基于隐患报告的事故预测模型及预警方法研究

发布时间：2018-04-25 21:04

本文选题：事故隐患 + 事故预测　；参考：《中国地质大学(北京)》2017年硕士论文

【摘要】：近年来,我国的安全生产形势依然严峻,虽然安全生产事故的起数有所下降,但是,重、特大事故的发生依然触目惊心,全国安全生产形势不容乐观。这也暴露出了我国安全生产工作中事故隐患大量存在,监管不合理,预防措施不到位,预警不及时等等重大问题。在此背景下,加强对事故发生规律的研究,尤其是事故和隐患间的关系,从而对事故进行预测并及时预警成为安全研究的关键。海因里希应用数理统计分析的方法,通过统计大量的事故数据得出了死亡和重伤、轻伤、无伤害事故的比例,并据此提出预防事故的方法论,即要想预防重特大事故的发生,就必须减少甚至消除无伤害事故。但是这一结论也存在一定的缺陷,即它并未指出要注意隐患和险兆事件这些容易被忽略的因素的影响,也就是说它没有关注导致事故发生的根本源头。没有重视系统的本质安全,就不能从根本上避免事故,因此海因里希法则存在一定的局限和缺陷,需要对其进行深化、扩展,从隐患、危险源入手,提高系统本质安全,从而最大程度地减少事故发生。本文从事故预测及隐患研究的现状分析和相关理论入手,通过大量阅读国内外的事故预测及隐患的相关文献资料,具体明确了国内外学者在事故预测方面的研究现状以及传统的“海因里希法则”所存在的欠缺和不足。在此基础上,通过我国近几年数据的统计规律,建立了扩展后的“事故灯塔模型”,和严重事故、死亡人数、一般事故与隐患的回归模型,同时建立了基于隐患的事故预测模型,并对我国未来几年可能存在的事故隐患进行了预测。另外,通过安全生产“十三五”规划确定了2020年的死亡人数控制目标,进而计算出了隐患的控制目标并提出了一套系统的预警方法。为验证本文研究思路的科学性和实用性,本文选取了某公司的隐患排查数据进行实例分析,通过对该公司2012年-2016年的隐患数据进行统计分析,建立了该公司的隐患预测模型,从而对未来几年的隐患进行预测,在此基础上,根据2017年1-4月份的隐患统计数据,确定其预警情况。从而使企业能够采取更科学、更及时且有隐患数据做支撑的预警预控措施,将可能形成事故的隐患消灭于萌芽之中,以实现防止事故发生目的。
[Abstract]:In recent years, the situation of production safety in our country is still grim, although the number of accidents in production safety has declined, but serious, the occurrence of serious accidents is still shocking, the situation of national safety production is not optimistic. This also exposed the safety work in our country in the accident hidden danger exists in large quantities, the supervision is unreasonable, the prevention measure is not in place, the early warning is not in time and so on important question and so on. Under this background, the key to safety research is to strengthen the study of accident occurrence law, especially the relationship between accident and hidden trouble, so as to predict and warn the accident in time. By using the method of mathematical statistical analysis, Heinrich has obtained the proportion of death and serious injury, slight injury, and no injury accident through statistics of a large number of accident data. Based on this, the methodology of preventing accidents is put forward, that is, to prevent the occurrence of serious and serious accidents. It is necessary to reduce or even eliminate harmless accidents. However, this conclusion also has some defects, that is, it does not point out that attention should be paid to the influence of hidden dangers and dangerous and ominous events, that is to say, it does not pay attention to the root causes of accidents. If we don't pay attention to the essential safety of the system, we can't avoid the accident fundamentally. Therefore, there are certain limitations and defects in the Heinrich rule, which need to be deepened, expanded, and started from hidden dangers and dangerous sources to improve the essential safety of the system. Thus minimizing the occurrence of accidents. This paper starts with the analysis of the present situation and related theories of accident prediction and hidden trouble research, and through reading a large number of domestic and foreign literature on accident prediction and hidden danger. The present situation of the research on accident prediction at home and abroad and the shortcomings and shortcomings of the traditional "Heinrich's rule" are clarified. On this basis, the extended "accident lighthouse model" and the regression models of serious accidents, deaths, general accidents and hidden dangers are established through the statistical laws of the data in recent years in our country. At the same time, the accident prediction model based on hidden danger is established, and the potential accident hidden danger in the coming years in China is forecasted. In addition, according to the 13th Five-Year Plan of safe production, the target of controlling the number of deaths in 2020 is determined, and the control target of hidden danger is calculated and a set of systematic early warning method is put forward. In order to verify the scientificity and practicability of this research idea, this paper selects the hidden trouble survey data of a company to carry on the example analysis, through carries on the statistical analysis to the hidden danger data of the company from 2012 to 2016, establishes the hidden danger forecast model of the company. Based on the prediction of the hidden dangers in the next few years, according to the statistical data of the hidden dangers in January to April of 2017, the early warning situation is determined. So that enterprises can take more scientific, more timely and hidden danger data as the support of early warning and control measures, the potential accident hidden danger will be nipped in the bud, in order to achieve the purpose of preventing accidents.
【学位授予单位】：中国地质大学(北京)
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：X928.03

【相似文献】