基于贝叶斯分类的煤矿系统安全动态定量评价  

Dynamic and quantitative safety evaluation of mine system based on Bayesian classification

作　　者：董丁稳[1] 陈赞 

机构地区：[1]西安科技大学能源学院,陕西西安710054 [2]陕西汇森煤业开发有限责任公司,陕西西安710065

出　　处：《西安科技大学学报》2015年第3期320-324,330,共6页Journal of Xi’an University of Science and Technology

基　　金：西安科技大学博士启动基金资助(2012QDJ30)

摘　　要：为及时、合理评价煤矿生产系统安全状况,促进现场隐患闭环管理,将源于现场日常隐患排查得到的隐患信息,从不同角度提取系统安全属性作为评价指标,形成一个表达系统安全状态的逻辑结构,通过建立安全隐患风险等级与系统安全评价标准之间的映射关系对系统安全属性量化,用以定量表示系统安全逻辑结构中指标层的安全状态。运用贝叶斯分类法分析历史评价样本及指标量化数据,用最大后验概率表达评价样本对于评价等级的隶属关系来得出评价结果,实现了煤矿生产现场系统安全动态、定量评价,通过实例分析表明了该方法适用于现场短周期的系统安全动态评价。For the timely and reasonable safety evaluation of mine production system to promote closed loop of safety hazard on site,the system safety attributes as evaluation index were abstracted by daily hazard identification on site,which formed an logical structure for expressing system safety status,and the evaluation index was quantified by mapping the risk level of safety hazards with the system safety evaluation criterion to quantitantively characterize the safety state of the index layer. Then Bayesian classification method was used to analyze the historical data of evaluation sample and used the greatest posterior probability to express the affiliation for evaluation sample with evaluation level to determine the evaluation results,which achieved a dynamic,quantitative safety evaluation,the case study shows that this method is suitable for short periodical dynamic safety evaluation of mine production system.

关 键 词：煤矿 系统安全 动态评价 

分 类 号：X936[环境科学与工程—安全科学]