节点文献

基于突变论的智能建筑中突发事件处理的决策支持与虚拟仿真

Decision-support and Virtual Simulation Based on Reasearch of Catastrophe Theory for Dealing with Unexpected Events in IB

【作者】 魏伟

【导师】 余庄;

【作者基本信息】 华中科技大学 , 建筑设计及其理论, 2009, 博士

【摘要】 论文针对智能建筑面向突发事件处理存在的问题,建立建筑突发事件模型,利用数学分析的方法分析建筑突发事件的特点,构建了突发事件的处理机制,包括智能建筑系统集成,突发事件决策支持和虚拟现实三大模块,并对这些模块进行了研究设计和具体实现。论文的主要创新性工作有:(1)提出突变论是描述建筑突发事件的最佳方法,并给与相关理论证明,结合突变论的知识,加深了对建筑突发事件特点的理解,建立了相关的数学模型,填补了建筑突发事件“定性分析”的空白。将建筑突发事件分为九类基本突发事件:火灾、非法侵入、设备故障与报警、恐怖袭击、爆炸、化学物品泄漏、建筑结构扭曲或倒塌、地震、自然灾害,其中设备故障与报警、爆炸属于尖点突变类突发事件;火灾、建筑结构扭曲或倒塌属于燕尾突变类突发事件;地震、自然灾害、非法侵入、化学品泄露属于折迭突变类突发事件。通过数学分析,辅助制定相应的决策,为对建筑突发事件的深入研究奠定基础。(2)目前多是针对人的逃生行为的研究,以及人在逃生过程中逃生心理的研究。论文研究利用建筑设备进行减灾运行并指导帮助人逃生。(3)为了快速得出决策,提出利用向量余弦法进行案例的相似度搜索,相比文字查询方法在时间上少了一个数量级。(4)提出利用数据仓库技术实现BAS,CAS和IAS的数据集成,并结合现有系统,实现智能建筑应对突发事件的系统联动控制,为今后实现智能建筑更高层次的集成奠定基础。(5)针对虚拟火灾突发事件,鉴于现有火焰模型的不足,论文提出了通过一组粒子建立流体模型,它们的统计特性和基于流体方程变化的粒子相同,且在数值上比方程的确定性解更加稳定。同时,在火焰中加入燃烧成分,实现了火焰的动态燃烧,增强了火焰的真实性。论文所建立的建筑突发事件突变模型,为今后进一步研究建筑突发事件提供了理论基础;论文对智能建筑系统集成、决策支持以及虚拟现实等方面问题的研究和具体实现,为实际工程中智能建筑突发事件处理提供了开发的参考。

【Abstract】 This paper orients the problems of unexpected events for intelligent building and sets up mathematics models, through which analysises the characters of building unexpected events. A mechanism is proposed by this paper including three modules to deal with building unexpected events, which are intelligent building system integration, decision support and virtual reality. This paper provides a design method of the mechanism and a way to realize it.The main innovative works of this paper: (1) Catastrophe theory is proved to be the best way to describe intelligent building unexpected events, which gives a better understanding of building unexpected events through establishing a mathematical model. This work fills the blank of "qualitative analysis" of intelligent building unexpected events. Intelligent building unexpected events are divided into nine basic categories: fires、illegal intrusion、equipment malfunction and alarm、terrorist attacks、explosions、chemical spills、twist and collapse of building、earthquakes、natural disasters. Equipment malfunction and alarm、explosions are cusp-type unexpected events; fire、twist or collapse of building structures belong to swallowtail-type unexpected events; earthquakes、natural disasters、illegal intrusion、leaking chemicals are fold-type unexpected events. The result of mathematical analysis of catastrophe models can support the decision-making of dealing the intelligent building unexpected events and lay a foundation for in-depth study. (2) At present most of the research are aimed at people’s escape behavior, as well as the psychological. This Paper focuses on equipment of the intelligent building for disaster reduction and to provide guidance to help the escaping. (3) In order to quickly get the decision-making, vector cosine law is used for searching similarity cases. Compared to the use of words searching, it takes more less time to get the most similarity case. (4) This paper bring forward warehouse technology to realize BAS, CAS and IAS data integration, and based on existing systems, intelligent building handle Intelligent building unexpected events through systems integration, which lay a foundation for a more higher integration of Intelligent building. (5) In the aspect of virtual unexpected events, focus on the shortage of existing flame model, paper proposes a random Lagrangian method based on non-grid, which through a group of particles to modeling the fluid. Its statistical properties and the equation based on fluid changes in the same particles, but more stable. At the same time chemical composition is added into fire simulation which increases the authenticity of flame.The catastrophe models of unexpected events established in this paper provides a theoretical foundation for further study of building unexpected events. The research and concrete realization in this paper on intelligent building system integration, decision support, as well as virtual reality and other aspects not only provides for practical engineering of intelligent building, but also gives a reference for further development.

节点文献中: 

本文链接的文献网络图示:

本文的引文网络