arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2606.04033cs.LG
收藏

通过梯度优化与多群注意力神经网络架构的逆临界实验设计

Gradient-based optimization of nuclear criticality experiments using neural surrogate eigenvalue sensitivities

Massachusetts Institute of Technology(麻省理工学院)

Will Savage, Logan Burnett, Dean Price

AI总结:

提出一种结合深度神经网络代理模型和非参数梯度优化的方法,用于逆设计临界实验几何结构以最大化中子相似性相关系数$c_k$,并在TN-LC运输容器验证中取得高$c_k$值。

摘要:

先进核反应堆设计和燃料概念的验证需要与目标技术具有高中子相似性的临界实验。中子相似性由相关系数$c_k$量化,该系数捕捉了核数据不确定性引起的$k_\text{eff}$共享偏差。通常,实验需要$c_k\geq0.9$才能与目标技术充分相似。本文提出了一种临界实验逆设计方法。使用深度神经网络代理模型和非参数梯度优化来生成最大化$c_k$的实验几何结构。 深度神经网络基于OpenMC计算的网格临界实验几何结构的灵敏度向量进行训练。模型架构结合了U-Net卷积编码器-解码器与新颖的多群注意力池化层,引入该层以捕捉灵敏度的不同空间依赖性。多群注意力池化在性能上优于传统池化,并具有可解释的内部行为。代理模型的可微性使得能够对全组合设计空间进行基于梯度的优化,通过直接改变几何网格中每个位置的材料分配来最大化$c_k$。 该方法应用于TN-Americas TN-LC运输容器(使用HALEU燃料)的验证,该容器现有的临界实验覆盖有限。优化过程为三个感兴趣配置生成了实验几何结构,分别达到0.97757、0.81324和0.93276的$c_k$分数。该方法展示了深度学习和梯度优化在加速先进核技术开发方面的潜力。

英文摘要:

The validation of advanced nuclear reactor designs and fuel concepts will require the design of new critical experiments with high neutronic similarity to the target technology. Neutronic similarity can be quantified by the correlation coefficient $c_k$, which captures the shared bias in $k_\text{eff}$ induced by uncertainties in nuclear data. Generally, a $c_k\geq0.9$ is needed for an experiment to be sufficiently similar to a target technology. In this work, a physics-informed deep neural network is trained to predict the neutronic sensitivity of grid-based critical experiment geometries. The differentiability of the neural network is used to enable gradient-based design optimization of new experiment geometries to maximize $c_k$ with the sensitivity profile of a target technology. This approach allows for optimization over the combinatorial design space of potential material combinations within the grid, moving beyond traditional parametric optimization approaches. The method is applied to the validation of the TN-Americas TN-LC transportation cask with HALEU fuel, for which existing critical experiment coverage is limited. This application is shown to produce experiment geometries achieving $c_k$ scores of 0.97757, 0.81324, and 0.93276 for three configurations of interest.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