Search academic texts

Information × Registration Number 0224U032323, 0124U003876 , R & D reports Title Development of a methodology for calculating the microkinetic models of chemical synthesis using physics-guided machine learning methods: second stage popup.stage_title Пошук оптимальних гіперпараметрів штучної нейронної мережі для підвищення точності розрахунку частки вакантних каталітичних центрів у мікрокінетичній моделі синтезу Фішера-Тропша. Дослідження області значень похідної частки вакантних каталітичних центрів у мікрокінетичній моделі синтезу Фішера-Тропша по вхідних параметрах. Head Druchok Maksym Yu., Кандидат фізико-математичних наук Registration Date 10-09-2024 Organization Institute of Condensed Matter Physics of the National Academy of Sciences of Ukraine popup.description2 This work is a continuation of the previous project, in which an approach to calculating the microkinetic model of Fischer-Tropsch synthesis using a physics-informed neural network approach was proposed. This approach made it possible to generalize the equations for calculating the fraction of vacant catalytic sites for a wide class of catalytic materials. As part of this work, several neural network models were obtained for accurate and quick calculation of the fraction of vacant catalytic sites and the derivatives of this quantity on the input parameters. The ability to quickly calculate the value of derivatives allows one to expand the application of models. The calculation of errors relative to the classical methods of solving equations was carried out, as well as the estimation of the gain in time when applying the obtained models.  Product Description popup.authors Demchuk Taras V. popup.nrat_date 2024-10-01 Close