A variant of the U-Net convolutional neural network architecture is proposed to estimate linear elastic compatibility stresses in -Zr (hcp) polycrystalline grain structures. Training data was generated using VGrain software with a regularity of 0.73 and uniform random orientation for the grain structures and ABAQUS to evaluate the stress fields using the finite element method. The initial dataset contains 200 samples with 20 held from training for validation. The network gives speedups of around 200x to 6000x using a CPU or GPU, with significant memory savings, compared to finite element analysis with a modest reduction in accuracy of up to 10%. Network performance is not correlated with grain structure regularity or texture, showing generalisation of the network beyond the training set to arbitrary Zr crystal structures. Performance when trained with 200 and 400 samples was measured, finding an improvement in accuracy of approximately 10% when the size of the dataset was doubled.

中文翻译：

---

采用 U-Net 估计多晶 Zr 微结构中的线弹性应力

提出了 U-Net 卷积神经网络架构的变体来估计 -Zr (hcp) 多晶晶粒结构中的线弹性相容应力。使用VGrain软件生成训练数据，其规则性为0.73，晶粒结构均匀随机取向，并使用ABAQUS使用有限元方法评估应力场。初始数据集包含 200 个样本，其中 20 个是通过训练进行验证的。与有限元分析相比，该网络使用 CPU 或 GPU 可将速度提高约 200 倍至 6000 倍，并显着节省内存，而精度仅适度降低 10%。网络性能与晶粒结构规律性或纹理不相关，表明网络对训练集之外的任意 Zr 晶体结构的泛化。测量了使用 200 和 400 个样本进行训练时的性能，发现当数据集大小加倍时，准确度提高了约 10%。