Grebnev Y.V.1,Yarovoy A.V.2,3
1FSAEI HE Siberian Federal University, Institute of space and information technology
2Chief Directorate of the Emercom of Russia for the Krasnoyarsk Territory
3FSBEE HE Siberian Fire and Rescue Academy EMERCOM of Russia
The paper presents preliminary results of the investigation of the possibility of using neural network algorithms for forecasting natural emergency situations. Natural emergencies are natural forest fires and floods. In work the basic principles of forecasting of emergency situations of natural character with use of neural networks are disassembled. A forecasting model has been developed and an efficiency estimate has been made for application within the framework of the executive authorities in the constituent entities of the Russian Federation.