化工园区多米诺事故风险评价与防控技术综述
2017-04-07贾梅生陈国华胡昆
贾梅生,陈国华,胡昆
(华南理工大学安全科学与工程研究所,广东 广州 510640)
化工园区多米诺事故风险评价与防控技术综述
贾梅生,陈国华,胡昆
(华南理工大学安全科学与工程研究所,广东 广州 510640)
定量风险评价是化工厂区或园区预防多米诺效应的主要方法,各类事故预防与控制措施的制定及其有效性检验也依赖于多米诺效应定量风险评价的结果。然而,由于概率风险表征、多米诺场景概率结构、火灾与爆炸致损概率等存在多种不一致或解释不明确的方法模型,使风险评价结果差异性很大,无法更进一步对各类多米诺事故预防与控制技术进行评估。本文首先界定了多米诺效应的基本概念和原理,然后对当前文献中相关多米诺效应定量风险评价方法进行深入剖析,明确了多米诺场景概率以及火灾热辐射、爆炸冲击波、爆炸碎片致损概率的计算方法。另外,还从阈值与安全距离、安全容量及布局优化、安全屏障与防火隔热层、安全与应急管理决策支持、安防与脆弱性分析及灾害衍生事故等方面对多米诺事故防控技术进行了评述。最后,指出以目标设备为核心的各类致损概率计算方法,即设备易损性,是后续研究的关注焦点。
多米诺效应;风险评价;致损概率;设备易损性;安全;化工园区
化工园区是石油和化学工业发展的必然趋势,然而集中的园区管理模式在带来产业规模效应的同时,也带来了较高的风险水平。多米诺效应问题的涌现即是由危险化学品储存与工艺装置规模的扩大,以及土地使用成本的提高而导致的,很多历史事故案例已经证明了多米诺效应发生的可能性及其非常严重的事故后果[1-9]。
英国健康安全委员会(HSC)下属重大危险源咨询委员会在1976年、1979年、1984年公布的重大危险源控制系列报告中首次明确了多米诺效应概念[10],并在1978年、1981年开展的坎维岛风险评价报告中进行了详细考虑[11]。历史上,1984年发生的墨西哥城国家石油公司LPG储运站连锁爆炸事故[11],极大地促进了多米诺效应在工业界与学术界的受关注程度,科技文献的涌现即始于此。1989年美国化工过程安全中心(CCPS)第一版化工过程定量风险评价指南[12],以及1999年荷兰应用科学研究院(TNO)定量风险评价紫皮书[13]已经对多米诺效应进行了相关考虑。
多米诺效应的官方条款主要见于欧盟委员会1996年《塞韦索II指令》第8条及其2012年更新版第9条,主要规定:企业需向管理部门告知信息,管理部门有辨识多米诺效应的责任,相关企业需相互合作、互通信息,在编制重大事故预防策略、安全管理系统、安全报告、内部应急预案时,需考虑多米诺效应的影响,并告知公众[14]。2012年,国务院安委办(2012)37号文件“关于进一步加强化工园区安全管理的指导意见”中也强调:化工园区需合理布局,预防连锁事故发生[15]。
1 多米诺效应定义与原理
众多多米诺效应的概念研究中,RENIERS和COZZANI的定义更具广泛适用性,其核心是事故扩展传播与后果影响扩大,包含3个基本要素:①初始事故场景及其物理影响,如火灾热辐射、爆炸冲击波、爆炸碎片等;②潜在的二次或一阶扩展事故场景,源于初始事故的扩展传播,危险化学品发生泄漏,使后果影响扩大;③后果影响扩大的目标设备或单元[16-19]。
另外,火灾热辐射、爆炸冲击波、爆炸碎片是三类主要的物理影响因素,即致损因子。火灾热辐射可导致目标设备的壳壁或结构材料发生高温强度下降,使压力储存的液化烃或工业气体的容器内压升高,同时火焰还有点火引燃的危险。爆炸冲击波可使目标设备发生屈服、倒塌、破裂、分解、整体位移、连接管线断裂、安全装置功能失效等,其传播路径还可产生二次碎片危险。爆炸碎片可穿透容器、切断支撑或管线,高温碎片也有点火或加热的危险,碎片来源包括容器爆炸破裂产生的小碎片、爆炸分解的零部件等。一般而言,有毒物质泄漏不会直接导致进一步危化品泄漏扩散、火灾、爆炸等场景,但可使现场操作人员中毒,丧失正常工作能力,间接导致多米诺效应。关于连续或批处理工艺装置的基本控制系统与安全功能系统的连锁效应导致的事故,一般不属于多米诺效应的研究范畴。
2 多米诺事故统计
事故统计分析可以明确多米诺场景的某些特征。2000年,KOURNIOTIS等[1]对207起化工事故统计发现,多米诺场景频率与危险化学品类型相关,蒸汽烃类较液体燃料更危险,并拟合了多米诺效应的社会风险曲线。2003年,RONZA等[2]统计了港口区的675起化工事故,发现直接从火灾到爆炸的扩展序列有30个,频率可达4.44×10–2。2008年,GÓMEZ-MARES等[3]统计了84起与喷射火相关的化工事故,指出喷射火有50%的可能性可以触发多米诺场景。2010年,DARBRA等[4]对与多米诺效应相关的225起化工事故,按阶段、区域、涉及危险化学品、原因、后果、场所、扩展序列等多个方面进行了详细统计分析,并绘制了社会风险曲线与扩展序列概率事件树图。类似研究还包括ABDOLHAMIDZADEH等[5]、CHEN等[6],HEMMATIAN等[8]的统计分析。另外,ZHANG 和ZHENG[7]对2006—2010年期间国内发生的1632起危险化学品事故统计发现,固定场所多米诺事故占13.2%,移动运输场所多米诺事故占9%。2015年,HEMMATIAN等[9]又专门对127起与沸腾液体扩展蒸汽云爆炸(BLEVE)相关的多米诺事故进行统计分析,绘制了扩展序列概率事件树图,并指出事故延时从一分钟到几小时不等,需重点关注应急疏散与消防安全距离。
3 定量风险评价方法
