基于自适应重要抽样法非能动系统功能故障概率评估

针对非能动系统功能故障概率评估,提出一种新的自适应重要抽样方法。这种方法先对失效域进行预抽样,然后拟合出失效域中样本分布的密度函数,以之作为重要抽样密度函数。以1000 MW非能动先进压水堆(AP1000)非能动余热排出系统为研究对象,考虑模型和输入参数的不确定性,将响应面法和自适应重要抽样法相结合,对其进行功能故障概率评估。结果表明:与传统的概率评估方法相比,自适应重要抽样法具有较高的计算效率,同时又能保证很高的计算精度。     

XAPR中破口失水事故下堆芯自然循环冷却能力的功能可靠性研究

功能失效是导致自然循环系统运行失效的重要因素，需在其可靠性分析中予以考虑。针对多维不确定性参数及小功能失效概率问题，提出了一种将改进响应面法及重要抽样子集模拟法相结合的功能可靠性分析方法。以西安脉冲堆（XAPR）堆池水自然循环冷却为例，结合中破口失水事故，考虑输入参数及模型的不确定性，对其进行了功能可靠性评估和灵敏度分析。结果表明：XAPR堆芯自然循环功能失效概率为3.796×10^-3，需充分考虑系统功能的可靠性。本文方法具有较高的计算效率，同时又能保证很高的计算精度，对XAPR堆芯自然循环非线性功能函数具有很强的适应性。     

基于最优化线抽样法的XAPR自然循环功能失效研究与分析

针对多维不确定性参数、小失效概率的功能可靠性分析，提出了一种优化线抽样的可靠性分析方法。该方法采用遗传算法求解约束条件的优化模型来寻求最优化重要方向，进而得到失效概率的高效估计。以西安脉冲堆（XAPR）自然循环冷却堆芯能力的可靠性评价为例，考虑模型与输入参数的不确定性，对中破口失水事故下的自然循环功能失效概率进行了量化分析。结果表明：与其他概率评估方法相比，本文方法具有很高的计算效率，同时又能保证很好的计算精度；对隐式非线性的功能可靠性分析是有效可行的，具有很强的适应性。     

XAPR中破口失水事故下堆芯自然循环冷却能力的功能可靠性研究

功能失效是导致自然循环系统运行失效的重要因素,需在其可靠性分析中予以考虑。针对多维不确定性参数及小功能失效概率问题,提出了一种将改进响应面法及重要抽样子集模拟法相结合的功能可靠性分析方法。以西安脉冲堆(XAPR)堆池水自然循环冷却为例,结合中破口失水事故,考虑输入参数及模型的不确定性,对其进行了功能可靠性评估和灵敏度分析。结果表明:XAPR堆芯自然循环功能失效概率为3.796×10~(-3),需充分考虑系统功能的可靠性。本文方法具有较高的计算效率,同时又能保证很高的计算精度,对XAPR堆芯自然循环非线性功能函数具有很强的适应性。     

灵敏度分析方法在非能动系统可靠性研究中的应用

在研究核电站安全时,热工水力非能动系统的可靠性研究基于所建立的热工水力学数值模型。模型通常极其复杂,具有多个输入参数,且输入参数具有不确定性,对模型输出的不确定性的影响又各不相同。灵敏度分析的目的是将各参数对模型输出的不确定性的影响进行排序,找出显著的影响参数。本文首先描述灵敏度分析的方法,然后应用秩转换回归分析方法计算HTR-10余热排出系统模型各参数的灵敏度,找出关键影响因素,将模型简化,并对简化模型应用响应面方法计算了失效概率。简化模型算得的失效概率与原模型的很接近。     

核电厂非能动余热排出系统的失效概率评价

先进核电厂设计中大量采用非能动安全系统提高反应堆安全性。但目前尚无系统性评价非能动系统的成熟方法,而且概率安全评价(PSA)也未考虑非能动系统自然循环现象不确定性导致的功能失效。在欧盟非能动系统可靠性评价研究项目(RMPS)研究成果的基础上,以压水堆二次侧非能动余热排出系统(PRS)为研究对象,基于统计学和热工水力计算确定了影响性能的参数重要度,进而利用蒙特卡罗抽样和响应面分析对全厂断电事故下的PRS自然循环失效概率进行了量化分析评价。初步评价结果表明:非能动系统功能失效概率为2.14×10-3,在PSA中应当充分考虑各种非能动系统的功能失效。本文的评价方法还可以为非能动安全系统设计优化提供支持。     

