AI-Guided Predictive Carbonation Mapping for Ultra-Low-Emission Cement Manufacturing Facilities
Annals of Civil Engineering and Management · 16.06.2026 · 03 (02) · 01-14
DOI: 10.33140/acem.03.02.06 ↗Kurz gefasst
Fachartikel in der Zeitschrift Annals of Civil Engineering and Management, online erschienen am 16.06.2026. Verfasst von Chinenye Elizabeth Onumadu.
Abstract (Originalsprache)
Carbonation curing can transform cement manufacturing from a CO₂ source to a CO₂ sink, but its industrial adoption is limited by unpredictable outcomes arising from feedstock variability, fluctuating flue gas compositions, and inhomogeneous chamber conditions. Current control strategies rely on static lookup tables that fail to account for realtime process dynamics, leading to underperformance or batch rejection. Here we develop and validate an AI-guided predictive carbonation mapping system that forecasts four key performance indicators simultaneously: CO₂ absorbed (wt%), carbonation depth (mm), 28-day compressive strength (MPa), and porosity reduction (%). Using a dataset of 10,800 production batches from a pilot carbonation chamber operating over 12 months (24 input features including raw meal XRF, chamber temperature, relative humidity, pCO₂, and curing duration), we trained a hybrid architecture combining a physics-informed neural network (PINN) embedding a simplified diffusion-reaction PDE with a gradientboosted residual corrector (XGBoost). The hybrid model achieves hold-out test R² values of 0.94 (CO₂ uptake), 0.91 (strength), 0.89 (carbonation depth), and 0.87 (porosity reduction), significantly outperforming standalone machine learning baselines (best baseline R² = 0.85). SHAP analysis reveals pCO₂ (33% contribution), curing temperature (22%), and water-to-cement ratio (18%) as dominant predictors, with reactive MgO content (9%) unexpectedly influential. The model generates 2D spatial carbonation maps revealing heterogeneous uptake (up to 20% variation across chamber corners), enabling targeted process redesign. Real-time inference runs at 47 ms per batch with uncertainty bounds (90% CI width ±1.8% CO₂ uptake), sufficient for closed-loop control. Simulated deployment reduces batch failure rates from 18% (conventional lookup-table control) to 6% while increasing average CO₂ uptake by 15% – equivalent to 75 kt additional CO₂ sequestered annually per 0.5 Mt plant. Key limitations include single-plant training chemistry and batch-averaged inputs; future work requires transfer learning across geologies and per-sample near-infrared sensors. This work demonstrates that hybrid AI, combining physical constraints with data-driven flexibility, can transform carbonation curing from an art to a predictable, optimizable industrial process.
Zitieren
APA
Onumadu, C. E. (2026). AI-Guided Predictive Carbonation Mapping for Ultra-Low-Emission Cement Manufacturing Facilities. Annals of Civil Engineering and Management, 03(02), 01-14. https://doi.org/10.33140/acem.03.02.06
Vancouver
Onumadu CE. AI-Guided Predictive Carbonation Mapping for Ultra-Low-Emission Cement Manufacturing Facilities. Annals of Civil Engineering and Management. 2026;03(02):01-14. doi:10.33140/acem.03.02.06
BibTeX
@article{Onumadu2026,
author = {Onumadu, Chinenye Elizabeth},
title = {AI-Guided Predictive Carbonation Mapping for Ultra-Low-Emission Cement Manufacturing Facilities},
journal = {Annals of Civil Engineering and Management},
year = {2026},
volume = {03},
number = {02},
pages = {01--14},
doi = {10.33140/acem.03.02.06}
}
EUVIMED