Surrogate-based modeling and dimension reduction techniques for multi-scale mechanics problems

Successful modeling and/or design of engineering systems often requires one to address the impact of multiple ＂design variables＂ on the prescribed outcome.There are often multiple,competing objectives based on which we assess the outcome of optimization.Since accurate,high fidelity models are typically time consuming and computationally expensive,comprehensive evaluations can be conducted only if an efficient framework is available.Furthermore,informed decisions of the model/hardware＇s overall performance rely on an adequate understanding of the global,not local,sensitivity of the individual design variables on the objectives.The surrogate-based approach,which involves approximating the objectives as continuous functions of design variables from limited data,offers a rational framework to reduce the number of important input variables,i.e.,the dimension of a design or modeling space.In this paper,we review the fundamental issues that arise in surrogate-based analysis and optimization,highlighting concepts,methods,techniques,as well as modeling implications for mechanics problems.To aid the discussions of the issues involved,we summarize recent efforts in investigating cryogenic cavitating flows,active flow control based on dielectric barrier discharge concepts,and lithium（Li）-ion batteries.It is also stressed that many multi-scale mechanics problems can naturally benefit from the surrogate approach for ＂scale bridging.＂

Numerical failure analysis of a continuous reinforced concrete bridge under strong earthquakes using multi-scale models

Previous failure analyses of bridges typically focus on substructure failure or superstructure failure separately. However, in an actual bridge, the seismic induced substructure failure and superstructure failure may influence each other. Moreover, previous studies typically use simplified models to analyze the bridge failure; however, there are inherent defects in the calculation accuracy compared with using a detailed three-dimensional （3D） finite element （FE） model. Conversely, a detailed 3D FE model requires more computational costs, and a proper erosion criterion of the 3D elements is necessary. In this paper, a multi-scale FE model, including a corresponding erosion criterion, is proposed and validated that can significantly reduce computational costs with high precision by modelling a pseudo-dynamic test of an reinforced concrete （RC） pier. Numerical simulations of the seismic failures of a continuous RC bridge based on the multi-scale FE modeling method using LS-DYNA are performed. The nonlinear properties of the bridge, various connection strengths and bidirectional excitations are considered. The numerical results demonstrate that the failure of the connections will induce large pounding responses of the girders. The nonlinear deformation of the piers will aggravate the pounding damages. Furthermore, bidirectional earthquakes will induce eccentric poundingsto the girders and different failure modes to the adjacent piers.

Analysis and modeling of resistive switching mechanisms oriented to resistive random-access memory

With the progress of the semiconductor industry,the resistive random-access memory（RAM） has drawn increasing attention.The discovery of the memristor has brought much attention to this study.Research has focused on the resistive switching characteristics of different materials and the analysis of resistive switching mechanisms.We discuss the resistive switching mechanisms of different materials in this paper and analyze the differences of those mechanisms from the view point of circuitry to establish their respective circuit models.Finally,simulations are presented.We give the prospect of using different materials in resistive RAM on account of their resistive switching mechanisms,which are applied to explain their resistive switchings.

Multi-scale modeling for prediction of mechanical performance in brazed GH99 thin-walled structure

Superalloy thin-walled structures are achieved mainly by brazing,but the deformation process of brazed joints is non-uniform,making it a challenging research task.This paper records a thorough investigation of the effect of brazing parameters on the microstructure of joints and its mechanical properties,which mainly inquires into the deformation and fracture mechanisms in the shearing process of GH99/BNi-5a/GH99 joints.The macroscopic-microscopic deformation mechanism of the brazing interface during shearing was studied by Crystal Plasticity(CP)and Molecular Dynamics(MD)on the basis of the optimal brazing parameters.The experimental results show that the brazing interface is mainly formed by(Ni,Cr,Co)(s,s)and possesses a shear strength of approximately 546 MPa.The shearing fracture of the brazed joint occurs along the brazing seam,displaying the characteristics of intergranular fracture.MD simulations show that dislocations disassociate and transform into fine twinning with increased strain.CP simulated the shear deformation process of the brazed joint.The multiscale simulation results are consistent with the experimental results.The mechanical properties of thin-walled materials for brazing are predicted using MD and CP methods.