CCPS化工过程定量风险评价指出,处理多米诺效应有两种逻辑[12]:①针对源设备或单元,失效频率与事故场景概率不变,多米诺效应使事故后果影响扩大,技术方法是扩展事件树的输出;②针对目标设备或单元,多米诺效应使失效频率增加,事故后果影响不变,技术方法是将多米诺效应处理为事故树的一类外部事件输入。但无论哪种逻辑,均需首先计算火灾热辐射、爆炸冲击波、爆炸碎片的致损概率,或根据安全距离确定哪些目标设备或单元需纳入多米诺效应的考虑范围。另外,由于多米诺效应的扩展场景与扩展模式比较复杂,多米诺场景概率结构有各种不同的计算方法,即多米诺风险依多米诺场景概率结构不同而有所差异。表1对文献中构建的各种多米诺效应定量风险评价方法进行了对比分析。
3.1 多米诺场景概率结构
场景概率结构表达多米诺效应定义及其风险计算方法,多场景、高阶扩展、扩展协同效应、应急措施干预、动态过程是其技术难点。如表1所示,
即使在概率风险评价的框架内,且明确增加失效频率与扩大事故后果影响的逻辑,多数多米诺场景概率结构依然自成体系,如COZZANI等[16,24-25]的分析方法、ABDOLHAMIDZADEH等[26-27]的蒙特卡洛模拟方法、BERNECHEA等[31]的事件树方法以及KHAKZAD等[32-34]的贝叶斯网络方法。导致混乱的原因除相关技术难点外,另一个重要原因是多米诺风险评价对象不明确或多米诺场景不统一。一般化工过程定量风险评价选择最坏事故场景,但是多米诺效应由于扩展场景与扩展模式比较复杂,各种组合情况非常多。因此,首先需要确定一种原则或方法以统一多米诺场景的选择问题。
3.2 火灾热辐射致损概率
火灾热辐射致损概率主要包括两种计算方法:①比例方法[11,20-23,29-30];②Probit模型,即应急及时概率[16,24-28,32-34,41]。比例方法缺乏严格的理论基础,随机性表达不明确。Probit模型表达应急耗时小于事故延时的可能性,其中,事故自然延时计算是一种确定的保守拟合关系式,所以主要反映的是应急措施的随机性。但是对于指定的化工园区或企业及其应急预案,应急耗时的概率分布是不一样的,即需要调整Probit模型系数。此外,保守拟合关系式的应用也有限制范围,而且事故延时本质上也是随机的,因此,如何分析与表达事故延时的随机性也需要另作研究。
3.3 爆炸冲击波致损概率
Probit模型是爆炸冲击波致损概率的主要计算方法,其核心是根据目标设备破坏失效的严重程度定义致损概率,概率单位与冲击波峰值超压相关,主要反映的是目标设备抵抗冲击波攻击的能力,方法的精确性与统计样本数量以及统计样本之间的一致性有关。方法建立在远场假设基础上,忽略了冲击波与目标设备之间复杂的交互过程,不适用于近场范围的破坏失效。KHAN和ABBASI[21-23]的Probit模型是建立在EISENBERG等[42]模型基础上的,以有效压力代替峰值压力,其他参数不变,使不同类型的目标设备可以通过曳引系数不同取值进行区别。COZZANI和SALZANO[43-47]首先将目标设备分为常压储罐、压力容器、塔设备、辅助设备四类,然后定义各类设备破坏失效的严重程度,在75个文献样本的基础上进行统计分析得到对应Probit函数。ZHANG和JIANG[48]在COZZANI和SALZANO样本数据的基础上,重新定义目标设备破坏失效的严重程度,得到新的Probit模型。
3.4 爆炸碎片致损概率
爆炸碎片致损概率的计算方法相对比较复杂,按研究方法可以分为两类:①基于碎片抛射轨迹方程的蒙特卡洛模拟;②比例方法。按研究对象可以分为三类:①碎片产生概率或碎片初始参数的概率分布,主要包括碎片数量、大小、形状、初始速度、抛射角等;②碎片击中概率;③碎片穿透概率。2001年,HAUPTMANNS[49-50]首次通过碎片抛射轨迹方程的蒙特卡洛模拟得到了碎片的击中概率,之后涌现出很多以此方法为基础的研究,主要集中于击中概率定义、随机参数概率分布确定、参数影响分析等[51-62]。比例方法主要是GUBINELLI等[63]提出的以概率几何原理为基础的击中概率模型,TUGNOLI等[64]通过一个事故案例对该方法的有效性进行了分析与验证。关于碎片初始参数及其概率分布的确定包括GUBINELLI等[65-66]、MÉBARKI等[67]、孙东亮等[68]、TUGNOLI等[69]的研究。关于碎片穿透概率的研究主要是NGUYEN 等[70]提出的基于临界残余壁厚的方法,以及陈刚等[71]提出的基于临界剩余强度系数的方法,由于两种方法同样是以碎片抛射轨迹方程的蒙特卡洛模拟为基础的,因此,可以直接获得碎片击中概率与穿透概率的乘积。
表1 多米诺效应定量风险评价方法对比分析
3.5 概率风险表征
多米诺场景概率结构与火灾热辐射、爆炸冲击波、爆炸碎片的致损概率最终都要集成于概率风险的框架内,以概率风险的原理进行结果表征。对于化工过程定量风险评价而言,个人风险与社会风险是常用的风险指标与风险结果描述方法,个人风险是事故场景发生概率与事故场景物理影响下个体伤亡概率的数学乘积,社会风险计算依赖于个人风险结果与人员密度分布情况[12-13]。个体伤亡概率一般基于Probit模型进行确定[12-13]。COZZANI等[16]提出的多米诺效应定量风险评价方法即应用个人风险与社会风险描述风险评价结果。KOURNIOTIS等[1]、DARBRA等[4]、HEMMATIAN等[8]从事故统计角度绘制了多米诺效应的社会风险曲线,可以作为风险评价结果的准则曲线。然而,对于更多的多米诺效应定量风险评价方法而言,多米诺场景概率即直接表示多米诺效应风险。
4 事故预防与控制技术
目标设备或单元在致损因子影响下的破坏失效主要与设备之间的安全距离、初始事故场景致损因子强度以及目标设备的防护措施有关,而关于降低源设备失效频率或初始事故场景物理影响的安全措施,如结构完整性管理、火源控制、保护层分析(LOPA)等,由于其在一般的过程安全管理中均有考虑,所以不列入详细讨论范围,但是初始事故的预防与安全措施也会显著降低多米诺风险,而且更为重要。
4.1 阈值与安全距离、安全容量及布局优化
