Mathematical intellectual algorithm for determining the fire hazard parameters of oil and oil refining products

To improve the accuracy of the obtained values when predicting the fire hazardous properties of organic substances, it was decided to modify the existing algorithm, supplementing it with deep learning technology. Verification of the data was carried out based on some reference and practical data. It was found that the values obtained in the course of computer simulation provide a satisfactory error, which means that the data obtained can be used in practice.