AI for Decarbonisation Innovation Programme: Stream 2 successful projects
Summary
This will lead to better forecasts of grid load allowing improved scheduling of generation assets, reducing network congestion and renewable energy curtailment. This project aims to develop an additional module for cases with highly seasonal power demand and extreme events (e.g., weather conditions, energy pricing etc), and optimising hydrogen delivery for multiple off-takers. The model will be based on a facility operated by RWE, a partner in this project, which has an installed renewable capacity of around 50 and will use hydrogen to replace for an off-gas-grid town.machine learning for solar forecasting for improved grid management and decarbonisationDescription: This project will create artificial intelligence models combining both sky images and numerical weather data for forecasting solar production (for very short-term, short-term and medium term) to significantly improve the prediction accuracy of meteorological parameters, reducing the power mismatch caused by solar forecast errors.Description: This project aims to create a novel neuromorphic computing unit which emulates the neural structure of the human brain and consumes a fraction of the power of conventional hardware. This project will take the first steps towards achieving a more efficient use of energy in computing by testing the principles underlying the proposed device and designing the microarchitecture ready for manufacturing.Adaptive for total substitution rate of alternative fuels in cement manufacturingDescription: The company has developed a platform which uses artificial intelligence / machine learning to optimise production control and delivers a reduction in fuel consumption in the manufacturing of cement. The size, shape, and moisture content of the concrete aggregate influence the quantity of binder, water and additives required to create a high-quality mix.