安全距离与安全容量是互补的、可以实现本质安全的多米诺事故防控技术,两者均与致损因子的阈值有关。阈值主要根据目标设备或单元类型及其破坏失效模式与严重程度进行确定,安全距离根据阈值与源设备事故场景特征确定,安全容量设置主要是限制事故场景物理响应范围。安全距离主要在工厂的布局设计阶段实现,而安全容量可以实现动态管理,并且两者均是布局优化与风险动态管理的基本参数。ALILECHE等[19]对标准与文献中关于多米诺效应的阈值与安全距离研究进行了详细评述。一般情况,通过实验或事故统计可以确定阈值,再根据事故后果物理影响的经验模型计算安全距离,但数据不一致仍然是常见的。对于爆炸冲击波而言:①缺乏对目标设备详细特征的描述;②破坏失效模式及其严重程度定义比较混乱。对于火灾热辐射而言,还容易忽略:①火灾场景不同致损机理,如接触点火或远距离热辐射;②火灾持续时间。相对比较全面细致的研究是COZZANI等[19,72]的推荐值,还从本质安全的角度,在事故后果物理影响经验模型的基础上,对指定事故场景的安全距离与安全容量关系进行了详细分析[38,73]。SPOELSTRA等[74]对荷兰国家标准PGS18、PGS19中LPG储罐区安全距离的确定方法进行了详细描述。
理论上,安全距离实现后,扩展场景或多米诺效应就不会发生了,但安全距离的确定仍有许多不确定或限制因素,只是达到了目前最好或最优水平,因此,布局优化有必要将多米诺风险作为一个目标函数,核心问题则是目标函数的构建及其求解方法。LEE等[75]、SO等[76]通过柯西最速下降法确定n个爆炸危险源的合理位置,以使构建的多米诺场景概率最低,前者针对新建厂区,后者针对扩建厂区。KHAN和AMYOTTE[77-78]构建的I2SI指数、TUGNOLI等构建的DHI指数[36-37]、COZZANI等[38]构建的DCP、DCA、UDI、TDI指数,均可作为布局优化的目标函数。JUNG等[79]、LÓPEZ-MOLINA等[80]、DE LIRA-FLORES等[81]将布局优化定义为混合整数非线性规划问题求解,不同之处是多米诺风险目标函数的构建与使用。BERNECHEA和ARNALDOS[82]、KHAKZAD和RENIERS[83]则通过对不同备选布局方案进行多米诺风险分析与评价,转化为多目标决策问题。
4.2 安全屏障与防火隔热层
安全屏障或安全装置、安全附件与LOPA中的保护层概念近似,是对相关硬件、软件、管理措施功能抽象的集合描述,从事故预防与风险管理的角度理解,既包括降低事故场景发生概率的措施,也包括减小事故后果影响的措施。LOPA指南将保护层分为8个层次[84],即工艺设计、基本过程控制系统、关键报警和人员干预、安全仪表功能、物理保护(释放设施,如安全阀、爆破片等)、释放后物理保护(如防火堤、防爆墙等)、厂区应急响应、周围社区应急响应。按作用方式与功能,它们又可分为主动的与被动的措施,以及阻止性的(释放前)与减缓性的(释放后)措施。LOPA指南并没有提供方法以评估保护层在降低多米诺风险方面的作用,或指出如何合理布置保护层以降低多米诺风险。LANDUCCI等[85]基于LOPA方法,得到了预防火灾触发多米诺场景时,各类安全屏障措施的有效性与要求时失效概率数据。JANSSENS等[86]通过元启发式算法,构建了一个合理布置安全屏障以减小火灾事故场景后果影响,增加扩展事故场景延时,降低多米诺风险的决策模型。
防火隔热层的概念与方法早于多米诺效应,主要指阻隔火焰直接接触影响或降低壳壁吸收热量速率以增加扩展事故场景延时的涂层或夹套,是一种被动的防火措施。BLEVE是一种后果极其严重的火灾导致的液化烃压力容器扩展事故场景,尤其在运输行业,由于经常不可避免地经过一些人员密集场所,风险非常高,因此,早期有许多关于防火隔热层的设计与实验研究,但是其应用也伴随着腐蚀检测与消防射流冲击的问题[11]。LANDUCCI等[87-88]通过对LPG储罐的实验与有限元模拟,分析验证了防火隔热层在增加BLEVE事故场景延时方面的有效性,实验材料包括环氧树脂发泡材料、蛭石、矿物纤维棉、水泥基无机材料,并提出3个指标以评估和指导设计。PALTRINIERI等[89]进一步指出,防火隔热层的应用可以使LPG供应链的个人风险降低一个数量级,使社会风险的期望值提高50%。DI PADOVA等[90]、TUGNOLI等[91]基于风险矩阵原理提出一个防火分区辨识方法,以合理应用防火隔热层,降低火灾触发多米诺效应风险。此外,由于防火隔热层也具有一定的结构强度,因此,可以探索研究新的过程设备设计方法,或者开发设计新的保护层材料或结构,以综合考虑火灾热辐射、爆炸冲击波、爆炸碎片的致损作用。
4.3 安全与应急管理决策支持
风险分析与评价本质上即是一种分级的安全管理思想与方法,多米诺效应的风险分析与评价也不例外。应急概念则是一种事故控制思想,即假设事故场景发生后如何采取措施以减小事故后果影响,多米诺效应概念也包含初始事故场景的假设,所以应急研究对于预防与控制多米诺事故非常重要。KHAN和ABBASI[92]在多米诺效应定量风险评价方法研究的基础上,开发了DOMIFECT软件,以辅助决策。COZZANI等[93]将多米诺效应定量风险评价方法集成于Aripar-GIS软件。RENIERS等[94-95]提出了集成危险与可操作性分析(HAZOP)、What-If分析、风险矩阵的Hazwim框架,使化工园区的企业之间可以相互合作,共同预防外部多米诺效应。之后,RENIERS和DULLAERT[96-97]从化工园区范围内扩展序列优先级的角度开发了DomPrevPlanning决策支持软件。另外,RENIERS等[98-99]还应用博弈论方法建模分析了化工园区企业在预防多米诺效应时的投资决策过程。ZHANG和CHEN[100]提出了限制化工园区或企业危险源扩展网络的离散孤岛模型。KHAKZAD等[33]应用贝叶斯网络与冲突分析方法建模了危化品安全容量的配置问题。关于应急管理,ZHOU等[35,101]分别应用事件序列图法与Petri网络建模分析了应急措施干预对火灾触发多米诺场景的影响,以及大规模并发火灾的应急响应问题。
