Mesoscale simulations of fission gas release (FGR) in nuclear fuel provide a powerful tool for understanding how microstructure evolution impacts FGR, but they are computationally intensive. In this study, we present an alternate, data-driven approach, using deep learning to predict instantaneous FGR flux from 2D nuclear fuel microstructure images. Four convolutional neural network (CNN) architectures with multiscale regression are trained and evaluated on simulated FGR data generated using a hybrid phase field/cluster dynamics model. All four networks show high predictive power, with $R^{2}$ values above 98%. The best performing network combine a Convolutional Block Attention Module (CBAM) and InceptionNet mechanisms to provide superior accuracy (mean absolute percentage error of 4.4%), training stability, and robustness on very low instantaneous FGR flux values.

翻译：核燃料中裂变气体释放（FGR）的介观尺度模拟为理解微观结构演化如何影响FGR提供了有力工具，但此类模拟计算成本高昂。本研究提出一种基于数据驱动的替代方法，利用深度学习从二维核燃料微观结构图像预测瞬时FGR通量。我们训练了四种具有多尺度回归能力的卷积神经网络（CNN）架构，并在基于混合相场/团簇动力学模型生成的模拟FGR数据上进行评估。所有四种网络均表现出高预测能力，$R^{2}$值超过98%。性能最优的网络结合了卷积块注意力模块（CBAM）与InceptionNet机制，在极低瞬时FGR通量值上展现出卓越精度（平均绝对百分比误差为4.4%）、训练稳定性及鲁棒性。