基于深度学习的超高速碰撞碎片云生成模型

数据驱动的模型已经被广泛研究,并成功应用到了计算力学。基于深度学习技术,提出一种新的采用数据驱动的碎片云生成模型。此模型可以学习SPH数值模拟结果,然后在多种控制条件下快速生成碎片云。在模型训练前的数据预处理阶段,对SPH模拟结果进行空间网格划分和质量聚合,实现了改善数据分布规律、加速模型训练和提升模型泛化性的目的。以高速靶球撞击薄壁圆筒后的碎片云质量分布为例,模拟并测试了多种控制条件下深度学习模型计算结果的正确性和稳定性,以及计算速度的高效性。实验证明,深度学习模型可以从训练集学习碎片云的物理规律,然后在训练集控制参数范围内进行良好的推理及插值;并且可以在训练数据集控制参数范围外,进行小范围推理预测;同时深度学习模型的计算速度远快于SPH方法。通过深度学习方法建立碎片云模型,可能是一种在空间飞行器防护结构原型设计阶段,实现碎片云实时生成的潜在方案。     

因子模糊化BP神经网络在磨粒识别中的应用

在引入磨粒形态学描述了提取磨损颗粒显微形态特征的基础上,用人工神经网络技术,编制了用于磨损颗粒自动识别的ＢＰ网络计算机模拟程序,应用所引入的因子模糊化训练法可使训练速度加快,以异或问题为例,速度可提高５～１０倍。用此网络对磨粒测试库进行识别实验发现,识别速度快且正常率在９０％以上,优于传统的磨粒识别方法。     

电感式磨粒传感器中铁磁质磨粒的磁特性研究

在电感式磨粒传感器中,铁磁质磨粒主要通过磁化作用改变传感器线圈的磁场分布,进而改变线圈的等效电感.建立了线圈中含有铁磁质磨粒时的磁场模型,得出了磨粒磁化场关于退磁因子的磁感应强度表达式.以球磨粒为例,通过计算球磨粒磁化场,得到球磨粒引起线圈电感变化率的解析式,并用有限元法计算了线圈的磁场,分析后发现:理论计算解与数值分析解相符,所建模型可以反映客观实际;磨粒的磁化强度由其磁导率和退磁因子共同决定,球磨粒的磁化强度可近似认为与磁导率无关;球磨粒引起的传感器线圈电感变化率随线圈单位长度上匝数的增加而减小,并趋向于一极限值;传感器线圈的电感变化率与球磨粒半径的三次方成正比,球磨粒半径在100μm以内,电感变化率在10-7数量级.本研究结论可为电感式磨粒传感器的设计提供理论指导.     

金属板料拉延二次成形的有限元法模拟

建立了二次拉延成形加工的有限元分析计算模型;采用一次成形-回弹计算-二次成形的连续计算过程模拟了实际加工过程;有限元计算采用动力显式计算程序MSC/DYTRAN;用主从面（master surface-slave surface)模型定义板料和模具的接触,摩擦力用库仑定律计算;利用动力松弛法对成形过程中的回弹进行了计算。模拟结果和实际零件比较,证明模型合理,自满稳定,结果可靠,具有良好的应用价值。     

深部软岩新型相似模拟材料的研制

在总结以往相似模拟材料特点的基础上,结合深部软岩强度、变形特征,本着相似材料组分简单、模型制作便捷等原则,研制了仅以河砂为骨料,松香酒精溶液为粘结剂的深部软岩新型相似模拟材料。通过对不同配比试样进行单、三轴压缩、巴西劈裂试验,获得了不同配比条件下该相似模拟材料的强度、变形、破坏特征。研究表明,通过简单的配比调配,该相似模拟材料可以较好地模拟深部大多类型的软岩,且具有组分简单、模型制作便捷、力学性能稳定、无毒无污染、价格低廉、一定程度上可重复利用等特点。     

Linking Morphology of Porous Media to Their Macroscopic Permeability by Deep Learning

Flow, transport, mechanical, and fracture properties of porous media depend on their morphology and are usually estimated by experimental and/or computational methods. The precision of the computational approaches depends on the accuracy of the model that represents the morphology. If high accuracy is required, the computations and even experiments can be quite time-consuming. At the same time, linking the morphology directly to the permeability, as well as other important flow and transport properties, has been a long-standing problem. In this paper, we develop a new network that utilizes a deep learning (DL) algorithm to link the morphology of porous media to their permeability. The network is neither a purely traditional artificial neural network (ANN), nor is it a purely DL algorithm, but, rather, it is a hybrid of both. The input data include three-dimensional images of sandstones, hundreds of their stochastic realizations generated by a reconstruction method, and synthetic unconsolidated porous media produced by a Boolean method. To develop the network, we first extract important features of the images using a DL algorithm and then feed them to an ANN to estimate the permeabilities. We demonstrate that the network is successfully trained, such that it can develop accurate correlations between the morphology of porous media and their effective permeability. The high accuracy of the network is demonstrated by its predictions for the permeability of a variety of porous media.

     

浅议水泥搅拌桩在高速公路软基处理中的应用

实例表明,水泥搅拌桩应用于高速公路软基处理时,须重视几个关键性参数的择取及直接影响软基处理效果的几点重要关系。     

Paired and Unpaired Deep Learning Methods for Physically Accurate Super-Resolution Carbonate Rock Images

Transport in Porous Media - X-ray micro-computed tomography (micro-CT) has been widely leveraged to characterise the pore-scale geometry of subsurface porous rocks. Recent developments in...     

磨粒的三维表面特征描述

介绍了三维表面粗糙度参数和表面纹理指数,采用三维表面高度偏差Sa、均方根Sq、表面斜度Ssk、表面峭度Sku等粗糙度参数和表面纹理指数Stdi描述磨粒的三维表面特征,选用油液分析获得磨粒,对其表面特征进行实例分析.结果表明,运用上述参数能够很好区分不同类型的磨粒表面特征.     

Detailed Cloud Microphysics Simulation for Investigation into the Impact of Sea Spray on Air-Sea Heat Flux

This paper reports on the development of a numerical weather simulation model combined with a detailed spectral-bin cloud microphysics model that can explicitly consider the droplet motion and droplet-atmosphere interactions of sea spray. Sea spray is composed of liquid droplets ejected from the sea surface into the evaporation layer, where it enhances heat as well as momentum exchanges between the atmosphere and the sea. In our study, we analyzed the results of idealized 3D simulations to investigate the impact of sea spray on latent heat exchanges and their consequent impact on boundary layer cloud development. The results show that sea spray enhances the latent heat flux by up to 62 % for the surveyed 10m-height velocities, which ranged from 12 to 42 m/s. They also show that sea spray moistening significantly enhances boundary layer cloud development.