Отрывок: In order to reduce the error incursion and improve the quality of prediction, it was proposed to include in the loss function the results of predicting the neural network several steps ahead in the recurrent mode. As a result of such training, the quality of prediction has improved significantly (Table 1), and the graphs of neural network prediction ...
Полная запись метаданных
Поле DC Значение Язык
dc.contributor.authorMalsagov M. YU.ru
dc.contributor.authorMikhalchenko E. V.ru
dc.contributor.authorKarandashev I. M.ru
dc.contributor.authorNikitin V. F.ru
dc.coverage.spatialхимическая кинетикаru
dc.coverage.spatialмодель горенияru
dc.coverage.spatialartificial neural networksru
dc.coverage.spatialchemical kineticsru
dc.coverage.spatialmodel of combustionru
dc.coverage.spatialhydrogenru
dc.coverage.spatialoxygenru
dc.coverage.spatialводородru
dc.coverage.spatialкислородru
dc.coverage.spatialискусственные нейронные сетиru
dc.creatorMalsagov M. YU., Mikhalchenko E. V., Karandashev I. M., Nikitin V. F.ru
dc.date.accessioned2022-10-06 14:29:44-
dc.date.available2022-10-06 14:29:44-
dc.date.issued2022ru
dc.identifierRU\НТБ СГАУ\488523ru
dc.identifier.citationModeling the dynamics of hydrogen combustion using the neural network UNET / M. YU. Malsagov, E. V. Mikhalchenko, I. M. Karandashev, V. F. Nikitin // International Conference on Physics and Chemistry of Combustion and in Extreme Environments (Samara, Russia, 12-16 July 2022) / V. N. Azyazov, A. M. Mayorova. - Samara : Publishing OOO “Insoma-Press”, 2022. - P. 38.ru
dc.identifier.urihttp://repo.ssau.ru/handle/International-Conference/Modeling-the-dynamics-of-hydrogen-combustion-using-the-neural-network-UNET-98905-
dc.language.isoengru
dc.sourceInternational Conference on Physics and Chemistry of Combustion and in Extreme Environments (Samara, Russia, 12-16 July 2022)ru
dc.titleModeling the dynamics of hydrogen combustion using the neural network UNETru
dc.typeTextru
dc.textpartIn order to reduce the error incursion and improve the quality of prediction, it was proposed to include in the loss function the results of predicting the neural network several steps ahead in the recurrent mode. As a result of such training, the quality of prediction has improved significantly (Table 1), and the graphs of neural network prediction ...-
Располагается в коллекциях: International Conference on Combustion Physics and Chemistry

Файлы этого ресурса:
Файл Размер Формат  
978-5-4317-0481-9_2022-38.pdf512.58 kBAdobe PDFПросмотреть/Открыть
Показать базовое описание ресурса Просмотр статистики
Поделиться:



Все ресурсы в архиве электронных ресурсов защищены авторским правом, все права сохранены.