Optimization of gas detector placement considering scenario probability and detector reliability in oil refinery installation

Gas detection system is a critical layer of protection in process safety. Leak scenario probability and detector reliability are two key factors in the optimization of gas detector placement. However, they are easily neglected in previous studies, which may lead to an inaccurate evaluation of the optimization solutions. In this study, a stochastic programming (SP) optimization method is proposed considering these two factors. In order to quantitatively represent the probability of leak scenarios, a complete accident scenario set (CASS) is built combining leak sources and wind fields. Then, the computational fluid dynamics (CFD) method is adopted for consequence modeling of gas dispersion. The Markov model is developed to predict the detector reliability. With the objective of minimal cumulative detection time (MCDT), the SP formulation considering scenario probability and detector reliability (MCDT-SPR) is proposed. By introducing the particle swarm optimization (PSO) algorithm, the optimization formulations can be solved. A case study is investigated on a diesel hydrogenation refining unit. Results validate this approach is promising to improve the detection efficiency. This method is more practical and matching with the actual industrial environment, where the leak scenarios and the detector reliability can change dynamically in real process setting.     

Evaluation on consequences prediction of fire accident in emergency processes for oil-gas storage and transportation by scenario deduction

This paper takes the safety in emergency processes as the starting point, from the perspective of scenario deduction, to study the consequences of fire accidents for oil-gas storage and transportation. Through the statistical analysis of actual accident cases, 19 frequently occurring basic scenarios in emergency processes are summarized. The scenario evolution paths of fire accidents for oil-gas storage and transportation are given by analyzing the evolution law of the accident development. Fuzzy numbers are introduced to express experts' qualitative judgment on accident scenarios. The empirical probabilities of scenario nodes are obtained by defuzzification calculation, and the state probability of each scenario node is calculated by using the dynamic Bayesian network joint probability formula. Under the comprehensive consideration about the probability statistics of actual accident cases, the critical scenario nodes on the evolution path and their final scenario probabilities are jointly determined to realize the optimization of the scenario evolution path. By constructing the correlation between the optimized scenario evolution path and the accident consequences, an accident consequence prediction model is established. The occurrence probability of accident consequences is calculated by the defuzzification method and dynamic Bayesian network. The accuracy of the consequence prediction model is verified by the July 16 Dalian's Xingang Harbor oil pipeline explosion accident. The research results provide scientific basis for helping decision makers to make the effective emergency measures that are most conducive to the rapid elimination of accidents and reducing the severity of accident consequences.     

生产事故引发环境事件应急演练的典型问题剖析

当前基层组织的突发环境事件应急演练重点不明确,多数裹挟着厚重的生产事故救援内容,甚至偏离了环境应急演练主题,这对真实突发环境事件应对是不利的。基于最坏情景演练思想,剖析生产事故与环境事件的演练重点,厘清生产事故引发环境事件的处置流程和演练模式,并借助化工企业甲醇泄漏事故引发环境事件的桌面推演实例,廓清复杂环境事件应急演练的构设方法与实施步骤。结果表明:生产事故引发环境事件应急演练策划设计中,生产事故应急部分可有效集成为向环境应急提供信息的通道,提供关键处置结果作为环境应急主题演练的响应信息即可;生产事故引发环境事件应急演练组织实施中,重点凸显环境应急决策和部署情况,可根据演练场景需要,重点设置应急监测、污染控制、事故调查等应急功能（组）,以考查或展示环境综合应急能力。     

性能化设计思想在化工设施布局安全设计中的应用

指令性规范是目前化工设施平面布局安全设计的主要依据,但其在应用中存在着条款僵化、安全理论基础不全面、可拓展性不明确等问题。本文将性能化设计思想引入到化工企业平面布局安全设计中,初步提出了一种包含危险辨识、性能化目标确定、后果评估及安全设施效用评估等主要组成部分的基于性能化设计思想的平面布局安全设计体系。对体系中最重要的部分—性能化目标确定进行了重点分析,将化工设施布局安全设计分为装置内设备之间的位置设计、厂区内装置区之间的位置设计以及厂区与外部单位之间的位置设计三个级别,从工艺关联性等角度对各级别布局设计中事故场景选择及设施可接受受损水平进行分析,并提出了通过匹配事故场景和设施可接受受损水平来确定性能化目标的方法。     

Scenario derivation and consequence evaluation of dust explosion accident based on dynamic Bayesian network

