基于贝叶斯网络的公路养护作业风险评价与控制研究

发布时间：2018-07-06 16:10

本文选题：公路 + 养护　；参考：《长安大学》2017年硕士论文

【摘要】：公路养护作业面广量大,在开放交通的复杂环境下作业,各种风险因素随时都有可能导致安全生产事故的发生,严重威胁养护作业人员的生命财产安全。随着我国公路网规模的日益庞大[1],公路养护与管理的工作量与日俱增,公路养护作业中越来越多的生产安全问题突显出来,养护作业过程中的伤亡事故率不断攀升。同时,公路养护作业过程具有复杂、多变的特点,使得公路养护安全作业的管理不能单纯依靠经验,还需要先进的技术手段。因此,本文以降低公路养护作业安全生产事故率,提高作业安全水平和作业区通行能力为目的,展开公路养护作业风险评价与控制研究,主要研究内容包括以下几方面:首先,在深入挖掘公路养护作业事故数据的基础上,剖析了公路养护作业安全事故特征,总结了公路养护作业不同于公路施工作业的原因,指出了公路养护作业主要的事故类型;提出运用系统安全理论进行公路养护作业风险因素分析的基本思路,设计并开展了覆盖陕西省的1272份问卷调查,利用SPSS软件对回收的1073份有效问卷进行因子分析,最终提炼出17个主要影响指标,据此构建了简明实用,操作性强的公路养护作业风险评价指标体系。其次,借助贝叶斯网络分析软件GeNIe2.0,结合专家知识和机器学习,确定各节点间的逻辑关系,建立贝叶斯网络模型拓扑结构,并确定贝叶斯网络的参数,进而构建了完整的公路养护作业风险评价模型;利用模型进行了逆向推理、敏感性分析、影响强度分析,得出了目前公路养护作业中最为危险的风险因素,并提出了针对性的安全生产建议。最后,结合实例,阐述了评价模型的应用过程,并验证了模型的可靠性、实用性和准确性。
[Abstract]:In the complex environment of open traffic, all kinds of risk factors may lead to the occurrence of safety accidents at any time, which seriously threaten the safety of life and property of maintenance workers. With the increasing scale of highway network in our country, the workload of highway maintenance and management is increasing day by day, more and more production safety problems are highlighted in highway maintenance operation, and the casualty accident rate in the course of maintenance operation is rising. At the same time, the operation process of highway maintenance is complex and changeable, which makes the management of highway maintenance safety operation not only depend on experience, but also need advanced technical means. Therefore, the purpose of this paper is to reduce the accident rate of road maintenance work safety, improve the safety level and capacity of operation area, and carry out the risk assessment and control research of highway maintenance operation. The main research contents include the following aspects: first, On the basis of deeply excavating the accident data of highway maintenance operation, this paper analyzes the characteristics of highway maintenance operation safety accident, summarizes the reasons why highway maintenance operation is different from highway construction operation, and points out the main accident types of highway maintenance operation. This paper puts forward the basic idea of risk factor analysis of highway maintenance operation by using systematic safety theory, designs and carries out 1272 questionnaires covering Shaanxi Province, and makes factor analysis of 1073 valid questionnaires collected by SPSS software. Finally, 17 main impact indexes were extracted, and a simple and practical risk evaluation index system of highway maintenance operation was constructed. Secondly, with the help of Bayesian network analysis software GeNIe2.0, combining expert knowledge and machine learning, the logical relationship between nodes is determined, and the topological structure of Bayesian network model is established, and the parameters of Bayesian network are determined. Then a complete risk assessment model of highway maintenance operation is constructed, and the most dangerous risk factors in highway maintenance operation are obtained by reverse reasoning, sensitivity analysis and impact strength analysis. Some suggestions on safety production are put forward. Finally, the application process of the evaluation model is expounded, and the reliability, practicability and accuracy of the model are verified.
【学位授予单位】：长安大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：U418.2

【参考文献】