4.4 安防与脆弱性分析及灾害衍生事故
触发化工园区或企业多米诺效应的原因,除常见的火灾、爆炸初始事故场景外,蓄意或恐怖袭击以及自然灾害也受到广泛关注。RENIERS等[102]通过抽象化工园区范围内所有危险源及其相互影响关系,首次提出了分离节点网络的概念,以辅助安防管理。之后,RENIERS和AUDENAERT[103]在分离节点网理络研究的基础上,提出单元装置脆弱性分析方法,以确定预防恐怖袭击时单元装置的防护优先级。RENIERS等[104]还应用物理效应衰减与恢复能力的概念对化工园区与企业的安防管理进行了研究,指出分离节点网络的方法可以提高化工园区遭受攻击后的快速恢复能力。LANDUCCI等[105]基于TNT当量法研究分析了自制爆炸装置触发多米诺场景时单元设施的脆弱性问题。关于灾害衍生事故(NaTech),COZZANI等[24]在多米诺效应定量风险评价方法的基础上,构建了NaTech定量风险评价方法。
5 研究展望
(1)由于多米诺效应概念还有许多争议,包括范围界定、场景定义等,使相关政策、法规、标准的制定以及定量风险评价方法的构建存在很多不一致或不相容的解释与结果,所以迫切需要在多米诺效应概念范围以及场景选择方面达成一种共识,以更好地促进工业实践。
(2)火灾热辐射、爆炸冲击波、爆炸碎片致损概率的计算方法与结果构成多米诺效应定量风险评价框架的基石,当前选择的比例方法或Probit函数,是由于缺乏过程设备易损性基础数据,为简化定量风险评价过程复杂性与运算量而进行的折中,因此,需要探索研究新的、高可靠度、高精确度的致损概率计算方法,即设备易损性分析方法,并且在过程设备设计的交付文档中提供易损性数据及其详细说明。
(3)在过程设备设计阶段进行的致损概率计算或易损性分析,可以提高多米诺效应定量风险评价的有效性,使布局优化、安全屏障布置、安全与应急管理决策支持更可靠,也可为过程设备的本质安全设计及评估提供理论与方法支持,以更好地从本质安全的角度预防与控制多米诺效应。
(4)可以探索研究新的过程设备设计方法,或开发新的保护层材料或结构,以综合考虑火灾热辐射、爆炸冲击波、爆炸碎片的致损作用。
[1]KOURNIOTIS S P,KIRANOUDIS C T,MARKATOS N C.Statistical analysis of Domino chemical accidents[J].Journal of Hazardous Materials,2000,71(1/2/3):239-252.
[2]RONZA A,FÉLEZ S,DARBRA R M,et al.Predicting the frequency of accidents in port areas by developing event trees from historical analysis[J].Journal of Loss Prevention in the Process Industries,2003,16(6):551-560.
[3]GÓMEZ-MARES M,ZÁRATE L,CASAL J.Jet fires and the Domino effect[J].Fire Safety Journal,2008,43(8):583-588.
[4]DARBRA R M,PALACIOS A,CASAL J.Domino effect in chemical accidents:main features and accident sequences[J].Journal of Hazardous Materials,2010,183(1/2/3):565-573.
[5]ABDOLHAMIDZADEH B,ABBASI T,RASHTCHIAN D,et al.Domino effect in process-industry accidents ——an inventory of past events and identification of some patterns[J].Journal of Loss Prevention in the Process Industries,2011,24(5):575-593.
[6]CHEN Y T,ZHANG M G,GUO P J,et al.2012 International symposium on safety science and technology:investigation and analysis of historical Domino effects statistic[J].Procedia Engineering,2012:45,152-158.
[7]ZHANG H D,ZHENG X P.Characteristics of hazardous chemical accidents in China:a statistical investigation[J].Journal of Loss Prevention in the Process Industries,2012,25(4):686-693.
[8]HEMMATIAN B,ABDOLHAMIDZADEH B,DARBRA R M,et al.The significance of Domino effect in chemical accidents[J].Journal of Loss Prevention in the Process Industries,2014,29:30-38.