In order to clarify the correlation between the evolution path of dust explosion accidents and emergency decision-making, and to accurately predict the disaster damage levels of various disaster bearing bodies. This paper extracts 56 key scenario elements from four aspects, namely state, answer, goal and environment, based on the analysis of typical dust explosion accident cases. At the same time, a general scenario evolution path of dust explosion accident is constructed. Using fuzzy number set theory and dynamic Bayes joint probability model, the accurate solution of scenario state probability was realized. With the help of accident cases and dynamic Bayes approach, the dust explosion consequence prediction index system and evaluation criteria were constructed, covering factors such as dust explosion intensity, casualties, direct economic losses, equipment damage, building damage, environmental damage and other factors. A polyethylene wax dust explosion accident in a city of China was used to verify the dust explosion accident scenario evolution model and consequences prediction model. The predicted results were in good agreement with the actual damage of various carriers of the accident, which indicated that the model could be used for dust explosion accident prediction and disaster loss prediction. The research results provided reference and technical support for the prediction of dust explosion accident evolution direction, emergency aid measures decision and deployment, disaster damage prediction and evaluation.     

Real-time risk assessment of explosion on offshore platform using Bayesian network and CFD

Combustion or explosion accident resulting from accidental hydrocarbon release poses a severe threat to the offshore platform's operational safety. Much attention has been paid to the risk of an accident occurring over a long period, while the real-time risk that escalates from a primary accident to a serious one was ignored. In this study, a real-time risk assessment model is presented for risk analysis of release accidents, which may escalate into a combustion or explosion. The proposed model takes advantage of Fault Tree-Event Tree (FT-ET) to describe the accident scenario, and Bayesian network (BN) to obtain the initial probability of each consequence and describe the dependencies among safety barriers. Besides, Computational Fluid Dynamics (CFD) is applied to handle the relationship between gas dispersion and time-dependent risk. Ignition probability model that considering potential ignition sources, gas cloud, and time series are also integrated into this framework to explain the likelihood of accident evolution. A case of release accidents on a production platform is used to test the availability and effectiveness of the proposed methodology, which can be adopted for facilities layout optimization and ignition sources control.     

Tractor rollover fatalities,analyzing accident scenario

IntroductionIn many countries, traditional data sources for collecting injuries of workers covered by compulsory accident insurance have recently been integrated by new observatories whose results may differ. A comparative analysis of the Italian data collection systems related to fatal tractor accidents in agriculture was performed focusing on tractor rollover fatalities with the aim of analyzing the accident scenario.MethodData from the Operational Archives of the Italian Workers Compensation Authority (INAIL), which collects injuries of workers covered by compulsory accident insurance and those of the National Surveillance System (INAIL_ASL), which provides narrative text reports of work-related fatal accidents have been analyzed and compared to the information collected by the INAIL Observatory. The INAL Observatory was recently set up to complement the collection of fatal accidents involving agricultural machinery. Italian data were then compared to data available at an international level. Fatal tractor accidents vary considerably with respect to fatal accidents in agriculture, being 10.6 and 43.7% for the Operational Archives and Surveillance System, respectively. National Surveillance System records, implemented with narrative texts allowed the accident scenario to be defined.Results71.7% of fatal tractor-related accidents refer to non-ROPS equipped vehicles and of these, 26.5% involved machines originally mounted with a ROPS that had been removed or was inoperative in the folded-down position during the rollover event. Just one fatal event from a collapsed ROPS on the overturned tractor was recorded. It is interesting that 16.6% of fatal accidents involved a clear environmental factor.Practical applicationA campaign to train tractor drivers on the correct use of the combination ROPS and seatbelt can contribute to decreasing rollover events with fatal outcomes. Contemporarily a strict requirement to install ROPS and a seatbelt on tractors, combined to an official inspection at the farm level, can increase the chance of survival in a rollover accident.     

基于案例推理的突发环境事件应急决策模型

突发环境事件不确定性高、演化过程复杂,给企业及政府的应急工作带来很大的困难,故如何提高应急决策的质量,是相关应急工作的重点之一。为了增强以往经验的可用性,利用本体模型进行案例的存储,根据情景构成要素提取方法对情景进行多维的空间表达,在此基础上优化情景匹配算法,实现基于情景的应急决策。最后,通过突发环境事件实例应用,验证该方法能够快速建立实例与本体模型之间的关系,提高知识的可用性和共享性,实现情景的多维表达,进而满足“情景-应对”的应急决策要求。     