基于子集模拟法非能动系统功能故障概率评估

针对非能动系统多维不确定性参数和小功能故障概率问题,提出基于马尔可夫链蒙特卡罗子集模拟的可靠性分析方法。该方法通过引入适当的中间失效事件,将小功能故障概率表达为一系列较大的中间失效事件条件概率乘积的形式,进而利用马尔可夫链模拟的条件样本点来计算条件失效概率。以AP1000非能动余热排出系统为研究对象,考虑热工水力学模型和输入参数的不确定性,对其进行功能故障概率评估。结果表明：与其它概率评估方法相比,子集模拟法具有较高的计算效率,同时又能保证很高的计算精度;对非能动安全系统非线性功能函数有很强的适应性。     

基于PSO优化神经网络响应面技术的非能动系统可靠性分析

在非能动可靠性分析数学模型的基础上，结合某型核动力装置非能动余热排出系统原理性试验装置和改进的热工水力程序的运行数据，识别了输入参数的不确定性，比较了不同神经网络响应面技术替代热工水力程序的精度和优度，分析了粒子群优化算法（PSO）优化神经网络响应面分类准确率。数值结果表明，该响应面具有较高的拟合优度，且能够较为准确的对非能动系统系统可靠性进行判定。      

基于响应面拟合方法中国铅基研究实验堆非能动余热排出系统可靠性分析

非能动系统已广泛地应用于新一代堆的设计中,其可靠性分析成为新型反应堆概率安全评价(Probabilistic Safety Analysis,PSA)的重要内容。本文提出一种用于非能动系统可靠性分析的响应面拟合方法,并应用于中国铅基研究实验堆反应堆容器空气冷却系统(Reactor Vessel Air Cooling System,RVACS)的可靠性分析。采用流体计算软件Fluent模拟RVACS系统的输入输出作为求解响应面性能函数的输入样本,利用最小二乘法和bootstrap方法估计响应面性能函数的系数,以响应面模型代替Fluent模型分析RVACS系统的非能动失效概率。分析表明,在所有能动余热排除系统不可用的情况下,RVACS四组并联排热管中的两组也能够可靠地导出反应堆余热。RVACS系统可靠性高。     

中国铅基研究堆非能动余热排出系统可靠性分析

铅冷快堆是第四代核能系统推荐堆型之一,世界上多个铅冷快堆采用非能动余热排出系统。非能动系统中作为驱动的自然力与阻力在数量级上接近,由周边环境、材料参数的变化引起的波动不可忽略,因此需要研究非能动系统可靠性。改进了常用的响应面分析法,并应用于中国铅基研究堆反应堆容器空气冷却系统(Reactor Vessel Air Cooling System,RVACS)中。分析中使用流体计算软件Fluent模拟中国铅基研究堆RVACS系统的余热排出过程,研究了输入参数的不确定性对系统可靠性及反应堆安全产生的影响。在大量模拟数据的基础上结合神经网络法建立了输入参数不确定性和结果不确定性之间的映射关系,并以此分析RVACS非能动失效概率。分析结果表明在全厂断电的情况下,RVACS四组并联排热管中的两组也能够可靠地导出反应堆余热。     

基于响应曲面和重要性抽样方法的热力系统参数失效概率计算

本文研究了将响应曲面与重要性抽样相结合的方法用于复杂热力系统参数失效概率的计算。建立了热力系统物理过程参数失效的数学模型,在此基础上研究了将响应曲面与重要性抽样相结合的算法模型,并给出了热力系统组成设备的性能退化模型和基于重要性抽样的仿真流程,进而对反应堆净化系统工作过程中参数失效问题进行了分析计算。研究表明,对于高维、非线性特性明显并考虑性能退化的复杂热力系统参数失效概率的计算,重要性抽样法较直接抽样能以较高效率获得满意精度的计算结果,而响应曲面法存在局限;响应曲面和重要性抽样相结合的方法是分析热力系统物理过程参数失效的有效方法。     

Passive system reliability analysis using Response Conditioning Method with an application to failure frequency estimation of Decay Heat Removal of PFBR