[9]HEMMATIAN B,PLANAS E,CASAL J.Fire as a primary event of accident Domino sequences:the case of BLEVE[J].Reliability Engineering & System Safety,2015,139:141-148.
[10]Advisory Committee on Major Hazards.The control of major hazards - first/ second/ third report[R].London:Healthy and Safety Commission,1976/1979/1984.
[11]MANNAN S.Lees’loss prevention in the process industries:hazard identification,assessment and control[M].3rd ed.Amsterdam:Elsevier,2005.
[12]Center for Chemical Process Safety.Guidelines for chemical process quantitative risk analysis[M].2nd ed.New York:American Institute of Chemical Engineers,2000.
[13]UIJT DE HAAG P A M,ALE B J M.Guideline for quantitative risk assessment ——‘purple book’[M].The Hague:Advisory Council on Dangerous Substances,2005.
[14]Seveso - III(Directive 2012/18/EU)[EB/OL].European Parliament and Council,2012.[2016-09-05]. http://eur-lex.europa.eu/ legalcontent/EN/TXT/?uri=CELEX:32012L0018.
[15]国务院安委会办公室.关于进一步加强化工园区安全管理的指导意见[EB/OL]. 2012. [2016-09-05]. http://www.chinasafety. gov.cn/newpage/Contents/Channel_4976/2012/0809/174959/conten t_174959.htm.The Work Safety Committee Office of The State Council.Guidance on strengthening the safety management of chemical parks[EB/OL].2012.[2016-09-05].http://www.chinasafety.gov.c n/newpage/Contents/Channel_4976/2012/0809/174959/content_1749 59.htm.
[16]COZZANI V,GUBINELLI G,ANTONIONI G,et al.The assessment of risk caused by Domino effect in quantitative area risk analysis[J].Journal of Hazardous Materials,2005,127(1/2/3):14-30.
[17]RENIERS G.An external Domino effects investment approach to improve cross-plant safety within chemical clusters[J].Journal of Hazardous Materials,2010,177(1/2/3):167-174.
[18]RENIERS G,COZZANI V.Domino effects in the process industries——modelling,prevention and managing[M].Amsterdam:Elsevier,2013.
[19]ALILECHE N,COZZANI V,RENIERS G,et al.Thresholds for Domino effects and safety distances in the process industry:a review of approaches and regulations[J].Reliability Engineering & System Safety,2015,143(s1):74-84.
[20]BAGSTER D F,PITBLADO R M.The estimation of Domino incident frequencies-an approach[J]. Process Safety and Environmental Protection,1991,69(4):195-199.
[21]KHAN F I,ABBASI S A.Models for Domino effect analysis in chemical process industries[J].Process Safety Progress,1998,17(2):107-123.
[22]KHAN F I,ABBASI S A.Studies on the probabilities and likely impacts of chains of accident(Domino effect)in a fertilizer industry[J].Process Safety Progress,2000,19(1):40-56.
[23]KHAN F I,ABBASI S A.An assessment of the likelihood of occurrence,and the damage potential of Domino effect(chain of accidents)in a typical cluster of industries[J].Journal of Loss Prevention in the Process Industries,2001,14(4):283-306.
[24]COZZANI V,ANTONIONI G,LANDUCCI G,et al.Quantitative assessment of Domino and NaTech scenarios in complex industrial areas[J].Journal of Loss Prevention in the Process Industries,2014,28(s1):10-22.
[25]ANTONIONI G,SPADONI G,COZZANI V.Application of Domino effect quantitative risk assessment to an extended industrial area[J].Journal of Loss Prevention in the Process Industries,2009,22(5):614-624.
[26]ABDOLHAMIDZADEH B,ABBASI T,RASHTCHIAN D,et al.A new method for assessing Domino effect in chemical process industry[J].Journal of Hazardous Materials,2010,182(1/2/3):416-426.
[27]RAD A,ABDOLHAMIDZADEH B,ABBASI T,et al.FREEDOM II:an improved methodology to assess Domino effect frequency using simulation techniques[J].Process Safety and Environmental Protection,2014,92(6):714-722.
[28]KADRI F,CHATELET E,CHEN G P.Method for quantitative assessment of the Domino effect in industrial sites[J].Process Safety and Environmental Protection,2013,91(6):452-462.
[29]张新梅.化工园区事故多米诺效应风险仿真原理及应用研究[D].广州:华南理工大学,2009.ZHANG X M.Study on simulation principle of accident risk caused by Domino effect and its application in chemical industrial parks[D].Guangzhou:South China University of Technology,2009.
[30]ZHANG X M,CHEN C.Mechanism analysis and risk assessment ofescalation scenario in chemical industry zones[J].Process Safety and Environmental Protection,2013,91(1/2):79-85.
[31]BERNECHEA E J,ANTONIO VILCHEZ J,ARNALDOS J.A model for estimating the impact of the Domino effect on accident frequencies in quantitative risk assessments of storage facilities[J].Process Safety and Environmental Protection,2013,91(6):423-437.
[32]KHAKZAD N,KHAN F,AMYOTTE P,et al.Domino effect analysis using Bayesian networks[J].Risk Analysis,2013,33(2):292-306.
[33]KHAKZAD N,KHAN F,AMYOTTE P,et al.Risk management of Domino effects considering dynamic consequence analysis[J].Risk Analysis,2014,34(6):1128-1138.
[34]KHAKZAD N.Application of dynamic Bayesian network to risk analysis of Domino effects in chemical infrastructures[J].Reliability Engineering & System Safety,2015,138:263-272.
[35]ZHOU J F,RENIERS G,KHAKZAD N.Application of event sequence diagram to evaluate emergency response actions during fire-induced Domino effects[J].Reliability Engineering & System Safety,2016,150:202-209.
