Carbon Re is an industrial artificial intelligence (AI) company that focuses on decarbonizing energy-intensive industries such as cement and steel. Carbon Re’s software solution, Delta Zero, merges continuous improvement techniques with cutting-edge AI to reduce greenhouse gas emissions in manufacturing and save a substantial amount of electricity, fuel, and carbon costs per plant per year. The company’s AI platform can deliver £2 million in fuel cost savings and 50 kilotons of CO₂ savings per plant.
Challenges: steel and cement CO₂ emissions
The decarbonization of energy-intensive industries, such as cement and steel, is essential to the global effort to combat climate change. These industries are significant contributors to greenhouse gas emissions, and their decarbonization is necessary to meet the target of limiting global warming to 1.5 ºC above pre-industrial levels. However, decarbonizing these industries presents significant challenges due to their complex processes and the high energy demands involved.
Cement production is responsible for approximately 8% of global CO₂ emissions. The process involves heating limestone and other raw materials to high temperatures in a kiln to produce clinker, which is then ground into cement. This process requires large amounts of energy, mainly in the form of fossil fuels like coal and natural gas. Therefore, decarbonizing the cement industry requires the adoption of new technologies and the use of alternative fuels, such as biomass and waste-derived fuels, that emit fewer greenhouse gasses.
One of the main challenges of decarbonizing the cement industry is the lack of commercially viable alternatives to the traditional manufacturing process. New technologies such as carbon capture, utilization, and storage (CCUS) and alternative binders that reduce the need for clinker production are still in the development stage and are not yet widely available. Furthermore, the high capital costs and the lack of policy incentives make it difficult for cement manufacturers to invest in new technologies.
Another energy-intensive industry that poses challenges for decarbonization is steel production. The production of steel accounts for approximately 7% of global CO₂ emissions. The process involves the reduction of iron ore in a blast furnace using coke as a reducing agent. This process emits large amounts of CO₂, and the energy required is mainly derived from coal and natural gas. Decarbonizing the steel industry requires the adoption of new technologies such as hydrogen-based direct reduction and electric arc furnaces that use renewable energy sources.
The main challenge of decarbonizing the steel industry is the lack of low-carbon technologies that are commercially available at scale. The adoption of new technologies such as direct reduction requires significant investment in research and development, and the transition to new technologies can be challenging due to the high capital costs and the need to retrofit existing facilities.
Moreover, the transition to low-carbon steel production requires a significant shift in the entire value chain, including the sourcing of raw materials, the development of new infrastructure, and the training of a new workforce. This shift requires a coordinated effort among policymakers, industry players, and other stakeholders.
Carbon Re AI platform
Carbon Re uses cutting-edge AI technology to decarbonize energy-intensive industries of cement and steel. The company is building the world’s most advanced AI platform called Delta Zero, which is designed to achieve operational efficiencies, reduce costs, and reduce annual carbon emissions by gigatons.
How can AI help decarbonize?
AI technology can now generate a ‘digital twin’ of an individual plant by analyzing the sensor data collected by Distributed Control Systems on a minute-by-minute basis. This creates a high-resolution digital model that captures the properties of a given kiln or furnace and delivers an accurate representation of the real world performance in that specific plant. This digital twin provides a platform for AI ‘agents’ to learn complex process control in a sophisticated simulation.
By applying digital twins and reinforcement learning to fuel use in material production and processing, AI agents are trained to explore and find the optimal parameters of the production system concerning one or more goals (e.g., minimal fuel cost, maximal throughput), which are then applied to the real plant.
Deep reinforcement learning is the process by which AI learns what actions to take in different contexts through repeated interaction with an “environment”. In this context, learning is the gradual improvement of a neural network, which can be viewed as the AI’s brain. The AI’s choice of action is influenced by the ‘reward function’ which can be tailored to induce the desired behavior in the problem solution.
For example, when optimizing performance in a cement plant, the choice of reward function could incorporate a carbon cost to encourage decisions that reduce emissions while maintaining production. It can also include specific constraints such as NOₓ emissions or torque limits to ensure safe operation.
Carbon Re claims that its Delta Zero AI platform has the potential to reduce CO₂ emissions by more than 50 kilotons per plant. By using AI to reduce carbon emissions, Carbon Re is connecting the biggest challenge of our time – climate change – with the biggest opportunity – advances in AI.
Artificial intelligence market
Many industries, including finance, healthcare, automotive manufacturing, and technology, have been transformed by advances in artificial intelligence, which have introduced new capabilities and opportunities for innovation. The global artificial intelligence market was valued at USD 136.55 billion in 2022 and is projected to expand at a compound annual growth rate (CAGR) of 37.3% from 2023 to 2030.
However, adoption in foundational industries such as cement, steel, and glass has been slower due to a lack of tools specifically tailored to the needs of plant operators. Carbon Re aims to develop and deploy its Delta Zero AI platform to help these energy-intensive industries reduce CO₂ emissions.
Carbon Re Products
Carbon Re has developed the Delta Zero AI platform, which is currently used in high-polluting industries such as cement, steel, and glass. The platform models the production environment of each plant and uses machine learning and AI to achieve efficiencies in operation, identifying optimal processes for the lowest possible carbon output and fuel use, bringing down operational costs and carbon emissions in the cement industry.
For example, Delta Zero Cement AI platform simulates the chemical and physical processes in a plant, enabling AI agents to find solutions tailored to each plant.
The AI software analyses feed rates, sensor data and control parameters to provide clear quantified recommendations to reduce the mass of CO₂ emitted per useful heating value (kgCO₂/UHV). This enables lower fuel costs and lower emissions. The AI platform has the potential to cut fuel use by 10% and fuel-derived CO₂ emissions by 20%, equivalent to more than 50 kilotons of annual CO₂ emissions per cement kiln.
Carbon Re Funding
Carbon Re Investors
Carbon Re is funded by 7 investors, including UCL Technology Fund, University of Cambridge Enterprise, Clean Growth Fund, Planet A Ventures, UCLB, Parkwalk Advisors, and Blue Impact Ventures. Planet A Ventures and UCL Technology Fund are the most recent investors.
Carbon Re Founder
Carbon Re CEO
Sherif Elsayed-Ali is CEO.