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 (e.g., 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.