Font Size: a A A

The Research And Practice Of Fire Intelligent Analysis Based On BP Neural Network Algorithm

Posted on:2017-09-23Degree:MasterType:Thesis
Country:ChinaCandidate:S L ZhangFull Text:PDF
GTID:2322330482976792Subject:Computer technology
Abstract/Summary:PDF Full Text Request
According to the fire code,more than a certain area of buildings must be equipped with automatic fire alarm system,some large unit will have multiple buildings,in order to avoid duty personnel too much,often multiple automatic fire alarm system are integrated together,this also brought new problems: fire alarm events frequently,basic it is false alarm.How to let the personnel on duty better judgment misstatement becomes very important.Therefore it is necessary to conduct fire incident intelligent analysis.Based on the analysis of the false alarm of fire automatic alarm system,We introduce the BP neural network to solve the above difficulties,The main works in this paper are as follows:Firstly,according to the analysis of the intelligent fire alarm information related theory,mainly by the factor of fuzzy clustering analysis of time and space,the color of the video monitoring intelligent identification model,dynamic texture spectrum analysis theory to the analysis of the intelligent fire alarm information such as the establishment of the model;Secondly,based on the model of fire alarm information intelligent analysis,through the BP neural network algorithm convergence time improvement,and the improved BP neural network algorithm for intelligent analysis and video factors of space and time for training,finally got to the fire alarm information intelligent analysis algorithm.Thirdly,In order to better reflect the application of the above algorithm,this paper intelligent fire alarm information analysis algorithms used in practice,and the operating results of intelligent analysis algorithms fire alarm network platform for analysis.Through theoretical analysis and practice it solved the problem of false positives fire.But also it can be used in security intelligent evacuation system.But there are deficiencies,such as:alarm events are not frequent or less building where the effect is not very obvious problems,it can also combine extensive analysis alarm data,etc.to better address correlation.
Keywords/Search Tags:Automatic fire alarm, BP neural network, intelligent analysis, Time and space factor, accidental alarm
PDF Full Text Request
Related items