Mechanical response of a tunnel subjected to strike-slip faulting processes,based on a multi-scale modeling method

The stick-slip action of strike-slip faults poses a significant threat to the safety and stability of underground structures.In this study,the north-east area of the Longmenshan fault,Sichuan,provides the geological background;the rheological characteristics of the crustal lithosphere and the nonlinear interactions between plates are described by Burger’s viscoelastic constitutive model and the friction constitutive model,respectively.A large-scale global numerical model for plate squeezing analysis is established,and the seemingly periodic stick-slip action of faults at different crust depths is simulated.For a second model at a smaller scale,a local finite element model(sub-model),the time history of displacement at a ground level location on the Longmenshan fault plane in a stick-slip action is considered as the displacement loading.The integration of these models,creating a multi-scale modeling method,is used to evaluate the crack propagation and mechanical response of a tunnel subjected to strike-slip faulting.The determinations of the recurrence interval of stick-slip action and the cracking characteristics of the tunnel are in substantial agreement with the previous field investigation and experimental results,validating the multi-scale modeling method.It can be concluded that,regardless of stratum stiffness,initial cracks first occur at the inverted arch of the tunnel in the footwall,on the squeezed side under strike-slip faulting.The smaller the stratum stiffness is,the smaller the included angle between the crack expansion and longitudinal direction of the tunnel,and the more extensive the crack expansion range.For the tunnel in a high stiffness stratum,both shear and bending failures occur on the lining under strike-slip faulting,while for that in the low stiffness stratum,only bending failure occurs on the lining.

NANO SCIENCE AND ENGINEERING IN SOLID MECHANICS

According to National Science Foundation （NSF） Director A. Bement, ‘Transformative research is... research driven by ideas that stand a reasonable chance of radically changing our understanding of an important existing scientific concept or leading to the creation of a new paradigm or field of science is also characterized by its challenge to current understanding or its pathway to new frontiers.＇ Nanotechnology is one of such frontiers. It is the creation of new materials, devices and systems at the molecular level--phenomena associated with atomic and molecular interactions strongly influence macroscopic material properties with significantly improved mechanical, optical, chemical, electrical... properties. Former NSF Director Rita Colwell in 2002 declared, ‘nanoscale technology will have an impact equal to the Industrial Revolution＇. The transcendent technologies include nanotechnology, microelectronics, information technology and biotechnology as well as the enabling and supporting mechanical and civil infrastructure systems and materials. These technologies are the primary drivers of the twenty first century and the new economy. Mechanics is an essential eleraent in all of the transcendent technologies. Research opportunities, education and challenges in mechanics, including experimental, numerical and analytical methods in nanomechanics, carbon nano-tubes, bio-inspired materials, fuel cells, as well as improved engineering and design of materials are presented and discussed in this paper.

基于指数随机图模型的发明者合作网络形成机制研究

聚焦探讨不同因素是否以及如何影响发明者合作网络形成,基于2004—2021年华为在中国申请的专利数据,构建发明者合作网络,运用指数随机图模型探讨行动者属性和内生结构效应对发明者合作网络形成的影响机理。结果表明:发明者合作网络是稀疏网络,呈现星形结构和闭合三角形结构共存的网络结构模式;发明者协同创新能力抑制新合作关系的形成,但发明者合作伙伴多样性和合作深度均促进新合作关系的形成,且具有相同水平合作伙伴多样性或合作深度的发明者之间更容易建立合作关系;发明者倾向于与处于中心位置的发明者建立合作关系以及与具有共同合作伙伴的直接合作伙伴建立新合作关系。

