基于SVR的高层建筑复合式火灾预警系统设计

Design of Composite Fire Pre-warning System for High-rise Buildings Based on the SVR

针对传统单一功能火灾探测预警系统可靠性差的缺陷,本文设计了一种高层建筑感烟-感温复合式火灾智能预警系统。该系统根据高层建筑火灾预警的非线性、历史数据少及影响因素多等特征,为了提高预警系统的准确性,引入支持向量机回归(SVR)算法,建立了基于SVR模型的高层建筑复合式火灾预警系统。最后以木材火和普通火燃烧标准历史数据为例,通过Matlab仿真模拟,对构建的SVR预警系统进行实证分析,从而验证了所建立的火灾预警系统的可行性和可靠性。该研究成果可为高层建筑火灾准确预警提供一种可靠的决策支持系统。

In view of the poor reliability of traditional single-function fire pre-warning system,this study dc signs and proposes a new composite fire pre warning system for the high-rise buildings. At the same time, considering the characteristics of the fire pre-warning system for high-rise buildings, such as being nonlin ear,few historical data,many influence factors and so on. In order to improve the accuracy of the prc warn ing system,the study uses the support vector regression（SVR） algorithm,and builds a composite fire pro warning system for high-rise buildings based on the SVR. Then it takes the historical data of wood burning and the ordinary fire as examples to make empirical analysis for the new system through the Matlab simulation. The research results provide a reliable decision support system for the accurate pre-warning of high rise building fire.