【作者】 路志强；

【导师】 李书全；

【作者基本信息】 河北农业大学 ， 水利水电工程， 2005， 硕士

【摘要】 建筑产业是我国国民经济的支柱产业之一,建筑行业的良性发展对我国社会主义市场经济建设、国民经济的发展具有至关重要的作用。但长期以来建筑业的安全生产问题一直是困扰业界人士和广大学者的难题。尤其,随着我国经济建设的迅速发展,建筑产业更是突飞猛进,近几年来,建筑安全事故屡有发生,每年的建筑安全事故发生数和建筑安全事故死亡人数有不断上升的趋势。我国政府各级领导历来都十分重视安全生产问题,政府有关部门也采取了不少的对策和措施。但是,这些对策和措施,仍然不能有效遏制安全事故的频繁发生,这不能不引起我们对我国目前所采取的安全生产管理体系的深刻反思。本文根据一定的安全科学方法和步骤以及相应的安全法规、条例、标准等,在充分调查研究的基础上,分析施工过程中各方面的安全因素,对施工的安全状况进行了分析,从而为制定安全评价和预警提供了科学依据。 本文综合利用安全系统工程、计算机软件工程等理论和先进技术,对安全评价和预警进行了深入地、系统的研究,从而建立了基于支持向量机的建筑施工安全预警模型。论文的主要内容如下: 1 在研究我国建筑业安全生产管理的现状和存在的问题的同时,介绍了世界发达国家建筑安全生产管理的经验,分析在我国当前形势下,构建建筑施工安全评价与预警的必要性和迫切性。 2 从体系的建立和系统结构的量化等方面对安全评价框架—安全评价准则体系的建立进行了研究。 3 对支持向量机理论进行基本概念上的介绍,并深入探讨了SVM算法的基本理论。然后对两种具体的SVM算法和LS-SVM算法进行了进一步的深入研究。 4 研究了支持向量机在安全预警工作中的工作机理,并将支持向量机应用于安全预警中。 5 总结了本文的研究工作,对今后的研究进行了展望。更多还原

【Abstract】 The construction industry is one of the principle industries of the national economy. And for building the socialism market economy and developing the national economy, it is very important to keep it developing well. But, for a long time, the safety in the process of the constructing has puzzled the people who are engaged in the construction industry and the scholars as a problem. Especially, with the rapid development of national economy, the construction industry has gained amazingly quick development. In recent year, there happened many accidents in the constructing, and the number of the accidents in one year takes on an ascending trend. The state always pays much attention on the management of the safety, and takes many measures to prevent the happening of the accidents. However, the measures can’t work efficiently. So we must study the safety management system from a new perspective. In this paper, based on ample researching and study, construction enterprise safety alert system is defined to analyze every type of safety factors and to early-warning the safety status in the construction process according to a certain safety scientific approaches, steps, relevant code, statute, criterion, etc. Thus it can provide scientific evidence for enacting fundamental safety early warning.By synthetically using safety system engineering, computer software engineering theories and advanced technologies, this paper has made a profound and systemic research into safety assessment and early-warning working mechanism. Also, it advanced the comprehensive design framework of Safety Early-Warning Model based on support vector machine. The following are the main study contents.1 While analyzing the present situation of safety producing and safety management and introducing the experience or the safety management in the developed countries, point out imminently to establish a new system of safety management in the construction industry.2 Researched into the framework of safety assessment —Safety Assessment Criteria System in the term of the system foundation and system structural quantifying.3 The basic concept of the support vector machine theory has been introduced and the basic theoretical properties have been gone deep into discussed. Then two classical SVM algorithms and LS-SVM algorithm have been done more research.4 This paper studied Support Vector Machine’s (SVM) working mechanism in safety early warning, and applied it to safety early warning.5 Last section draws a conclusion from the whole research and takes a look into the forth-coming work.更多还原