[36]TUGNOLI A,KHAN F,AMYOTTE P,et al.Safety assessment in plant layout design using indexing approach:implementing inherent safety perspective:Part 1 guideword applicability and method description[J].Journal of Hazardous Materials,2008,160(1):100-109.
[37]TUGNOLI A,KHAN F,AMYOTTE P,et al.Safety assessment in plant layout design using indexing approach:implementing inherent safety perspective:Part 2 - Domino hazard index and case study[J].Journal of Hazardous Materials,2008,160(1):110-121.
[38]COZZANI V,TUGNOLI A,SALZANO E.The development of an inherent safety approach to the prevention of Domino accidents[J].Accident Analysis & Prevention,2009,41(6):1216-1227.
[39]KHAKZAD N,RENIERS G.Using graph theory to analyze the vulnerability of process plants in the context of cascading effects[J].Reliability Engineering & System Safety,2015,143(s1):63-73.
[40]NI Z J,WANG Y Z,YIN Z G.Relative risk model for assessing Domino effect in chemical process industry[J].Safety Science,2016,87:156-166.
[41]LANDUCCI G,GUBINELLI G,ANTONIONI G,et al.The assessment of the damage probability of storage tanks in Domino events triggered by fire[J].Accident Analysis & Prevention,2009,41(6):1206-1215.
[42]EISENBERG N A,LYNCH C J,BREEDING R J.Vulnerability model:a simulation system for assessing damage resulting from marine spills[R].Springfield:National Technical Information Service,1975.
[43]COZZANI V,SALZANO E.The quantitative assessment of Domino effects caused by overpressure:Part I.probit models[J].Journal of Hazardous Materials,2004,107(3):67-80.
[44]COZZANI V,SALZANO E.The quantitative assessment of Domino effect caused by overpressure:Part II.case studies[J].Journal of Hazardous Materials,2004,107(3):81-94.
[45]COZZANI V,SALZANO E.Threshold values for Domino effects caused by blast wave interaction with process equipment[J].Journal of Loss Prevention in the Process Industries,2004,17(6):437-447.
[46]SALZANO E,COZZANI V.The analysis of Domino accidents triggered by vapor cloud explosions[J].Reliability Engineering & System Safety,2005,90(2/3):271-284.
[47]SALZANO E,COZZANI V.A fuzzy set analysis to estimate loss intensity following blast wave interaction with process equipment[J].Journal of Loss Prevention in the Process Industries,2006,19(4):343-352.
[48]ZHANG M G,JIANG J C.An improved Probit method for assessment of Domino effect to chemical process equipment caused by overpressure[J].Journal of Hazardous Materials,2008,158(2/3):280-286.
[49]HAUPTMANNS U.A procedure for analyzing the flight of missiles from explosions of cylindrical vessels[J].Journal of Loss Prevention in the Process Industries,2001,14(5):395-402.
[50]HAUPTMANNS U.A Monte-Carlo based procedure for treating the flight of missiles from tank explosions[J].Probabilistic Engineering Mechanics,2001,16(4):307-312.
[51]PULA R,KHAN F I,VEITCH B,et al.A model for estimating the probability of missile impact:missiles originating from bursting horizontal cylindrical vessels[J].Process Safety Progress,2007,26(2):129-139.
[52]张新梅,陈国华.化工罐区爆炸碎片多米诺效应影响概率计算模型[J].化工学报,2008,59(11):2946-2953.ZHANG X M,CHEN G H.Impact probability of Domino effect caused by explosion fragments in chemical tank area[J].CIESC Journal,2008,59(11):2946-2953.
[53]ZHANG X M,CHEN G H.The analysis of Domino effect impact probability triggered by fragments[J].Safety Science,2009,47(7):1026-1032.
[54]MÉBARKI A,NGUYEN Q B,MERCIER F.Structural fragments and explosions in industrial facilities:Part II projectile trajectory and probability of impact[J].Journal of Loss Prevention in the Process Industries,2009,22(4):417-425.
[55]钱新明,徐亚博,刘振翼.球罐BLEVE碎片抛射的Monte-Carlo分析[J].化工学报,2009,60(4):1057-1061.QIAN X M,XU Y B,LIU Z Y.Monte-Carlo analysis of fragments projectile from spherical tank BLEVE[J].CIESC Journal,2009,60(4):1057-1061.
[56]钱新明,徐亚博,刘振翼.球罐BLEVE碎片抛射的危害性研究[J].高压物理学报,2009,23(5):389-394.QIAN X M,XU Y B,LIU Z Y.Research on hazard of fragments from spherical tank BLEVE[J].Chinese Journal of High Pressure Physics,2009,23(5):389-394.
[57]陈刚,朱霁平,武军,等.爆炸球罐尺寸对抛射碎片击中相邻罐体概率的影响[J].化工学报,2010,61(6):1599-1604.CHEN G,ZHU W P,WU J,et al.Influence of size of bursting spherical tank on impact probability between fragments and adjacent vessels[J].CIESC Journal,2010,61(6):1599-1604.
[58]陈刚,朱霁平,武军,等.化工储罐间距和体积对爆炸碎片多米诺效应概率的影响[J].火灾科学,2011,20(1):37-42.CHEN G,ZHU W P,WU J,et al.Impact of gap and volume of storage tanks on Domino effect of explosion fragments[J].Fire Safety Science,2011,20(1): 37-42.
[59]SUN D L,JIANG J C,ZHANG M G,et al.Parametric approach of the Domino effect for structural fragments[J].Journal of LossPrevention in the Process Industries,2012,25(1):114-126.
[60]SUN D L,JIANG J C,ZHANG M G,et al.Influence of the source size on Domino effect risk caused by fragments[J].Journal of Loss Prevention in the Process Industries,2015,35:211-223.
