Yaroslav V. Grebnev1,2, Alexander K. Moskalev1, Dinara I. Shagidulina1, Alexander V. Antonov2,3, Dmitry V. Ivanov3
1Siberian Federal University, Krasnoyarsk, Russia
2Russian Emergencies Ministry for the Krasnoyarsk Territory, Krasnoyarsk, Russia
3Siberian Fire and Rescue Academy EMERCOM of Russia, Zheleznogorsk, Russia
The active development of the Arctic, which has become a trend of the last decade, has become a big challenge for the fragile ecosystem of the Arctic and requires proper control over technological processes at enterprises, since over the past two years a number of accidents have occurred in the Arctic associated with spilling oil products on the soil and water bodies located in these areas. Applicable measures to ensure technosphere safety are currently insufficient, as evidenced by the increasing number of emergencies and the use of modern technologies for monitoring and forecasting risks is required in order to adjust action plans in the event of an emergency and calculate the required forces and means. Modeling of oil and oil products spills is carried out in order to carry out preventive measures to prevent emergency situations. Methods for estimating the area of oil spills currently used, especially in the Arctic zone, have a number of limitations. In this work, the modeling of the process of bottling oil products was carried out in order to calculate the distribution of the concentration of pollutants and neural network prediction of the area of pollution up to the moment of its liquidation. To simulate an emergency, there was the PHOENICS software product and a neural network forecasting technique using the Scikit-Learn library in the Python programming language. The simulation results were correlated with the data obtained in the analysis of an accident that occurred due to a pipeline depressurization when pumping oil from a vessel to a tank farm on the Khatanga River in the Arctic zone of the Krasnoyarsk Territory.
Open » // Ссылка для цитирования этой статьи