000 (~$7.6k US) and ยฃ4,000 (~$5k US), respectively, and their research contributions have the potential to be deployed by Open Climate Fix to produce more accurate solar power forecasts for the UKโs National Grid Electricity System Operator.
When the competition ran for the first time as ClimateHack.AI 2022 between January and March 2022, the winning model was the most accurate model for predicting cloud movements that Open Climate Fix had seen to date.
Students interested in taking part should visit climatehack.ai, check out the competition page and join our community Discord server.
Students may take part from the following universities: Bristol, Caltech, Carnegie Mellon University, Harvard University, Illinois Urbana-Champaign, Imperial College London, King's College London, Manchester, Michigan, Oxford, Princeton, St Andrews, Toronto, UC Berkeley, UCLA, University College London (UCL), Warwick and Waterloo.