基于粗糙集的油建施工作业安全预警模型的研究与实现

发布时间：2018-02-26 18:10

本文关键词： 安全预警指标体系粗糙集 Rosetta 规则　出处：《西南石油大学》2015年硕士论文　论文类型：学位论文

【摘要】：油建工程是油气田开发核心之一,因其有野外施工、工种繁多、流动性大、作业时间长、环境较复杂、工序较多等特征,所以在作业的过程中,常常伴有意外事故发生,风险较大。如何保证油建施工作业的安全进行,预防事故的发生始终是油建行业板块需要重点关注的问题,建立油建施工作业安全预警系统是刻不容缓的。本文提出基于粗糙集建立油建施工作业安全预警模型。该模型为构建油建施工作业预警子系统奠定了基础。本文利用粗糙集理论的数据挖掘和知识发现的能力,对油建施工作业安全预警样本中的各种数据进行研究分析并从中获取规则,作为安全预警的依据。本文具体研究主要有以下几个方面： (1)在已有安全预警及相关理论研究的基础上,针对油建施工作业过程中存在的危险、危害因素以及可能发生的油建工程事故的特征,建立一套合适的预警指标体系。 (2)根据各预警指标的发生次数及事故程度,构建决策表。运用粗糙集理论知识,对决策表中条件属性数据进行离散化处理、属性约简,最后生成相应的规则。 (3)依据可信度、覆盖度对规则进行处理,获得有效规则。把获得的规则,导入安全预警规则库,完成匹配算法。 (4)设计并实现基于粗糙集的油建施工作业安全预警模块。 (5)运用数据对安全预警模型进行有效性和实用性验证。
[Abstract]:Oil construction engineering is one of the core of oil and gas field development. Because of its characteristics such as field construction, various types of work, large mobility, long working time, complex environment, more working procedures, etc., accidents often occur in the course of operation. The risk is high. How to ensure the safety of oil construction operation and prevent accidents is always the key problem that the oil construction industry should pay attention to. It is urgent to establish a safety early warning system for oil construction operation. In this paper, a safety early warning model of oil construction operation based on rough set is proposed, which lays a foundation for the construction of early warning subsystem of oil construction operation. This paper makes use of the ability of data mining and knowledge discovery of rough set theory to study and analyze all kinds of data in early warning samples of oil construction operation and obtain rules from them. As the basis of security early warning, this paper mainly studies the following aspects:. 1) on the basis of existing safety early warning and related theoretical research, a set of appropriate early warning index system is established in view of the hazards, harmful factors and the characteristics of possible oil construction accidents in the course of oil construction operation. 2) according to the occurrence times and accident degree of each early warning index, the decision table is constructed, and the conditional attribute data in the decision table is discretized by using rough set theory, and the attribute reduction is made, and the corresponding rules are generated at last. 3) deal with the rules according to the credibility and coverage degree, obtain the effective rules, and import the obtained rules into the security early warning rule base to complete the matching algorithm. Design and implement the oil construction safety warning module based on rough set. The validity and practicability of the security early warning model are verified by using the data.
【学位授予单位】：西南石油大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TE48;TP18

【参考文献】