证据理论在事故树不确定性处理中的应用
发布时间：2018-11-18 21:29
【摘要】：中密度纤维板在满足工农业生产和人民生活需要等方面起着重要的作用。近年来，每年在中国都会有大量的木业加工厂建立，但是伴随着大量的生产线的出现，因发生火灾而遭受严重的损失也日益增加，其中中密度纤维板主生产线的物料输送过程中出现火星而引起火灾的概率最大，一旦发生火灾，木纤维的爆炸式燃烧使得火势蔓延发展极快，难以扑灭。事故树分析模型是针对火灾事故最重要的安全评价模型之一，并广泛应用于各种领域的事故评价中。事故树特有的分析结构可以对火灾诱发原因做出总体至部分按树形结构逐步地细化，能有效地对火灾现场进行监控和评价。然而火灾事故的发生原因、发生形式、发生环境、可燃物种类及火源分布等诸多因素是不确定的，传统事故树分析很难表达出火灾起因中的不确定性，这些因素制约了事故树分析模型在火灾评价中的发展。 本文结合事故树分析评价模型与证据理论，，在事故树的定量分析阶段中运用证据理论原理来处理事故调查中的不确定信息，利用贝叶斯近似法来优化原有的证据融合算法，实现融合不同领域专家诊断的信息。此方法应用于东f木业中纤板生产线防火系统中，提出了更科学更有效率的防火安全制度和安全操作规程，成功提升了该系统火灾防护的能力。 在本文的结尾处对全文的研究工作进行了总结，对未来的研究方向提出了工作重心，并且指出了本文所编制的防火事故树有待改进的地方，和证据理论算法的深入研究的必要性。最后本文对未来在证据理论算法和事故树定量分析结合的研究方向进行了展望。
[Abstract]:MDF plays an important role in meeting the needs of agriculture and industry. In recent years, a large number of wood processing plants have been set up in China every year, but with the emergence of a large number of production lines, the fire has suffered serious losses. The probability of fire caused by Mars in the main production line of MDF is the greatest. Once the fire occurs, the explosion of wood fiber makes the fire spread rapidly and it is difficult to extinguish. Accident tree analysis model is one of the most important safety assessment models for fire accident and is widely used in various fields of accident evaluation. The unique analysis structure of the accident tree can make the whole to part of the cause of the fire induced by the tree structure gradually detailed, and can effectively monitor and evaluate the fire site. However, many factors, such as the cause, the form, the environment, the type of fuel and the distribution of fire source, are uncertain. The traditional analysis of accident tree is difficult to express the uncertainty in the cause of fire. These factors restrict the development of accident tree analysis model in fire assessment. Combined with accident tree analysis and evaluation model and evidence theory, this paper applies evidence theory principle to deal with uncertain information in accident investigation in the stage of quantitative analysis of accident tree, and optimizes the original evidence fusion algorithm by using Bayesian approximation method. The information of expert diagnosis in different fields can be fused. This method has been applied to fire prevention system of medium fiber board production line in Dongfen wood industry. A more scientific and efficient fire prevention safety system and safety operation rules have been put forward, and the fire protection capability of this system has been improved successfully. At the end of this paper, the research work of this paper is summarized, the focus of the future research is put forward, and the improvement of the fire prevention tree is pointed out. And the necessity of further research on evidence theory algorithm. Finally, the research direction of combining evidence theory algorithm and accident tree quantitative analysis in the future is prospected.
【学位授予单位】：武汉科技大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：TS68;X928.7
本文编号：2341308
[Abstract]:MDF plays an important role in meeting the needs of agriculture and industry. In recent years, a large number of wood processing plants have been set up in China every year, but with the emergence of a large number of production lines, the fire has suffered serious losses. The probability of fire caused by Mars in the main production line of MDF is the greatest. Once the fire occurs, the explosion of wood fiber makes the fire spread rapidly and it is difficult to extinguish. Accident tree analysis model is one of the most important safety assessment models for fire accident and is widely used in various fields of accident evaluation. The unique analysis structure of the accident tree can make the whole to part of the cause of the fire induced by the tree structure gradually detailed, and can effectively monitor and evaluate the fire site. However, many factors, such as the cause, the form, the environment, the type of fuel and the distribution of fire source, are uncertain. The traditional analysis of accident tree is difficult to express the uncertainty in the cause of fire. These factors restrict the development of accident tree analysis model in fire assessment. Combined with accident tree analysis and evaluation model and evidence theory, this paper applies evidence theory principle to deal with uncertain information in accident investigation in the stage of quantitative analysis of accident tree, and optimizes the original evidence fusion algorithm by using Bayesian approximation method. The information of expert diagnosis in different fields can be fused. This method has been applied to fire prevention system of medium fiber board production line in Dongfen wood industry. A more scientific and efficient fire prevention safety system and safety operation rules have been put forward, and the fire protection capability of this system has been improved successfully. At the end of this paper, the research work of this paper is summarized, the focus of the future research is put forward, and the improvement of the fire prevention tree is pointed out. And the necessity of further research on evidence theory algorithm. Finally, the research direction of combining evidence theory algorithm and accident tree quantitative analysis in the future is prospected.
【学位授予单位】：武汉科技大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：TS68;X928.7
【参考文献】
相关期刊论文 前10条
1 唐立建;黄德镛;;基于动态贝叶斯网络分析法的炸药库爆炸事故分析[J];安全与环境学报;2010年06期
2 倪国强,梁好臣;基于Dempster-Shafer证据理论的数据融合技术研究[J];北京理工大学学报;2001年05期
3 李勇;王德功;杨佐龙;;模糊神经网络与证据理论的飞机目标敌我识别[J];吉林大学学报(信息科学版);2012年01期
4 章洪池;;浅论纤维干燥系统火灾之消防[J];中国人造板;2007年01期
5 丁丹;;中密度纤维板厂工艺设计中的防火设计[J];中国人造板;2012年04期
6 蔡明;张兆宁;王莉莉;;基于模糊事故树分析法的飞行碰撞风险研究[J];航空计算技术;2011年03期
7 李创起;景国勋;张永强;;基于事故树方法的人机系统可靠性分析[J];机电工程;2011年12期
8 李岳峰，刘大有;证据理论中的近似计算方法[J];吉林大学自然科学学报;1995年01期
9 杨波;徐姣;倪晓阳;;高速公路隧道交通安全事故树分析[J];安全与环境工程;2011年03期
10 刘培兴;;中密度纤维板厂配电设计及电气火灾预防[J];林产工业;2010年03期
本文编号：2341308
本文链接：https://www.wllwen.com/kejilunwen/anquangongcheng/2341308.html
