【作者】 田梅；

【导师】 赵忠杰；

【作者基本信息】 长安大学 ， 交通信息工程及控制， 2008， 硕士

【摘要】 火灾是一种失去人为控制的燃烧过程,产生火灾的基本要素是可燃物、助燃物和点火源,燃烧过程中的物理和化学现象是可以被探测到的。传统的探测方法大多通过采集单一的火灾特征参数信息进行判断和识别,但由于干扰因素多,其误报率一直比较高。近年来,在探测器的灵敏度、可靠性方面做过许多技术改进,使火灾报警准确度得到一定的提高,但还不能满足火灾探测系统的自动化的要求。对火灾信息固有特征,要对其进行全面、准确的描述才能达到减少误报的目的。因而,多判据的火灾探测方法是目前火灾探测领域的主要研究方向。公路隧道火灾在发生初期是最容易扑救和控制的,如果能够及早地探测到隧道火灾的发生,在火灾发生的初期就给出警报,及时进行救灾处理,就能够避免火灾的扩大,减少人员伤亡和财产损失。因此,“及时发现、准确报警”成为隧道火灾研究中的重要问题。本课题的研究内容是对数据融合技术进行研究,并在此基础上设计隧道火灾探测的方案。本论文的主要理论研究工作包括三方面的内容,即隧道火灾探测的原理与方法、数据融合技术的基本原理及体系结构和基于数据融合的隧道火灾自动探测方法。依据数据融合的基本原理,充分利用多个传感器资源,把多个传感器在空间和时间上的冗余或互补信息依据某种准则来进行融合,提出了一种基于数据融合技术的隧道火灾探测算法,以感烟、感温和感光传感器的模拟量为输入,利用人工神经网络和模糊逻辑技术对多传感器信号进行融合,设计出一种快速、准确和有效的隧道火灾自动探测融合系统,可缩短报警时间,降低误报率。克服了以往火灾探测算法单一使用固定阈值的弊端。同时,通过隧道监控系统将火灾信息发布至信息显示设备,及时作出消防救灾方案。本文最后对模糊逻辑推理进行仿真,从而验证了本文提出的隧道火灾探测算法的有效性和可行性。更多还原

【Abstract】 The fire is a combustion process that lost control, the basic of fire are combustible and combustion-supporting stuff, fire fountain. The physics and chemistry phenomenon in combustion process could be detected. The traditional survey methods merely carry on the judgment and the recognition through gathering sole fire characteristic parameter information. So it is disturbed by the environment inevitably. The question about the system’s high mistakenly reporting rate is prominent. In recent years, the accuracy offire alarm becomes better, while the sensibility and reliability of detectors are improved.But it can’t match the request of the automation of the fire detection system. It can decrease mistakenly reporting only by describing fire inherent characteristic completelyand exactly. Thus fire detection algorithm based on multi-criteria is the main research direction in the present fire surveying field.Highway tunnel fire in the initial stage is easy to save and control, if we can detect the tunnel fire as early as possible, give the alarm and do some measure, it could avoid the spread and reduce the loss of personnel and property. So the important problem in tunnel fire study is“discover in time, exactly alarm”.The project which the thesis based on is to do someresearches on the data fusion technology, then to design and implement tunnel fire detecting system. The theoretic in this thesis consists of three parts: basical principle and system structure ofdata fusion technology; Principle and method of tunnel fire detecting and tunnel fire detecting system based on data fusion system.According to the rationale of data fusion, the space-time redundancy and complemented sensor information was composing by the rule, proposed a kind of tunnel fire detection based on the data fusion technology. The input analog signals are provided by smoke, temperature and light sensors, and the algorithm synthesizes those signals by the neural network and the fuzzy logic technology, designed a kind of quick, exact and effective tunnel fire detecting system, which can shorten the time of alarm and reduce the rate of misinformation. The results showed that it conquered the disadvantage of using fixed threshold in fire detection. Meanwhile, tunnel fire alarm system releases information to the displaying devices and administrators will make fire disaster relief programme in time.In the end, The paper simulates the fuzzy-logic discursion. Simulating model can proves the feasibility and validity of the tunnel fire detection proposed in this paper。更多还原