【作者】 丁宏军；

【导师】 姜德生；

【作者基本信息】 武汉理工大学 ， 通信与信息系统， 2013， 博士

【摘要】 随着我国现代化建设迅速推进,电力在工业中的应用越来越广泛,电力传输安全问题也引起人们高度重视。国内长距离输电主要靠电缆隧道,但电缆隧道火灾频繁发生,而且电缆隧道环境恶劣,火灾扑救困难、易复燃、扑救时间长、损失大。近几年,火灾探测技术不断进步,新技术层出不穷,特别是以光纤传感技术为基础的光纤喇曼火灾探测技术和光纤光栅感温火灾探测技术,以其突出的技术优势,已在石油石化等易燃易爆场所和公路隧道等恶劣环境替代传统技术,成为主流的火灾探测技术。但是,由于电缆隧道结构复杂,火灾成因多,现有光纤感温火灾探测技术难以直接用于电缆隧道火灾探测中。本文针对目前缺乏电缆隧道火灾初起阶段模型研究,光纤感温火灾探测系统缺乏设计依据的问题,以电缆隧道为应用环境,建立理论模型计算不同火灾场景的温度场分布,通过电缆隧道火灾模拟再现实验,研究不同线型感温火灾探测器的响应,并结合实验修正理论模型,指导建立探测器安装、使用和验收规范。本文首先研究现有光纤喇曼和光纤光栅感温火灾报警系统的基础理论、探测机理和产品技术指标,从环境因素、火灾探测影响因素等方面对这两种技术做出分析和比较。然后,统计分析现有火灾案例,将火灾分为电缆过热火灾、初起小规模火灾和大规模火灾,并分别制定了建立模拟火灾场景的方案。调查光纤感温火灾探测技术的电缆隧道应用环境,建立不同应用环境的传热模型,分析计算不同火灾场景的温度场分布及其动态趋势。根据传热模型对火灾温度场的预估和对现有火灾探测技术的了解,制定实验方案；搭建实验平台,恢复应用环境或者在实体工业环境中搭建实验平台。最后,分析火灾探测器在不同火灾场景模拟中的技术指标,比较实验中和理论模型中所得到温度场差异,分析其产生的原因,修正传热模型,优化实验方案和探测器的安装或设置参数。得到的结论为：电缆过热火灾只能影响过热电缆所在层的温度分布,而对相邻的上下层影响较小,相邻的电缆温度都会升高,只有采用直接接触方式进行敷设,才可能有效地探测到电缆的温度变化；初起小规模火灾的热释放速率及热辐射规模均较低,引起火灾发生位置温度的迅速上升然后逐步下降,但是影响范围很小,由于发生位置的不确定性,无论采用何种安装方式均存在探测盲点；大规模火灾的发生会引起隧道顶部的温度迅速上升,火灾发生位置正上方的温升达到30℃/min。线型火灾探测器可悬挂安装在隧道顶部,光纤光栅温度探测单元的间距设置不超过6m,适合采用差温报警功能,且升温速率的报警阈值可设置为5℃/min。本文的主要研究成果和创新点在于：1)在实体电缆隧道中,构建了电缆过热,初起小规模和大规模火灾等三种典型的电缆隧道火灾模拟场景和实验数据采集系统；2)基于有限元热传导分析方法,建立了电缆隧道不同火灾的温度模型,通过仿真和模拟实验,分级了基于光纤感温技术的模拟温度场和实验温度场的分布和变化规律,提出了光纤感温火灾检测系统在电缆隧道中的安装和使用设计规范；3)通过对三种典型电缆隧道火灾的模拟实验,建立了初起阶段火灾的传热模型,根据此模型能预估不同火灾场景中温度场的分布和变化规律。更多还原

【Abstract】 With the rapid improvement of China’s modernization, electrical power is applied in industry more and more widely and power transmission security issues aroused great attention. But the cable tunnel fire occurred frequently, which bear long-distance power transmission, and the harsh environment in cable tunnel causes difficult fire fighting, easy resurgence, long-time fighting and big loss.In recent years, fire detection technology continuously improve and new technologies come out one after the other, especially fiber Raman fiber grating fire detection technology and thermal fire detection technology that base on optical fiber sensing technology. Since their outstanding technical advantages, they have displaced traditional fire detection technologies in explosive areas including petroleum and petrochemical and harsh environment including road tunnels. However, due to the complex structure of the cable tunnel and fire causes, existing fiber thermal fire detection technology can not be used directly in the cable tunnel fire detection. Aiming at the issues of the lack of cable tunnel fire early stage model and the lack of design basis for new fire detection technologies in cable tunnel, this paper establish theoretical model for cable tunnel, calculate temperature distribution for different fire scenes, study the response of different linear heat detectors in different fire scenes reproduction experiments, fix theoretical models combining with experiment results to guide the standard establishment for the detector installation, application and acceptance check.Firstly, the theoretical basis, detection mechanism and product specifications of existing fiber Raman and FBG temperature fire alarm systems are studied, which are analyzed and compared in terms of environmental factors and fire detection factors. Then, statistical analyzing existing fire cases, the fire is divided into cable over-heating fire, small-scale fire and large-scale fires and fire scenes reproduction programs are developed. Investigating cable tunnel application environment for fiber temperature sensing fire detection technology, application environments transfer model is established to calculate and analyze the temperature distribution and dynamic trend for different fire scenes. According fire temperature field prediction from heat transfer model and understanding of existing fire detection technologies, experimental program is developed and experimental platform is established in real or rebuild industrial environment. Finally, technical indicators of fire detectors in different fire scenarios simulation are analyzed, comparing temperature field differences between experimental and theoretical models are compared and their causes are analyzed to modified heat transfer model and optimize the experimental program and detector installation or setup parameters. The conclusions are following:1) overheating cable only affects its layer where neighboring cable temperature will rise, but effects to adjacent lower and upper layers is little. Only detector that lay on layer and contact directly could detect the temperature change of the cable and give alarm signal effectively.2) Since heat release rate and thermal radiation of small-scale fire are low, the temperature at fire location increases rapidly and then decreases gradually. Because its affected area is small and occurrence location is uncertain, every installation method has detection blind spots.3) Large-scale fire will cause temperature rapid increasing in the top of tunnel and the increasing rate will reach30℃/min just above the pool fire location. Linear fire detectors can be suspended under the tunnel ceiling. The distance setting between two FBG temperature detection units does not exceed6m, differential temperature alarm function is suitable and heating rate alarm thresholds can be set to5℃/min.The main research results and innovation of this paper is to:1) three typical cable tunnel fire simulation scenarios including cable overheating fire, small-scale fire and large-scale fire and experimental data acquisition system are established in a physical cable tunnel;2) Based on finite element heat conduction analysis, cable tunnel different fires temperature model are established, through simulation and modeling experiments, temperature field distribution and variation of simulation and experimental results are analyzed and installation and using design specifications for optical fiber temperature sensing fire detection system in cable tunnel are proposed;3) through three typical cable tunnel fire simulations establishing a heat transfer model of the early stages fire, according to this model, it can estimate different fire scenarios in the temperature field distribution and variation.更多还原