[61]SUN D L,JIANG J C,ZHANG M G,et al.Investigation of multiple Domino scenarios caused by fragments[J].Journal of Loss Prevention in the Process Industries,2016,40:591-602.
[62]LISI R,CONSOLO G,MASCHIO G,et al.Estimation of the impact probability in Domino effects due to the projection of fragments[J].Process Safety and Environmental Protection,2015,93:99-110.
[63]GUBINELLI G,ZANELLI S,COZZANI V.A simplified model for the assessment of the impact probability of fragments[J].Journal of Hazardous Materials,2004,116(3):175-187.
[64]TUGNOLI A,MILAZZO M F,LANDUCCI G,et al.Assessment of the hazard due to fragment projection:a case study[J].Journal of Loss Prevention in the Process Industries,2014,28(SI):36-46.
[65]GUBINELLI G,COZZANI V.Assessment of missile hazards:evaluation of the fragment number and drag factors[J].Journal of Hazardous Materials,2009,161(1):439-449.
[66]GUBINELLI G,COZZANI V.Assessment of missile hazards:identification of reference fragmentation patterns[J].Journal of Hazardous Materials,2009,163(2/3):1008-1018.
[67]MÉBARKI A,MERCIER F,NGUYEN Q B,et al.Structural fragments and explosions in industrial facilities:Part I probabilistic description of the source terms[J].Journal of Loss Prevention in the Process Industries,2009,22(4):408-416.
[68]孙东亮,蒋军成,张明广,等.基于最大熵原理的卧罐爆炸碎片数量概率分布[J].化工学报,2011,62(s1):219-224.SUN D L,JIANG J C,ZHANG M G,et al.Probabilistic distribution of fragments number from explosion of horizontal tank based on maximum entropy principle[J].CIESC Journal,2011,62(S1):219-224.
[69]TUGNOLI A,GUBINELLI G,LANDUCCI G,et al.Assessment of fragment projection hazard:probability distributions for the initial direction of fragments[J].Journal of Hazardous Materials,2014,279:418-427.
[70]NGUYEN Q B,MÉBARKI A,SAADA R A,et al.Integrated probabilistic framework for Domino effect and risk analysis[J].Advances in Engineering Software,2009,40(9):892-901.
[71]陈刚,朱霁平,武军,等.储罐爆炸抛射碎片对目标储罐的破坏概率研究[J].防灾减灾工程学报,2012,32(2):216-222.CHEN G,ZHU W P,WU J,et al.Damage probability model of target storage tank impacted by explosion fragments[J].Journal of Disaster Prevention and Mitigation Engineering,2012,32(2):216-222.
[72]COZZANI V,GUBINELLI G,SALZANO E.Escalation thresholds in the assessment of Domino accidental events[J].Journal of Hazardous Materials,2006,129(1/2/3):1-21.
[73]COZZANI V,TUGNOLI A,SALZANO E.Prevention of Domino effect:from active and passive strategies to inherently safer design[J]. Journal of Hazardous Materials,2007,139(2):209-219.
[74]SPOELSTRA M,MAHESH S,KOOI E,et al.Domino effects at LPG and propane storage sites in the Netherlands[J].Reliability Engineering & System Safety,2015,143(s1):85-90.
[75]LEE J Y,LEE J W,KO J W,et al.Optimization for allocating the explosive facilities in order to minimize the Domino effect using nonlinear programming[J].Korean Journal of Chemical Engineering,2005,22(5):649-656.
[76]SO W,KIM Y H,LEE C J,et al.Optimal layout of additional facilities for minimization of Domino effects based on worst-case scenarios[J].Korean Journal of Chemical Engineering,2011,28(3):656-666.
[77]KHAN F I,AMYOTTE P R.Integrated inherent safety index(I2SI):a tool for inherent safety evaluation[J].Process Safety Progress,2004,23(2):136-148.
[78]KHAN F I,AMYOTTE P R.I2SI:a comprehensive quantitative tool for inherent safety and cost evaluation[J].Journal of Loss Prevention in the Process Industries,2005,18(4/5/6):310-326.
[79]JUNG S H,NG D,DIAZ-OVALLE C,et al.New approach to optimizing the facility sitting and layout for fire and explosion scenarios[J].Industrial & Engineering Chemistry Research,2011,50:3928-3937.
[80]LOPEZ-MOLINA A,VAZQUEZ-ROMAN R,MANNAN M S,et al.An approach for Domino effect reduction based on optimal layouts[J].Journal of Loss Prevention in the Process Industries,2013,26(s1):887-894.
[81]DE LIRA-FLORES J,VAZQUEZ-ROMAN R,LOPEZ-MOLINA A,et al.A MINLP approach for layout designs based on the Domino hazard index[J].Journal of Loss Prevention in the Process Industries,2014,30:219-227.
[82]BERNECHEA E J,ARNALDOS J.Optimizing the design of storage facilities through the application of ISD and QRA[J].Process Safety and Environmental Protection,2014,92(6):598-615.
[83]KHAKZAD N,RENIERS G.Risk-based design of process plants with regard to Domino effects and land use planning[J].Journal of Hazardous Materials,2015,299:289-297.
[84]Center for Chemical Process Safety.Layer of protection analysis -simplified process risk assessment[M].New York:American Institute of Chemical Engineers,2001.
[85]LANDUCCI G,ARGENTI F,TUGNOLI A,et al.Quantitative assessment of safety barrier performance in the prevention of Domino scenarios triggered by fire[J].Reliability Engineering & System Safety,2015,143(SI):30-43.
[86]JANSSENS J,TALARICO L,RENIERS G,et al.A decision model to allocate protective safety barriers and mitigate Domino effects[J].Reliability Engineering & System Safety,2015,143(s1):44-52.
[87]LANDUCCI G,MOLAG M,REINDERS J,et al.Experimental and analytical investigation of thermal coating effectiveness for 3m3LPG tanks engulfed by fire[J].Journal of Hazardous Materials,2009,161(2/3):1182-1192.
