{"title":"Sensitivity and Uncertainty Information Incorporated Loss of Flow Accident Analyses for Research Reactors","authors":"Tao Liu, Zeyun Wu","doi":"10.1115/icone2020-16242","DOIUrl":null,"url":null,"abstract":"\n This paper outlines a system level safety analysis procedure for research reactors incorporating sensitivity and uncertainty components. The protected loss of flow (LOF) accident was selected as an exemplified design basis accident to demonstrate the analysis procedure. The conceptual NIST (National Institute of Standards and Technology) horizontally split-core based research reactor was adopted as a research reactor model in the study. Two system level dynamics codes, RELAP5-3D and PARET, were employed in this work in a comparison study manner. The primary objective of the present work is to demonstrate the analysis capability of integrating sensitivity and uncertainty information in addition to traditional predictions of the system code models for the study of the thermal-hydraulics (T/H) safety characteristics of research reactors under accidental transient scenarios. The canonical transient predictions on the LOF accident yielded from the two system codes mentioned above have demonstrated some noticeable yet acceptable discrepancies. To better understand the discrepancies observed in the simulations, sensitivity and uncertainty analyses were performed by coupling the RELAP5-3D model and the data analytic engines provided by the RAVEN framework developed by INL. The sensitivity information reveals the significances of key figure of merits such as the peak cladding temperature varies with different boundary and initial parameters in both normal operation and design basis transients. The uncertainty analysis informs the deviations of the responses contributed by the errors of various input components. Both the sensitivity and uncertainty information will be incorporated into a safety analysis framework as part of the safety characteristic predictions delivered by the framework.","PeriodicalId":414088,"journal":{"name":"Volume 3: Student Paper Competition; Thermal-Hydraulics; Verification and Validation","volume":"86 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2020-08-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Volume 3: Student Paper Competition; Thermal-Hydraulics; Verification and Validation","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1115/icone2020-16242","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 1
Abstract
This paper outlines a system level safety analysis procedure for research reactors incorporating sensitivity and uncertainty components. The protected loss of flow (LOF) accident was selected as an exemplified design basis accident to demonstrate the analysis procedure. The conceptual NIST (National Institute of Standards and Technology) horizontally split-core based research reactor was adopted as a research reactor model in the study. Two system level dynamics codes, RELAP5-3D and PARET, were employed in this work in a comparison study manner. The primary objective of the present work is to demonstrate the analysis capability of integrating sensitivity and uncertainty information in addition to traditional predictions of the system code models for the study of the thermal-hydraulics (T/H) safety characteristics of research reactors under accidental transient scenarios. The canonical transient predictions on the LOF accident yielded from the two system codes mentioned above have demonstrated some noticeable yet acceptable discrepancies. To better understand the discrepancies observed in the simulations, sensitivity and uncertainty analyses were performed by coupling the RELAP5-3D model and the data analytic engines provided by the RAVEN framework developed by INL. The sensitivity information reveals the significances of key figure of merits such as the peak cladding temperature varies with different boundary and initial parameters in both normal operation and design basis transients. The uncertainty analysis informs the deviations of the responses contributed by the errors of various input components. Both the sensitivity and uncertainty information will be incorporated into a safety analysis framework as part of the safety characteristic predictions delivered by the framework.
本文概述了一个包含灵敏度和不确定度成分的研究堆系统级安全分析程序。并以保护失流(LOF)事故为例,对分析过程进行了说明。本研究采用概念型NIST (National Institute of Standards and Technology)水平裂芯研究堆作为研究堆模型。本文采用RELAP5-3D和PARET两种系统级动力学代码进行对比研究。本工作的主要目的是证明除了传统的系统代码模型预测之外,集成灵敏度和不确定性信息的分析能力,用于研究反应堆在事故瞬态情景下的热工水力(T/H)安全特性。从上述两种系统代码得到的LOF事故的典型瞬态预测显示出一些明显但可接受的差异。为了更好地理解模拟中观察到的差异,通过将RELAP5-3D模型与INL开发的RAVEN框架提供的数据分析引擎耦合进行敏感性和不确定性分析。灵敏度信息揭示了在正常运行和设计基础瞬态中,包层峰值温度随不同边界和初始参数变化等关键参数的意义。不确定性分析告知了由各种输入分量的误差引起的响应偏差。敏感性和不确定性信息都将被纳入安全分析框架,作为框架提供的安全特性预测的一部分。