Intelligent fire-fighting fire identification method and system based on neural network algorithm

Abstract

The invention provides an intelligent fire-fighting fire identification method and system based on a neural network algorithm. The method comprises the following steps: reading environment sensing data, and preprocessing and classifying fire conditions; calculating a data change rate and smoothing the data to obtain to-be-classified data; acquiring original fire feature data; preprocessing the original fire feature data; establishing a fire feature database, smoothing fire feature data to construct a perception model, and establishing a feature project; reprocessing the sample input into the combustion stage prediction model; and classifying the environment perception data and the fire feature data after smoothing processing through the model, and judging the type and the combustion stage of the fire. The system comprises an intelligent fire-fighting cloud platform, an administrative management platform, a system access module, a maintenance module and a monitoring center. According to the invention, the fire can be found earlier, the response time is shortened, the fire type is predicted, targeted fire fighting is carried out, data of multiple sensors are cooperatively processed, and the false alarm probability is reduced through comprehensive research and judgment.

Description

technical field [0001] The invention relates to a method and system for intelligent fire identification based on a neural network algorithm, and belongs to the technical field of fire prevention and control. Background technique [0002] With social development and technological progress, smart fire protection systems have gradually become popular in various units. However, there are still many deficiencies in combination with actual use conditions. For example, although one or more detectors are combined to judge whether a fire has occurred, the threshold method used in the judgment method leads to a high false alarm rate of the system; the inability to judge the development of the fire leads to longer response time and thus Delaying the fire or taking excessive fire-fighting measures will aggravate the losses; more importantly, it is impossible to judge the type of fire. When the type of fire is not clear, untargeted firefighting is likely to cause greater losses and even ...

AI Technical Summary

Problems solved by technology

However, there are still many deficiencies in combination with actual use conditions. For example, although one or more detectors are combined to judge whether a fire has occurred, the threshold method used in the judgment method leads to a high false alarm rate of the system; the inability to judge the development of the fire leads to longer response time and thus Delaying the fire or taking excessive fire-fighting measures aggravates the loss; more importantly, it is impossible to judge the type of fire. When the type of fire is not clear, untargeted firefighting is likely to cause greater losses and even endanger the lives of firefighters.

Images