[88]LANDUCCI G,MOLAG M,COZZANI V.Modeling the performance of coated LPG tanks engulfed in fires[J].Journal of Hazardous Materials,2009,172(1):447-456.
[89]PALTRINIERI N,LANDUCCI G,MOLAG M,et al.Risk reduction in road and rail LPG transportation by passive fire protection[J].Journal of Hazardous Materials,2009,167(1/2/3):332-344.
[90]DI PADOVA A,TUGNOLI A,COZZANI V,et al.Identification of fireproofing zones in Oil&Gas facilities by a risk-based procedure[J].Journal of Hazardous Materials,2011,191(1/2/3):83-93.
[91]TUGNOLI A,COZZANI V,DI PADOVA A,et al.Mitigation of fire damage and escalation by fireproofing:a risk-based strategy[J].Reliability Engineering & System Safety,2012,105(s1):25-35.
[92]KHAN F I,ABBASI S A.DOMIFFECT(DOMIno eFFECT):user-friendly software for Domino effect analysis[J].Environmental Modelling & Software,1998,13(2):163-177.
[93]COZZANI V,ANTONIONI G,SPADONI G.Quantitative assessment of Domino scenarios by a GIS-based software tool[J].Journal of Loss Prevention in the Process Industries,2006,19(5):463-477.
[94]RENIERS G L L,DULLAERT W,ALE B J M,et al.The use of current risk analysis tools evaluated towards preventing external Domino accidents[J].Journal of Loss Prevention in the Process Industries,2005,18(3):119-126.
[95]RENIERS G L L,DULLAERT W,ALE B J M,et al.Developing an external Domino accident prevention framework:Hazwim[J].Journal of Loss Prevention in the Process Industries,2005,18(3):127-138.
[96]RENIERS G L L,DULLAERT W.DomPrevPlanning©:user-friendly software for planning Domino effects prevention[J]. Safety Science,2007,45(10):1060-1081.
[97]RENIERS G L L,DULLAERT W.Knock-on accident prevention in a chemical cluster[J].Expert Systems with Applications,2008,34(1):42-49.
[98]RENIERS G,DULLAERT W,SOUDAN K.Domino effects within a chemical cluster:a game-theoretical modeling approach by using Nash-equilibrium[J].Journal of Hazardous Materials,2009,167(1/2/3):289-293.
[99]RENIERS G,CUYPERS S,PAVLOVA Y.A game-theory based Multi-plant Collaboration Model(MCM)for cross-plant prevention in a chemical cluster[J].Journal of Hazardous Materials,2012,209/210:164-176.
[100]ZHANG X M,CHEN G H.Modeling and algorithm of Domino effect in chemical industrial parks using discrete isolated island method[J].Safety Science,2011,49(3):463-467.
[101]ZHOU J F,RENIERS G.Petri-net based simulation analysis for emergency response to multiple simultaneous large-scale fires[J].Journal of Loss Prevention in the Process Industries,2016,40:554-562.
[102]RENIERS G L L,DULLAERT W,AUDENAERT A,et al.Managing Domino effect-related security of industrial areas[J].Journal of Loss Prevention in the Process Industries,2008,21(3):336-343.
[103]RENIERS G L L,AUDENAERT A.Preparing for major terrorist attacks against chemical clusters:intelligently planning protection measures w.r.t. Domino effects[J],Process Safety and Environmental Protection,2014,92(6):583-589.
[104]RENIERS G L L,SOERENSEN K,KHAN F,et al.Resilience of chemical industrial areas through attenuation-based security[J]. Reliability Engineering & System Safety,2014,131:94-101.
[105]LANDUCCI G,RENIERS G,COZZANI V,et al.Vulnerability of industrial facilities to attacks with improvised explosive devices aimed at triggering Domino scenarios[J].Reliability Engineering & System Safety,2015,143(s1):53-62.
Review of risk assessment and pre-control of Domino effect in Chemical Industry Park
JIA Meisheng,CHEN Guohua,HU Kun
(Institute of Safety Science and Engineering,South China University of Technology,Guangzhou 510640,Guangdong,China)
Quantitative risk assessment of Domino effects(DMQRA)is the key method to be used to prevent Domino accidents in chemical plants or parks. Also,the various pre-control measures and their availability evaluations are based on the result of DMQRA. However,there are many inconsistent or unclear interpreted modes and methods for presentation of probability risk,probability structure of Domino scenarios and damage probabilities of target equipment exposed to fire or blast such that the obtained DMQRA results are in a great variability. Moreover,it is hard to evaluate the specific pre-control measures. In this paper,firstly,the concept and basic principles of Domino effect are clearly defined. Then through carefully analyzing of the various DMQRA methods in literature,the probability structure of Domino scenario and damage probabilities of target equipment attacked by fire thermal radiation,blast wave overpressure and blast fragments are outlined clearly. In addition,damage thresholds,safety distances,safety inventories,layout optimization,safety barriers,thermal insulations,decision supporting for safety and emergency management,security,resilience and natural hazards triggering for domino effect are reviewed in detail. Finally,it is figured out that the advanced methods for calculating of damage probabilities of target equipment exposed to fire,blast and othertriggering vectors,i.e. equipment vulnerability,will be focused upon in the future.
Domino effect;risk assessment;damage probability;equipment vulnerability;safety;Chemical Industry Park
TQ086;X937
A
1000–6613(2017)04–1534–10
10.16085/j.issn.1000-6613.2017.04.050
2016-09-07;修改稿日期:2016-11-04。
国家自然科学基金(21576102)及国家重点研发计划(2016YFC0801500)项目。
贾梅生(1988—),男,博士研究生。联系人:陈国华,教授,博士生导师,从事工业安全与风险评价研究。E-mail:mmghchen @scut.edu.cn。