Method for constructing commercial area building fire risk prediction grade model

Abstract

A method for constructing a commercial area building fire risk prediction grade model comprises the following steps: S1, obtaining fire historical data, processing the fire historical data, analyzingthe fire historical data, screening out fire characteristics for model training from the fire historical data, obtaining a fire risk prediction grade model, generating a fire characteristic data set from the fire historical data according to the fire characteristics, and dividing the fire characteristic data set into a training set and a test set; S2, constructing a commercial area building fire risk prediction grade model based on a random forest machine learning algorithm according to the training set; S3, calculating the importance of each fire characteristic, and correspondingly adjustinga commercial area building fire risk prediction level model; S4, inputting the test set into the commercial area building fire risk prediction grade model to obtain a fire risk prediction grade, and evaluating the commercial area building fire risk prediction grade model. Building fire disasters in commercial areas can be effectively reduced, and the fire-fighting inspection efficiency of fire-fighting workers in the areas is improved.

Description

technical field [0001] The invention relates to the field of fire risk prediction level models, in particular to a method for constructing a building fire risk prediction level model in a commercial area. Background technique [0002] Fire safety is related to the fundamental interests of the people and is the foundation and guarantee of social development. In recent years, firefighting efforts and investment have been continuously increased, firefighting infrastructure has been continuously improved, and firefighting and rescue and fire prevention and control capabilities have also been continuously improved. However, the total number of fires is still relatively large, and the risk of major fire accidents still exists. With the continuous advancement of urban modernization, today's fires are characterized by complex and diverse causes, increased difficulty in rescue, and increased casualties and property losses. For example, according to the literature [NetEase. Graphical...

AI Technical Summary

Problems solved by technology

In view of the large number of buildings in the current urban commercial area, dense vertical pipe wells, complex functions, dense personnel, and large fire load, it is easy to cause large areas of smoke filling and three-dimensional combustion after a fire, which will bring severe challenges to fire prevention and control and fire fighting and rescue work. To solve the problem, the present invention proposes a fire risk prediction model method created by machine learning. By analyzing and training the characteristics of fire hazards in commercial buildings, the fire risk assessment of commercial buildings can be carried out, and the fire risk level can be divided to help Firefighters eliminate fire hazards in time to reduce the risk of fire

Images