Abstract
Compartmental models are widely used to simplify the analysis of complex fluid dynamics systems, yet subjective compartment definitions and computational constraints often limit their applicability. The CompArt algorithm introduces an AI-driven framework that automates compartmentalization in Computational Fluid Dynamics (CFD) simulations, optimizing both accuracy and efficiency. By leveraging unsupervised clustering techniques such as Agglomerative Clustering, CompArt identifies coherent flow regions based on velocity and turbulent kinetic energy dissipation rate, ensuring a data-driven, physically consistent segmentation. The methodology integrates a connectivity-based clustering strategy, where compartments are dynamically optimized using the Silhouette score and adjacency matrix. This approach enables the reduction of high-resolution 3D CFD simulations into a network of interconnected sub-systems, significantly lowering computational costs while preserving system heterogeneity. The framework was tested on turbulent stirred tank simulations at 200 and 400 RPM, demonstrating its ability to capture both large-scale flow structures and fine-scale turbulence features. At higher RPM, increased mixing leads to a more fragmented dissipation pattern, which the model successfully adapts to. With scalability up to industrial scale meshes (~1M cells) and full automation of hyperparameter selection, CompArt enhances the applicability of CFD compartmental models in crystallization, multiphase flow, and process optimization. By bridging AI-driven clustering with physical modelling, this novel framework streamlines CFD simulations, making real-time industrial applications feasible without sacrificing predictive accuracy.