Multivariate linear regression fire risk evaluation method based on environmental big data

Abstract

The invention relates to a multivariate linear regression fire risk evaluation method based on environmental big data. The method comprises the following steps: dividing landforms, acquiring sample data based on the landforms, dividing environment categories, constructing index weights of the environment categories, establishing a multiple linear regression model, setting a classifier based on initial sample data, and constructing an environment classification-multiple linear regression network; and inputting a region to be evaluated into the environment classification-multiple linear regression network to obtain a fire risk evaluation value. On the basis of big data and machine learning, a multivariate linear regression model in a large environment is constructed through geomorphic features and environment categories, then a multi-layer network is established based on geomorphic features, evaluation is performed based on multivariate attributes, the method can be stably applied to a complex system through training, and then a fire risk evaluation system is perfected. The method is high in evaluation result accuracy, high in application degree and good in transportability.

Description

technical field [0001] The present invention relates to the technical field of data processing systems or methods specially suitable for administration, commerce, finance, management, supervision or forecasting purposes, in particular to a multiple linear regression fire risk assessment method based on environmental big data. Background technique [0002] Fire safety risk assessment refers to the use of scientific and reasonable hazard identification and risk assessment methods, through strict control of accidental harmful factors in the main firefighting work, and the formulation of risk control measures to eliminate hazards and avoid failures caused by inadequate measures and other reasons. The purpose of fire alarm and fire prevention is to avoid vicious accidents such as leakage and fire of toxic, harmful, flammable and explosive media. [0003] Furthermore, according to the results of evaluation and identification, targeted and highly operable preventive control measure...

AI Technical Summary

Problems solved by technology

[0004] Fire risk assessment methods can be divided into qualitative and quantitative methods; qualitative methods are suitable for the safety inspection of social units to identify the most unfavorable fire events, while quantitative methods require a large amount of historical data, through clear assumptions, data and mathematical associations. On-the-spot assessment of the model to determine the actual risk of fire; the former is relatively simple, but the accuracy of the assessment results is controversial, while the latter, although the assessment results are relatively accurate, is actually difficult to be conveniently and quickly used in the smart fire protection system

Images