冻融循环下玄武岩纤维混凝土冲击力学性能预测模型

冻融循环下纤维混凝土的劣化规律是寒区服役混凝土工程安全性和耐久性评价的重要依据,现有服役混凝土工程的安全性和耐久性评价的研究具有工作量大、成本高、周期长等特点,构建基于机器学习的高精度力学性能预测模型已成为本领域研究热点。为探究冻融循环后玄武岩纤维混凝土冲击力学性能的高精度预测模型,采用SHPB装置对冻融循环后BFRC开展动态冲击压缩力学性能试验,并构建机器学习-Optuna混合预测模型,对60组以玄武岩纤维体积掺量、冻融循环次数、动荷载冲击速度为影响因素建立的动态峰值应力样本数据集进行预测。结果表明:k近邻、Lasso、多层感知机、极度梯度提升树和随机森林5种经典机器学习模型的预测准确度均较高,说明机器学习算法对于冻融循环后BFRC动态力学性能预测具有良好的预测效果,其中随机森林算法为最优预测算法;RF-Optuna混合预测模型显示出0.9754的拟合优度,具有良好的预测精度;非数据集工况预测表明,该混合模型对于各影响因素均具有良好泛化能力。研究成果可为冻融循环条件下BFRC动态力学性能的快捷精准预测提供参考。

基于改进DeeplabV3+的水面多类型漂浮物分割方法研究

【目的】为解决传统图像处理方法鲁棒性差、常用深度学习检测方法无法准确识别大片漂浮物的边界等问题,【方法】提出一种基于改进DeeplabV3+的水面多类型漂浮物识别的语义分割方法,提高水面漂浮的识别能力。对所收集实际水面漂浮物进行分类,采用自制数据集进行对比试验。算法选择xception网络作为主干网络以获得初步漂浮物特征,在加强特征提取网络部分引入注意力机制以强调有效特征信息,在后处理阶段加入全连接条件随机场模型,将单个像素点的局部信息与全局语义信息融合。【结果】对比图像分割性能指标,改进后的算法mPA(Mean Pixel Accuracy)提升了5.73%,mIOU(Mean Intersection Over Union)提升了4.37%。【结论】相比于其他算法模型,改进后的DeeplabV3+算法对漂浮物特征的获取能力更强,同时能获得丰富的细节信息以更精准地识别多类型水面漂浮物的边界与较难分类的漂浮物,在对多个水库场景测试后满足实际水域环境中漂浮物检测的需求。

单轴压缩荷载作用下混凝土损伤破坏的细观力学分析

基于伸缩因子函数的变形方法和蒙特卡罗方法,建立了由骨料、砂浆和二者之间的界面过渡区构成的混凝土随机骨料模型。采用有限单元法对混凝土随机骨料模型的抗压强度、应力峰值、应力分布、损伤状态等力学性能进行了数值计算,从细观尺度上分析了混凝土的损伤破坏过程及裂缝形貌特征。结果表明:在单轴压缩荷载作用下,混凝土首先在薄弱区域的界面过渡区处发生拉应力和剪应力集中,产生初始损伤;然后损伤扩展至水泥砂浆,形成贯穿裂缝;最终,导致混凝土承载力失效。

基于改进CNN-LSTM和RF的铁水KR脱硫预测模型

为实现较高精度的脱硫剂加入量预测,有效提高生产效益,本文提出一种基于改进卷积神经网络(CNN)-长短期记忆(LSTM)网络和随机森林(RF)结合的铁水脱硫两步预测模型。考虑到模型输入数据的相关性,利用皮尔逊相关系数确定各输入参数的相关性并筛选特征。模型以CNN-LSTM为基础,增加卷积层和残差连接,在提高挖掘数据的高维特征信息的同时避免网络退化。为增加网络对特征的区分和关注能力,引入多头注意力机制,让网络更加关注特征中的重要信息。使用贝叶斯优化RF超参数构建误差预测模型从而实现残差推理,对改进的CNN-LSTM模型预测结果进行修正。以现场采集的数据进行实验,结果表明,与CNN-LSTM模型相比,本文模型的拟合精度R2提升了17.11%,平均绝对值误差MAE降低了24.85%,均方根误差RMSE降低了30.18%,平均绝对百分比误差MAPE降低了28.33%。

