OptiCap AI will deliver the first open-architecture, physics-based CO2 capture digital twin that uses AI to calibrate to process data and optimize capture efficiency. Combining rigorous first-principles models with physically constrained AI can cut energy use and cost of capture by at least 1015%, while de-risking scale-up across emission sources and process concepts. By natively embedding dynamic simulation, uncertainty propagation and sensitivity analysis, OptiCap AI will let operators see exactly which parameters dominate cost and drive risk, guiding faster, evidence-based engineering and investment decisions.
OptiCap AI advances carbon-capture simulation by concentrating on two key pillars: 1) Physically-constrained AI calibration and optimization using (grey-box) parameters that honour mass- and energy balances when fitting the twin to plant data, cutting months of manual tuning to hours. 2) An open, high-fidelity digital twin backbone in Julia/SciML, the next-generation technical and scientific computing language, which integrates natively with Q-props electrolyte thermodynamics for modelling of any solvent and flue gas composition accurately and fast. Together these pillars tackle the long-standing bottlenecks in CO2 capture modelling of slow model calibration, narrow solvent coverage and opaque black-box twins. To demonstrate this, OptiCap AI will be applied on three real-world capture processes (power generation, biogas production and photo-electrochemical DAC), for a world-first process-level solvent screening demonstration, and for systems analyses, showing how uncertainties and sensitivities can be used for next-generation technoeconomic and life-cycle analyses.
10-15 % energy savings for post-combustion and biogas production use cases, corresponding to unlocking ca. 1 MtCO2/yr of additional economically viable capture across Denmark’s 10.5 Mt/yr point-source potential. De-risked scale-up, faster troubleshooting, and ability to model novel solvents and process concepts, such as photo-electrochemical DAC. An MIT-licensed Julia core will be released for public beta in 2029, allowing Danish emitters and EPCs to test OptiCap AI.
