Spartan UK pilots AI technology to cut emissions and energy use
Spartan UK pilots AI technology to cut emissions and energy use
The AI at the British Spartan plant promises to reduce CO₂ emissions by up to 10% when heating steel slabs. Image: Metinvest
Spartan UK has launched a pilot project with industrial AI start-up Deep.Meta to reduce emissions and improve energy efficiency at its Newcastle upon Tyne plant.
The project deploys Deep.Optimiser-PhyX, an AI-powered digital twin that uses live furnace data to optimise heating cycles and production scheduling in real time. By making the process more efficient and stable, the system is designed to lower both CO₂ intensity and energy consumption. Spartan UK is a subsidiary of the Ukrainian steel group Mitinvest and operates a heavy plate rolling mill in Newcastle-upon-Tyne with a production capacity of up to 200,000 tonnes of heavy plate per year. Deep.Meta’s modelling for Spartan UK’s operations indicates the technology could deliver up to 10% reductions in CO₂ intensity, subject to validation in live production. The pilot focuses on using energy more intelligently, improving process stability and reducing variability – all critical factors in a market where energy and carbon costs make up a significant share of overall production costs. “Deep.Meta is a trusted partner, and we are piloting the Deep.Optimiser solution, because of the rising costs of energy and carbon,” said Michael Brierley, CEO of Spartan UK. “Increasing the efficiency of production is of high importance as energy costs form a significant part of our cost structure.” The collaboration is another important step of Metinvest Group towards decarbonisation and digital innovation across its assets. By testing advanced AI solutions at Spartan UK, the Group aims to scale successful approaches across its wider steel portfolio in the future. According to its own statements, British start-up Deep.Meta has developed Deep.Optimiser-PhyX, an AI-supported digital twin, an intelligent digital replica of the steel production process that combines physics and machine learning to optimise furnace operation, thereby simulating years of production in just a few hours. Source: Metinvest/Deep.Meta