基于多传感器融合技术的弹库火灾探测

提出了一种应用多传感器数据融合技术于舰载导弹弹库的火灾探测的方法.该方法采用多传感器探测火灾的多种参量,利用模糊推理技术、神经网络技术和事例推理技术的复合推理方法判别弹库火灾状态.     

基于多传感器数据融合技术的远程火灾预警系统

为提高火灾预警系统的集成性以及高效传输预警结果,设计了一种基于多传感器数据融合技术的远程火灾预警系统。在系统硬件设计方案中,重点介绍了传感器的信号调理电路、传感器和SIM300模块与单片机系统的连接方式。在系统软件设计方案中,重点讨论了硬件系统自检方法和多传感器数据处理方法。     

基于无线传感器网络的火灾探测方法研究

针对火灾发生时的智能探测系统和方法进行研究。使用无线传感网络技术实现火灾探测传感单元的组网,提高传感单元的检测距离和可靠性。研究基于D-S证据理论的多传感器信息融合的火灾探测判别方法,将多个传感器的检测数据进行综合考虑,得出火灾发生的概率,提高火灾探测准确率。研究结果表明,使用多传感器融合方法后,对温度、CO或是烟雾传感器的数据进行全面考虑,判别火灾发生的概率和结果准确率较好,即使存在烟雾干扰信号的情况,可以有效屏蔽干扰信号,准确判断火灾发生。     

浅谈基于视频的高大空间建筑火灾探测与扑救技术

基于传统的高大空间建筑火灾探测与扑救技术是有局限性的,本文结合该技术在实践中的应用趋势,提出了基于视频的高大空间建筑火灾探测与扑救技术的设计方法。     

受限空间火灾探测数据融合算法研究

为了实现对受限空间火灾发生情况多传感器自动探测信号处理,建立了基于CO浓度、烟气颗粒浓度、红外视频图像等多传感器探测的火灾探测数据融合分析系统.在受限空间内搭建基于多传感器火灾探测的数据融合分析实验平台.根据火灾红外视频图像的燃烧区域面积及圆形度等特征参数介绍了红外图像识别算法.在三种火灾探测方式特征参数的基础上提出了...     

孤立森林算法的灭火救援疏散路径规划方法

针对建筑火灾中人员疏散路径规划问题,提出基于孤立森林算法的灭火救援疏散路径规划的方法。运用布置在火灾现场的无线传感器网络采集火灾环境信息,构建火灾数据样本,随机分割并训练火灾数据样本,创建多个孤立二叉树组建孤立森林,识别火灾异常数据,获得着火点及障碍物位置,并以栅格法构建火灾救援环境动态地图为基础,通过更新位置节点当量距离、信息素浓度以及信息素挥发因子的改进蚁群算法,构建救援疏散路径组合优化模型,规划出最佳灭火救援疏散路径。测试结果表明：该方法可准确检测火灾中的着火点位置,可在多起点、多终点的救援疏散路径规划中更好地避开着火点和障碍物,快速、合理地规划出最佳灭火救援疏散路径。     

光纤传感器感温性能分析修正

摘 要：为提升光纤传感器在建筑火灾预警中的适用性,本研究针对单模、多模普通光纤开展基于常用实际火灾升温曲线的高温试验以及感温性能分析,进而遴选感温性能更适用于建筑火灾条件下的传感器。同时,基于试验数据建立升温误差修正模型,以提升测温精度。结果表明：在测温稳定性方面,多模光纤传感器远优于单模光纤传感器,多模光纤传感器感温性能在建筑火灾预警中适用性更好,且更具有修正价值,初步建立的升温误差修正模型使其测温平均绝对误差降低至13.12 ℃。利用分布式光纤进行高精度建筑火灾地图绘制,成为火灾监测预警、疏散救援以及结构倒塌预测的技术关键。     

基于Apriori算法综合管廊火灾识别预警系统

为了有效识别和预防由不安全因素引起的综合管廊火灾事故,利用多种传感器数据融合方法对综合管廊中可能导致火灾的各种危险因子进行监测,并提出了以PLC为下位机、NI公司开发的LabVIEW作为上位机的火灾危险识别与预警系统。该系统采用改进的关联规则Apriori算法对管廊中诸多引发火灾的危险因素进行了分析,并对分析结果进行了量化。结果表明,该系统能够迅速识别综合管廊中火灾事故的环境因素,并提前作出预警,为城市地下综合管廊火灾预警工程应用提供参考和借鉴。     

基于火灾公众责任险的保险费率的厘定

从火灾公众责任险的保险费率厘定的问题出发,探讨了基于安全系统工程思想的建筑火灾风险评价指标体系的建立方法,并在此基础上探讨了基于置信度和损失率的建筑火灾保险费率确定方法,主要包括建筑主要类别和各子类的费率调整因子的确定等,并利用模拟数据对该方法的具体应用进行了分析,为填补我国在该研究领域的空白提供有益的帮助与指导.     

