基于深度学习算法的楼宇消防系统设计  

Design of Building Fire Protection System Based on Deep Learning Algorithm

作　　者：乔元健 Qiao Yuanjian

机构地区：[1]齐鲁工业大学(山东省科学院)电气工程与自动化学院

出　　处：《智慧工厂》2020年第4期48-50,共3页Smart Factory

摘　　要：针对传统的消防控制系统性能稳定性差且容易出现误报和漏报的问题,设计了一种人工智能与嵌入式技术相结合的智能消防系统。该系统首先借助BP神经网络可以自学习的特性,将火灾现场产生的一氧化碳浓度信息、烟雾浓度信息以及温度信息作为BP神经网络的输入,然后通过设置网络的激活函数以及误差阈值输出火灾的有火概率、无火概率以及阴燃火概率,并对三种火灾类型设定阈值决策输出有无火灾的发生。仿真结果表明该系统具有更稳定的输出和更好的性能。Aiming at the problems of poor performance stability and prone to false positives and false negatives of traditional fire control systems,an intelligent fire protection system combining artificial intelligence and embedded technology was designed.The system first uses the self-learning characteristics of the BP neural network to take the carbon monoxide concentration information,smoke concentration information,and temperature information generated by the fire scene as the input of the BP neural network,and outputs the fire probability of the fire by setting the network's activation function and error threshold.,The probability of no fire and the probability of smoldering fire,and set thresholds for the three types of fire to determine whether a fire will occur.Simulation results show that the system has more stable output and better performance.

关 键 词：人工智能 BP神经网络 自学习 激活函数 阈值 

分 类 号：TP216[自动化与计算机技术—检测技术与自动化装置]