磁流变延时机构数字系孪生系统

针对当前磁流变延时机构延时性能预测存在实验相对困难、成本高和性能预测精确度不足等问题,将数字孪生技术与磁流变延时机构工作机理相结合,开发了磁流变延时机构数字孪生系统。建立磁流变延时机构数字孪生系统整体架构,规划并设计系统功能模块及运行机制,分析磁流变延时机构数字孪生系统物理空间、孪生空间、孪生数据及可视化服务的实现方法。融合机理仿真与实验数据,采用随机森林算法借助Unity3D可视化平台,实现基于数字孪生技术的磁流变延时机构延时性能高精度快速仿真预测及其虚拟可视化。设计开发磁流变延时机构数字孪生系统,并结合实验数据验证系统功能,为实现引信数字孪生全生命周期的数据交互及其可视化提供服务。

Analysis of grinding mechanics and improved grinding force model based on randomized grain geometric characteristics

Too high grinding force will lead to a large increase in specific grinding energy, resulting in high temperature in grinding zone, especially for the aerospace difficult cutting metal materials,seriously affecting the surface quality and accuracy. At present, the theoretical models of grinding force are mostly based on the assumption of uniform or simplified morphological characteristics of grains, which is inconsistent with the actual grains. Especially for non-engineering grinding wheel,most geometric characteristics of grains are ignored, resulting in the calculation accuracy that cannot guide practical production. Based on this, an improved grinding force model based on random grain geometric characteristics is proposed in this paper. Firstly, the surface topography model of CBN grinding wheel is established, and the effective grain determination mechanism in grinding zone is revealed. Based on the known grinding force model and mechanical behavior of interaction between grains and workpiece in different stages, the concept of grain effective action area is proposed. The variation mechanism of effective action area under the influence of grain geometric and spatial characteristics is deeply analyzed, and the calculation method under random combination of five influencing parameters is obtained. The numerical simulation is carried out to reveal the dynamic variation process of grinding force in grinding zone. In order to verify the theoretical model, the experiments of dry grinding Ti-6Al-4 V are designed. The experimental results show that under different machining parameters, the results of numerical calculation and experimental measurement are in good agreement, and the minimum error value is only 2.1 %, which indicates that the calculation accuracy of grinding force model meets the requirements and is feasible. This study will provide a theoretical basis for optimizing the wheel structure, effectively controlling the grinding force range, adjusting the grinding zone temperature and improving the workpiece machining quality in the industrial grinding process.

Research on failure criterion of composite based on unified macro- and micro-mechanical model

A new unified macro- and micro-mechanics failure analysis method for composite structures was developed in order to take the effects of composite micro structure into consideration. In this method, the macro stress distribution of composite structure was calculated by commercial finite element analysis software. According to the macro stress distribution, the damage point was searched and the micro-stress distribution was calculated by reformulated finite-volume direct averaging micromechanics （FVDAM）, which was a multi-scale finite element method for composite. The micro structure failure modes were estimated with the failure strength of constituents. A unidirectional composite plate with a circular hole in the center under two kinds of loads was analyzed with the traditional macro-mechanical failure analysis method and the unified macro- and micro-mechanics failure analysis method. The results obtained by the two methods are consistent, which show this new method＇s accuracy and efficiency.

Enhancing Tea Leaf Disease Identification with Lightweight MobileNetV2

Diseases in tea trees can result in significant losses in both the quality and quantity of tea production.Regular monitoring can help to prevent the occurrence of large-scale diseases in tea plantations.However,existingmethods face challenges such as a high number of parameters and low recognition accuracy,which hinders their application in tea plantation monitoring equipment.This paper presents a lightweight I-MobileNetV2 model for identifying diseases in tea leaves,to address these challenges.The proposed method first embeds a Coordinate Attention(CA)module into the originalMobileNetV2 network,enabling the model to locate disease regions accurately.Secondly,a Multi-branch Parallel Convolution(MPC)module is employed to extract disease features across multiple scales,improving themodel’s adaptability to different disease scales.Finally,the AutoML for Model Compression(AMC)is used to compress themodel and reduce computational complexity.Experimental results indicate that our proposed algorithm attains an average accuracy of 96.12%on our self-built tea leaf disease dataset,surpassing the original MobileNetV2 by 1.91%.Furthermore,the number of model parameters have been reduced by 40%,making itmore suitable for practical application in tea plantation environments.