边缘计算在火灾探测中的应用初探

考虑火灾探测系统边缘数据的爆发式增长,引入边缘计算概念,以动态网络和时序数据库为基础,融合机器学习算法,将计算量从云中心下移至网络边缘侧,设计了基于边缘计算的火灾探测模型架构。系统用ZigBee技术和SDN技术搭建网络拓扑;采用ViBe算法抽取背景模型,随机决策森林算法融合各环境因子;按照数据时序用途分级存储。系统测试结果表明,该方法提升了运算效率和检测精度。     

GA优化BP神经网络在多参量火灾探测中的应用

针对BP 神经网络的随机权重和阈值稳定性不高的问题,运用遗传算法(GA)对BP 神经网络的初始权重和阈值进行优化,提出了一种基于GA 优化BP 神经网络的多参量数据融合方法以实现火灾探测,提高火灾探测准确率和模型泛化性能,并利用该模型对标准明火和阴燃火中的温度、烟雾浓度和CO 浓度进行数据融合实现火灾探测。研究显示,相较单纯BP 神经网络,经GA 优化的BP 神经网络火灾探测算法能够更快速精确地实现火灾探测,探测精度有显著改善,火灾识别准确率提高至98.84%。     

基于注意力机制的3D卷积双波段烟火识别方法

为提高复杂环境下烟火识别的精度，提出一种基于3D卷积和时空注意力机制的双波段烟火识别方法，该方法融合近红外和可见光双波段图像数据，使用视频流中基于时间的动态特征和基于空间的静态特征降低漏报率、误报率。实验结果表明，该算法在双波段数据集上的烟火识别精度达到99.90%，优于其他基于3D卷积的烟火识别算法，同时，模型具有较小的参数量，能够满足实时推理需求。因此，使用双波段特征的同时，结合注意力机制充分利用视频的动态信息，可以有效提高烟火识别精度。     

Fire detection using smoke and gas sensors

Fire detection systems located in aircraft cargo compartments are currently based only on smoke detectors. They generate about 200 false alarms per year for US registered aircraft. The number of false alarms is growing as more planes are outfitted with smoke detectors and air travel expands. Moreover, the survivability of an aircraft in a fire scenario depends on the early detection of the fire. A fire detection system is developed based on the simultaneous measurements of carbon monoxide, carbon dioxide, and smoke. The combination of the rates of rise of smoke and either carbon monoxide or carbon dioxide concentration provides a potential fire alarm algorithm to increase the reliability of aircraft smoke detectors, and to reduce the time to alarm. The fire detection system with the alarm algorithm detected fires that were not alarmed by smoke sensors, and alarmed in shorter times than smoke sensors operating alone.     

Prediction of pullout interaction coefficient of geogrids by extreme gradient boosting model

Geogrids embedded in fill materials are checked against pullout failure through standard pullout testing methodology. The test determines the pullout interaction coefficient which is critical in fixing the embedment length of geogrids in mechanically stabilized earth walls. This paper proposes prediction of pullout interaction coefficient using data driven machine learning regression algorithms. The study primarily focusses on using extreme gradient boosting (XGBoost) method for prediction. A data set containing 220 test results from the literature has been used for training and testing. Predicted results of XGBoost have been compared with the results of random forest (RF) ensemble learning based algorithm. The predictions of XGBoost model indicates 85% accuracy and that of RF model shows 77% accuracy, indicating significantly superior and robust prediction through XGBoost above RF model. The importance analysis indicates that normal stress is the most significant factor that influences the pullout interaction coefficients. Subsequently pullout tests have been performed on geogrid embedded in four different fill materials at three normal stresses. The proposed XGBoost model gives 90% accuracy in prediction of pullout interaction coefficient compared to laboratory test results. Finally, an open-source graphical user interface based on the XGBoost model has been created for preliminary estimation of the pullout interaction coefficient of geogrid at different test conditions.     

