Level of fire danger of the local territory

Abstract

The object of the study is the fire risk of the local area. The problem to be solved is to take into account most of the significant parameters in the territorial placement of fire-rescue units of different functional capacities. As part of the solution to this problem, a technique for assessing the fire risk of a large-scale local area has been developed. The methodology is focused on local territories of a large area with a low population density. A special feature of the proposed method is the differentiated fire risk assessment of each point of the surface plane. For such an assessment, the parameters that are decisive from the point of view of impact on the fire hazard are analyzed and structured. The specified factors include the spatial distribution of population density and buildings, the transport and communication network, the spatial distribution of the density and type of vegetation, and statistical data on landscape fires. The use of existing geo-informational resources in real time is foreseen. A new approach of ranking the fire risk of the elementary plane of the territory in accordance with the necessary number of resources of rescue units to ensure the appropriate level of safety is proposed. Neural network data processing methods were used to compare local area parameters with fire risk ranks. A neural network capable of comparing the fire risk of the territory with its parameters was obtained. The functionality of the developed methodology was tested and the fire risk levels of an arbitrary area were graded with an average degree of correlation of 0.97. The proposed method allows for assessment and correction of the state of provision of local territories with civil protection resources. The developed methodology is especially relevant when creating new fire and rescue units of territorial communities.