An urban pipeline accident model based on system engineering and game theory

Urban pipeline accidents are caused by complex social-technical factors, in which urban communities and pipeline systems are involved. Such accidents can thus be investigated from the viewpoint of system engineering. System-Theoretic Accident Model and Processes (STAMP) is a systemic method for safety assessment, which has been adopted in many domains. This approach can provide deep insights of accident causes by considering direct and indirect factors. Meanwhile, competition and cooperation between stakeholders in accidents are observed. Therefore, these parties can also be analyzed with the game theory. That is, stakeholders in STAMP can be regarded as players in game. The aim of this paper is to provide a new insight to analyze urban pipeline accidents by considering both STAMP and game theory. In this paper, we proposed an accident model for urban pipelines, with a case study of China-Qingdao pipeline accident occurred in 2013. We concluded that accident reasons can be investigated in-depth and lessons can be learned from analyzing causal factors by using STAMP. Based on results generated from STAMP, we applied the game theory to analyze roles that government and companies act in the China-Qingdao urban pipeline accident. The results show that current punishment and incentive systems are incomplete, lacking of the driving force and constraining force for the stakeholders involved in the accident.     

Low rate releases of hazardous light gases under semi-confined geometry: A consequence based approach and case-study application

As evidenced by accident statistics, an important factor when considering the safe operation of process plants is the effective mitigation of the effects of gaseous flammable releases, either by a prevention, or a protection approach. A detailed historical analysis was performed considering accidental scenarios associated with the use and management of light gases, starting from raw data selected from FACTS database and analysed by a causal multi-layer method. Results revealed that the major part of the accidental releases involving methane, hydrogen, ethene, ammonia can be attributed to organizational or process/plant immediate causes. As expected, the most frequent scenarios following the release are fire and explosion. We focus our attention on the development of a short-cut method allowing preliminary evaluation of the maximum gaseous build-up under semi-confined conditions, limiting the effects of the fire/explosion scenario to a tolerable level. The limitations of the model that is applied to selected case-studies and require further experimental validation are critically discussed. The results of the application of the model, which can boast of being safe but not disproportionately conservative, can be set as a maximum threshold in proper designing technical measures aiming at limiting the effects to a tolerable level by protection methods, e.g. isolation, venting, suppression and containment.     

Mechanism analysis and risk assessment of escalation scenario in chemical industry zones

Many major hazard installations (MHIs) are located in chemical industry zones and escalation effect may be triggered when the fire or explosion occurs on a MHI. To investigate the mechanism of the accident escalation, a systematic quantitative assessment methodology is proposed by the considering the feature and uncertainty of the escalation scenario. The main accident energy carriers of the escalation are heat radiation, overpressure of blast and fragments. The escalation probability, joint influence of the three energy carriers and risk characterization of the accident scenarios are carried out. By the new methodology, the escalation scenario in chemical industry zones can be analyzed and the risk escalation morphology is demonstrated by the simulation software. The visualized risk cloud figure gives a supplementary way to prevent the escalation scenario in chemical industry zones planning.     

高硫炼油装置硫化氢泄漏场景集定量分析

随机规划方法已成为解决不确定条件下高硫炼油装置气体检测报警仪布置优化问题的重要途径,而构建接近真实情况的泄漏场景集则是实现随机规划的基础。目前,有毒气体泄漏场景集构建方面的研究鲜有报道,且传统以典型泄漏场景为代表的做法未能体现真实风险。为此给出一种融合泄漏概率和风场联合分布概率的定量构建硫化氢泄漏场景集的方法,使场景集包括重要泄漏场景并运用DNV的LEAK软件和历史气象数据定量预测场景的实现概率,为实现后续的气体泄漏检测报警仪布置随机规划提供技术支持。并以某柴油加氢装置为例,定量构建其硫化氢泄漏场景集,为后续的气体检测报警仪布置优化及其它基于定量风险分析的控制决策提供支持。     

基于演化博弈的危化品安全监管情景推演研究

针对危化品安全生产监管问题,基于演化博弈理论构建危化品安全监管演化博弈模型,并将危化品事故发生率引入模型,对比分析危化品企业与地方政府监管部门行为策略的演化稳定均衡。在此基础上进行情景推演模拟仿真,研究表明:危化品事故发生率,对危化品企业和地方政府监管部门的策略选择有显著影响,当危化品事故发生概率低于某一临界值时,危化品企业和监管部门都会疏忽安全投入和监管;地方政府承受危化品事故经济损失和信誉损失增大时,危化品安全监管系统演化呈现出周期性波动;引入上级政府惩罚机制情景下,当上级政府惩罚力度高于危化品企业未投入安全生产受到的处罚和地方政府安全监管成本时,其最终都选择安全投入和严格监管策略。研究结论为政府监管危化品安全生产提供新的思路和对策建议。     