基于O-YOLOv3 的无人机森林火灾识别研究

为解决传统森林火灾检测误报率高、响应速度慢等问题,提出了以无人机作为探测平台,地面站作为火灾识别系统,实现森林火灾的自动探测、识别和定位。开发了六旋翼无人机平台,通过所搭载的红外摄像机和机载计算机获取森林火灾现场图像并实时传回地面。利用地面站对所接收到的火灾图像进行处理,实现对森林火场的在线监测。在森林火灾识别算法方面,提出了O_YOLOv3 算法,采用Darknet 框架进行网络训练,使用K_means 方法自动生成锚点,有效提高火灾识别精度与响应速度。将O_YOLOv3 算法与其他几种算法进行对比实验验证本文算法的有效性。实验结果表明：O_YOLOv3 火灾识别算法能够快速、精准识别森林火灾;所研制的基于O_YOLOv3 的无人机森林火灾探测系统能够用于实际森林火灾探测。     

多传感器融合的火灾识别算法研究

针对传统火灾探测存在反应滞后,误报率高等问题,提出一种基于多传感器融合的火灾识别算法。采用运动和颜色检测提取可见光图像的疑似火灾区域,同时采用阈值分割提取红外图像的疑似火灾区域。将两者疑似火灾区域分别进行特征提取,再根据所获特征进行基于 SVM 的融合火灾识别,通过传感器检测实时的温度与烟雾浓度,从而实现可靠和快速的火灾识别。实验结果表明,该火灾识别算法在可接受的时间范围内准确率高且鲁棒性强。     

An algorithm for improving the reliability of detection with processing of multiple sensors''signal

Recent surveys show that most of the false alarms of fire alarm systems are given by deceptive phenomena caused by human actions, such as cigarette smoking and cooking. Software-controlled systems have a potential ability in increasing the reliability of detection signals by data processing. However, it is rather difficult to distinguish an early stage fire from those deceptive phenomena only by processing the data of a single sensor. The authors' new algorithm utilizes the signals from three different sensors, i.e. temperature, smoke and gas (CO) concentration. Incorporating a mathematical fire model, these signals are translated into source parameters such as heat release rate, and smoke and gas generation rates. The conditions of the fire are then analyzed using the cross-correlation function between these source parameters. Testing the algorithm has been done using some experimental data and a promising result has been obtained.     

优化神经网络在多传感器火灾探测中的应用

针对BP神经网络在拟合过程中探测精度低、容易陷入局部最优的问题,提出一种基于遗传算法（GA）和模拟退火算法（SA）共同改进的BP神经网络模型,该网络模型可以有效提高火灾识别准确率,同时避免网络过拟合现象,使预测结果跳出局部最优从而达到全局最优。首先,通过GA改进隐藏层结构部分,然后通过SA改进连接权重部分,最后利用优化后的GA-SA-BP模型对火灾实验数据进行信息融合实现火灾探测。实验研究表明,对比单一BP神经网络,经GA和SA改进后的BP神经网络能够有效改善网络拟合能力,并提升火灾探测精度至98.91%。     

Performance improvement of fire detectors by means of gas sensors and neural networks

Nowadays, fire detection systems are used world wide in order to protect life and goods. However, at the present time, detectors show poor features with regard to detection speed and reliability. They respond to only a few fire parameters like smoke particles and they do not take into account other important fire parameters such as gaseous products. In this paper, we present a new multi-sensor detector consisting of a commercial optical fire detector, a temperature sensor and selected semiconductor metal oxide gas sensors. The use of a multi-sensor detector requires a more sophisticated algorithm than the simple threshold rule. The new algorithms are typically based on pattern recognition systems, consisting of a pre-processing unit, a feature extraction unit and a classification unit. The choice of suitable methods for the feature extraction and the classification is difficult. Most often, the classifier depends on the type of the extracted features. In this paper two methods for the feature extraction with their suitable classifiers are presented and compared. However the classification is based on neural networks.The first algorithm consists of (i) a pre-processing unit; and (ii) a FFT-based feature extraction unit to resolve characteristic fire signatures. For that purpose a moving window has been introduced and a composed signal has been generated from the different pre-processed sensor responses. The algorithm also consists of (iii) a classification unit with a Learning Vector Quantization (LVQ) neural network to classify the extracted features to fire, not fire, or disturbing event.The second algorithm consists of a pre-processing unit and feature extraction method based on the scaling of the quadratic mean value. For this kind of feature extraction a back-propagation neural network as a classifier has been chosen.An important improvement towards the use of commercial detectors has been achieved using both of the above-described algorithms. The neural networks with suitable feature extraction methods were able to detect test fires more quickly than the commercial optical fire detector without generating false alarms by disturbing events.