Search academic texts

Information × Registration Number 0224U031521, 0124U002673 , R & D reports Title Development of a methodology for calculating the microkinetic models of chemical synthesis using physics-guided machine learning methods popup.stage_title Head Druchok Maksym Yu., Кандидат фізико-математичних наук Registration Date 08-05-2024 Organization Institute of Condensed Matter Physics of the National Academy of Sciences of Ukraine popup.description2  The primary outcome of the research is a neural network that has been trained to forecast a functional resolution of equations within the framework of Fischer-Tropsch catalytic synthesis. Fischer-Tropsch synthesis denotes a catalytic process converting syngas (composed of CO and H2) into a petroleum-like product consisting of higher-order hydrocarbons. The input parameters for such a model (and neural network) comprise the thermodynamic characteristics of the chemical reactor and the kinetic constants specific to a given catalytic material. In contrast to previously available solutions, the proposed method enables the generalization of the microkinetic model's solution for various catalytic materials without necessitating additional retraining. An advantage of this approach over classical numerical methods lies in its demonstrated significant enhancement in calculation speed. Furthermore, this method facilitates the computation of reaction rates for each distinct reaction product (including n-paraffins and 1-olefins with varying carbon chain lengths). As part of the project, this approach was validated through comparison with classical methods and experimental data. Product Description popup.authors Demchuk Taras V. popup.nrat_date 2024-05-08 Close