Innovative nuclear reactor designs include passive means to achieve high reliability in accomplishing safety functions. Functional reliability analyses of passive systems include Monte Carlo sampling of system uncertainties, followed by propagation through mechanistic system models. For complex passive safety systems of high reliability, Monte Carlo simulations using mechanistic codes are computationally expensive and often become prohibitive. Passive system reliability analysis using recently proposed Response Conditioning Method, which incorporates the insights obtained from approximate solutions like response surfaces in simulations to obtain computationally efficient and consistent probability estimates, is presented in this paper. The method is applied to evaluate the reliability of passive Decay Heat Removal (DHR) system of Indian Prototype Fast Breeder Reactor (PFBR). The accuracy as well as efficiency of the method is compared with direct Monte Carlo simulation. The variability of the reliability values is estimated using bootstrap technique. The system abilities, to prevent critical structural damage as well as to ensure operational safety, are quantitatively ascertained. The system functional failure probabilities are integrated with hardware failure probabilities and the inclusion of passive system unreliability in Probabilistic Safety Assessment is demonstrated.     

Estimation of the functional failure probability of a thermal-hydraulic passive system by Subset Simulation

In the light of epistemic uncertainties affecting the model of a thermal-hydraulic (T-H) passive system and the numerical values of its parameters, the system may find itself in working conditions which do not allow it to accomplish its function as required. The estimation of the probability of these functional failures can be done by Monte Carlo (MC) sampling of the uncertainties in the model followed by the computation of the system response by a mechanistic T-H code. The procedure requires considerable computational efforts for achieving accurate estimates. Efficient methods for sampling the uncertainties in the model are thus in order.In this paper, the recently developed Subset Simulation (SS) method is considered for improving the efficiency of the random sampling. The method, originally developed to solve structural reliability problems, is founded on the idea that a small failure probability can be expressed as a product of larger conditional probabilities of some intermediate events: with a proper choice of the conditional events, the conditional probabilities can be made sufficiently large to allow accurate estimation with a small number of samples. Markov Chain Monte Carlo (MCMC) simulation, based on the Metropolis algorithm, is used to efficiently generate the conditional samples, which is otherwise a non-trivial task.The method is here developed for efficiently estimating the probability of functional failure of an emergency passive decay heat removal system in a simple steady-state model of a Gas-cooled Fast Reactor (GFR). The efficiency of the method is demonstrated by comparison to the commonly adopted standard Monte Carlo Simulation (MCS).     

多种影响因素下的设备易损性计算

抗震裕度评估是核电厂地震安全评估的方法之一，通过地震易损性分析计算高置信度低失效概率的抗震能力值是抗震裕度评估中很重要的一步。本文对于同时受到多种失效模式影响的设备易损性计算进行了研究，讨论了蒙特卡罗抽样方法和拉丁超立方分布抽样方法在设备易损性计算中的应用，对两种抽样方法的计算效率和准确度进行了评价。结果表明，在小样本抽样计算时拉丁超立方抽样方法有更好的计算效率和收敛速度，在1 000次样本数量时，两种抽样方法计算结果均可达到收敛。     

Quantitative functional failure analysis of a thermal–hydraulic passive system by means of bootstrapped Artificial Neural Networks

The estimation of the functional failure probability of a thermal–hydraulic (T–H) passive system can be done by Monte Carlo (MC) sampling of the epistemic uncertainties affecting the system model and the numerical values of its parameters, followed by the computation of the system response by a mechanistic T–H code, for each sample. The computational effort associated to this approach can be prohibitive because a large number of lengthy T–H code simulations must be performed (one for each sample) for accurate quantification of the functional failure probability and the related statistics.     

Efficient reliability estimate of passive thermal hydraulic safety system with automatic differentiation

An approach for efficient estimation of passive safety system functional reliability has been developed and applied to a simplified model of the passive residual heat transport system typical of sodium cooled fast reactors to demonstrate the reduction in computational time. The method is based on generating linear approximations to the best estimate computer code, using the technique of automatic reverse differentiation. This technique enables determination of linear approximation to the code in a few runs independent of the number of input variables for each response variable. The likely error due to linear approximation is reduced by augmented sampling through best estimate code in the neighborhood of the linear failure surface but in a sub domain where linear approximation error is relatively more. The efficiency of this new approach is compared with importance sampling MCS which uses the linear approximation near the failure region and with Direct Monte-Carlo Simulation. In the importance sampling MCS, variants employing random sampling with Box-Muller algorithm and Markov Chain algorithm are inter-compared. The significance of the results with respect to system reliability is also discussed.