压水堆下封头多层熔池模型敏感性分析

下封头熔池模型是熔融物堆内滞留（IVR）有效性评价的重要模型，已在典型压水堆安全评价中得到广泛应用。传统的2层熔池模型和近年来提出的3层熔池模型，主要模拟熔池内熔融物的成分及热量的分配与传递过程，具有关系式复杂和强非线性的特点。为了为熔池分层模型以及严重事故缓解策略的优化提供帮助，采用中国核动力研究设计院自研的全局敏感性分析工具SALib和熔池分析软件CISER V2.0对4种熔池多层模型进行了敏感性分析，得到了主要输入参数对各模型关键结果参数的影响程度，敏感性分析结果反映了各熔池模型的典型特点。下封头半径对4种熔池分层模型均有显著的影响，Salay&Fichot模型与2层熔池模型中影响关键结果参数的输入参数基本相同，熔融物初始质量对Esmaili模型影响最大，熔融物密度对Seiler模型影响最大。 

Sensitivity Analysis of Multi-Layer Molten Pool Model of PWR Lower Head

The molten pool model of the lower head is an important model to evaluate the effectiveness of In Vessel Retention (IVR), which has been widely used in the safety evaluation of typical pressurized water reactor (PWR). The distribution and transfer process of the composition and heat of the melt in the pool of the traditional two-layer pool model and the three-layer pool model proposed in recent years are simulated, which is with the characteristics of complex relationship and strong nonlinearity. In order to provide a support for the optimization of the molten pool delamination model and the mitigation strategy of serious accidents, both global sensitivity analysis library (SALib) and IVR analysis code (CISER V2.0) developed by Nuclear Power Institute of China are used to analyze the sensitivity of four molten pool multilayer models and obtain the influence degree of the main input parameters on the key result parameters of each model. The sensitivity analysis results reflected a typical characteristics of the molten pool model. The radius of the lower head has a significant impact on the four molten pool multilayer models, while the input parameters that have significant influence on the key result parameters are basically the same in the Salay & Fichot model and the two-layer molten pool model, the initial mass of molten material has the greatest impact on the Esmaili model, and the density of the molten material has the greatest impact on the Seiler model.