DEVELOPMENT OF FIRE PREDICTION AND PREVENTION DIGITAL SYSTEM ALGORITHMS

This research work is devoted to the development of algorithms for a digital system aimed at early detection, prediction and prevention of fire hazards. In the work, the process of fire hazard assessment is modeled on the basis of modern information technologies and artificial intelligence tools. The main focus is on collecting data in real time, analyzing it and creating algorithms that determine the level of danger. In the process of research, methods of data cleaning, normalization and determination of correlation between variables were used to process multidimensional data streams obtained from various sensors (temperature, smoke, gas concentration and humidity meters). The effectiveness of statistical models and machine learning algorithms (logistic regression, random forests, neural networks) in predicting the probability of fire occurrence was compared. As a result, the optimal algorithm was selected and tested in real conditions. The results of the research will serve to increase fire safety, reduce material and human losses, and ensure safety in industrial, transport, and residential infrastructure based on digital management.

Hali tarjima qilinmagan