石化企业平面布局安全设计中事故场景的选择研究

基于安全性能思想的设计方法是石化行业平面布局安全设计的一个重要发展方向,文章对这种设计方法中目前存在的主要不确定因素——事故场景的选择问题进行了研究。对《石油化工企业设计防火规范》等指令性规范中各级别平面布局安全设计的依据进行了统计分析,给出了各级别布局设计中主要的安全考虑和大致的安全间距范围,并对目前安全考虑中存在的模糊性进行了分析,给出了导致其存在的原因主要为＂事故场景考虑不同＂和＂‘相互影响’的含义不同＂;由此,结合事故伤害理论,对石化行业常见的9种19类常见事故场景的后果分析估算,得到了各事故场景对目标设施造成不同程度＂影响＂的范围,并以之为依据,分析给出了各级别布局安全设计中适合考虑的事故场景及适当的安全设计目标。     

情景-应对模式下考虑软时间窗的高铁应急资源调度优化

为提升高速铁路应急处置水平,考虑突发事件发生后应急资源需求的不确定性,通过对事故情景集的分析和计算,采用"情景-应对"模式,获取事故点的资源需求量;通过引入"软时间窗"概念,建立以超时惩罚成本、资源运输成本和救援点固定出动成本3者之和最小化为优化目标的应急资源调度模型,利用遗传算法进行求解,并基于实际事故案例验证方法和...     

安全生产效益的分析评价及其与安全投入的关系

通过把握和界定安全效益、安全投入等反映安全生产的经济活动特征和规律的经济学范畴的内涵,指出从企业的经营利润指标上看安全投入不能带来直接经济效益,只产生远期(长远)的、潜在的间接性效益,包括间接经济效益和非经济效益,安全投入的效益具有间接性、潜在性、隐形性、不确定性和迟滞性。由于生产安全事故风险不可能为零,当事故预防水平达到一定程度(可接受的程度及其以下)时,通过增加经济投入提高预防水平而对于事故风险的削减无明显的效果,就应考虑与经济效益的结合问题。笔者提出安全投入数量限度问题和安全投入的阈值;并进一步建立安全效益评价模型和安全投入与安全指标的函数关系模型,对安全投入与安全水平(安全指标)、安全效益的关联关系进行了定性分析。     

城市消防站选址布局优化及对雄安新区的启示

安全是雄安新区建设发展的重要保障和基础底线。消防救援队伍,承担火灾防范、火灾扑救、抢险救援等工作,是提升新区城市救援能力,保障城市安全的有生力量。消防站是消防救援队伍工作（执勤备战）的场所,优化消防站选址布局意义重大。面向雄安新区消防安全提升和应急救援能力提升需求,针对我国目前消防站建设过程中覆盖率低及消防站建设投资巨大等问题,建立位置集合覆盖模型和最大覆盖模型联合使用的选址方法,增加补充性约束条件对模型进行改进,提出基于GIS的选址模型应用流程,并通过实际案例应用研究,提出雄安新区消防站规划建设的启示建议。     

AFRA – Heuristic expert system to assess the atmospheric risk of sulphide waste dumps

In 2006, an unprecedented atmospheric confined space accident took place in a sampling shed at the Sullivan Mine in Kimberley, British Columbia. This accident suggests that a risk assessment should be carried out on a regular basis at mine reclamation sites for many years after closure. In this paper, an Atmospheric Fuzzy Risk Assessment (AFRA) tool is described that can assess atmospheric risk given heuristic and measured data at such sites. It can also serve to transfer knowledge about atmospheric hazards in an enclosed structure. The system uses fuzzy logic to input and output information and to perform weighted inferencing. The paper describes the developmental process as well as system verification and validation based on a number of known test and reference waste dumps. AFRA is a heuristic expert system based on fuzzy logic and the first tool that was developed to assess the atmospheric risk of mine waste dumps. The atmospheric risk is estimated by fuzzy Mamdani system given the values of four major elements of risk comprising of: gas generation, gas emission, gas confinement, and human exposure. The ability of AFRA to adapt its risk assessment to different climate conditions is explained. There are many physical, chemical, and environmental factors which fluctuate over time affecting oxygen-depletion in waste dumps. AFRA can help mining engineers and mine managers recognize this type of danger when conducting a confined space inventory at a reclamation site.     

Prediction of occupational accident statistics and work time loss distributions using Bayesian analysis

This paper uses Bayesian analysis tools for the stochastic evaluation of work time losses due to occupational accidents in a workplace. Models are developed for accident frequencies, duration of recovery from an accident, and the worker unavailability. The unavailability statistics are hereby derived considering a two state stochastic model, to provide estimates for the expected work time losses over a base period of workplace operation. The above models are applied on real multiyear accident data collected from the Greek Petrochemical Industry.