An ergonomic system for forecasting forest fires and the peculiarities of their spread in the conditions of man-made development of the world

Abstract

This paper deals with the development of an ergonomic system for forecasting wild fires and the features of their spread, which is the object of research. The objective of the study is to make a full-fledged system for forecasting and determining the features of wild fire spread. To achieve this objective, the following tasks are solved: to carry out a comparative analysis of existing systems for forecasting the spread of wild fires in order to identify the disadvantages and advantages of these systems; based on the analysis of existing systems, put forward requirements for the ergonomic system being developed; to develop the structure of a neural network capable of forecasting the spread of wild fires in accordance with the requirements; generate a data set for training and testing the developed neural network and test its operation. To solve the tasks, techniques of analyzing large data sets are used, for example, Data Mining, regression analysis, machine learning, data processing methods such as filtering, augmentation, and others. The practical relevance of this paper lies in the fact that the developed system can be used as an auxiliary system at environmental protection enterprises.