Neural network modeling in the problem of localization earthquake of Ukraine

An example of using the capabilities of neural network modeling in the problem of localizing the sources of earthquakes in the territory of Ukraine registered by the network of seismic stations of the Institute of Geophysics of the National Academy of Sciences of Ukraine: «Odessa», «Squira», «Poltav...

Повний опис

Збережено в:
Бібліографічні деталі
Дата:2020
Автори: Gerasymenko, O.O., Shumlyanska, L.O., Nadezhka, L.I., Pivovarov, S.P., Ganiev, O.Z., Ostapchuk, N.M., Shipko, N.L.
Формат: Стаття
Мова:Ukrainian
Опубліковано: Subbotin Institute of Geophysics of the NAS of Ukraine 2020
Теми:
Онлайн доступ:https://journals.uran.ua/geofizicheskiy/article/view/201743
Теги: Додати тег
Немає тегів, Будьте першим, хто поставить тег для цього запису!
Назва журналу:Geofizicheskiy Zhurnal

Репозитарії

Geofizicheskiy Zhurnal
Опис
Резюме:An example of using the capabilities of neural network modeling in the problem of localizing the sources of earthquakes in the territory of Ukraine registered by the network of seismic stations of the Institute of Geophysics of the National Academy of Sciences of Ukraine: «Odessa», «Squira», «Poltava», «Nikolaev». According to monitoring data 2007―2019, the authors conducted a continuous accumulation of a seismological database, including for organizing the functioning of a neural network, in the first place, the formation of a training set. Using the capabilities of a powerful statistical analysis tool ― neural networks, the authors built local hodographs of P-, S-waves of the territory of Ukraine, namely, earthquakes of the Ukrainian Shield, the Dnieper-Donets Depression and the Sea of Azov in the magnitude range 2.7―4.8 from the records of four institute seismic stations geophysicists in a form that allows them to be integrated into modern means of digital processing. To clarify the arrival times of the phases of seismic waves within the study region that are poorly visually assessed, the authors use a high level of programmable applications in simulated azimuths to process the signals. The article provides examples of network operation in operational mode. The simulation of the localization problem allows us to accurately design the foci of seismic events in the industrial regions of Ukraine, which confirms the examination of the results by global Jeffries―Bullen hodographs. The examples of localization of earthquakes of 2011, 2013 with magnitudes of 3.9 and 4.6 in the region of the Kryvyi Rih basin provide additional opportunities for analyzing the structural features of the lithosphere, and in the future, real-time assessments of the characteristics of the seismic process to prevent it.