FlowMat: a toolbox for modeling flow reactors using physics-based and machine learning approaches for modular simulation, parameter identification, and reactor optimization

This paper introduces a versatile, open-source MATLAB/Simulink toolbox for modeling and optimizing flow reactors. The toolbox features a modular architecture and an intuitive drag-and-drop interface, supporting a range of different modeling approaches, including physics-based, data-driven, and hybrid models such as physics-informed neural networks. We detail the toolbox's implementation and demonstrate its capabilities through real-world applications, including the simulation of flow reactors, identification of reaction parameters using experimental data (<i>e.g.</i>, transient data), and optimization of reactor operating points and configurations. Experimental validations illustrate the practical applicability and effectiveness of the toolbox, making it a valuable resource for researchers and engineers in the field with the potential of reducing the cost and time required for parameter determination and reactor optimization.