Multi-objective Optimization Capability for Heterogeneous LWR Fuel Assemblies Supercells and Cores

异构轻水堆燃料组件超级电池和堆芯的多目标优化能力

• 批准号: 1950497
• 金额: --
• 依托单位: University of Cambridge
• 依托单位国家: 英国
• 项目类别: Studentship
• 财政年份: 2017
• 资助国家: 英国
• 起止时间: 2017 至 无数据
• 项目状态: 已结题
• 来源: https://gtr.ukri.org/projects?ref=studentship-1950497
• 关键词: Multi; objective; Optimization; Capability; Heterogeneous

Due to boundary assumptions in the mathematical modelling of fuel assemblies it is suspected that gains could be achieved with larger unit models of the physical systems. However, due to the large computational load of modelling these systems it has not typically been efficient to model so-called supercells, or indeed entire cores, because 'black box' optimisation (where the optimiser repeatedly tests candidate solutions) take an exceptionally long period of time.The PhD will look at approaches to modelling larger areas than have been typically considered, and will cover the novel methods required to do this.

由于燃料组件数学建模中的边界假设，人们怀疑，物理系统的更大单元模型可能会带来收益。然而，由于对这些系统进行建模的计算量很大，因此对所谓的超级细胞或整个核心进行建模通常并不有效，因为“黑匣子”优化（优化器反复测试候选解决方案）需要非常长的时间。博士将研究对比通常考虑的更大区域进行建模的方法，并将涵盖实现这一目标所需的新方法。