基于随机游走的图扩散模型

图注意力网络(Graph Attention Networks,GAT)通过汇聚相邻节点计算中心节点特征时,缺少图的结构信息且没有利用高阶邻域节点。针对此问题,提出一种采用随机游走策略的图扩散模型。该模型通过随机游走访问邻域内的节点以提取结构信息,并通过设置游走时的重启参数和迭代次数,控制邻域范围以提取局部和全局节点信息,再通过图注意力的加权求和机制对特征进行转换。该模型在3个引文图数据集上进行实验比较,比传统GAT模型的准确率平均提升了1.1%,证明了随机游走策略在捕获节点结构信息方面发挥了重要作用。

随机复合材料结构非线性热-力耦合模拟的统计高阶多尺度方法

对于具有复杂随机细观构造的复合材料结构的非线性热-力耦合问题的随机多尺度建模和计算仍是一个具有挑战性的问题。本文发展了一个新的统计高阶多尺度方法,克服了随机多尺度问题直接模拟时巨大的计算量,实现了具有随机复合材料结构非线性热-力耦合问题的数值模拟。借助统计多尺度渐近分析和泰勒级数方法,本文严格推导了可以精确分析随机复合材料结构宏-细观尺度非线性热-力耦合响应的统计高阶多尺度计算模型。然后,通过局部误差分析证明了统计高阶多尺度计算模型中高阶校正项在保持计算模型局部能量和动量守恒的重要意义。进一步,建立了可以高效模拟随机复合材料结构非线性热-力耦合行为的具有离线和在线两阶段的时空多尺度算法。最后,通过数值实验验证了统计高阶多尺度方法的计算高效率和高精度。

基于注意力机制和随机像素擦除的双判别器单图生成对抗网络

针对单一自然图像生成模型的梯度消失和模式崩溃问题,提出了基于注意力机制和随机像素擦除的双判别器单图生成对抗网络模型。具体方法为:在生成器中引入CBAM模块以增强特征表示;采用多样性损失优化损失函数提高生成图像多样性;在图像上采样及过渡到下一阶段前增加随机像素擦除进一步丰富输出的多样性;在判别器中集成自注意力机制捕获更加全面的依赖关系;实施双判别器设计减轻模式崩溃问题。实验结果表明:与单一自然图像生成模型相比,本文方法在图像质量、多样性和训练稳定性方面均有显著提升。

基于深度融合模型的气膜密封端面状态识别方法

气膜密封装置是工业领域应用广泛的一种密封技术,其可靠的密封性能对于设备正常运行至关重要。气膜密封装置的动静密封环接触端面相对运动会产生摩擦,摩擦过程会产生复杂的声发射信号,这些信号往往隐含密封端面运行状况的重要信息。采用传统的方法往往难以准确识别和分类这些微弱的特征信号,因此需要开发更高精度的故障诊断方法。针对机械密封动、静环端面摩擦状态难以识别这一问题,以气膜密封装置为研究对象,提出了一种基于深度融合模型的气膜密封端面状态识别方法。首先,采用声发射传感器及采集设备,对密封端面的声发射信号进行了采集;其次,利用小波包变换方法对采集到的信号进行了滤波处理,并提取了时域和频域的微弱特征;然后,将深度随机森林(DRF)作为分类层融入卷积神经网络(CNN)形成了融合模型,对预先处理过的密封装置运行状态的特征信息进行了识别和分类;最后,根据实验的泄漏量,使用混淆矩阵和受试者工作曲线分析了两种模型的特征提取能力。研究结果表明:CNN-DRF融合模型对于密封端面声发射信号的两种特征识别精度分别为96%和98%,与传统的CNN模型相比,其可以充分提取信号特征信息,具有更出色的故障诊断能力。

景区间旅游客流网络形成的内-外生机制分析--基于加权指数随机图模型

景区间旅游客流网络的形成机制是旅游流网络研究的重要问题。已有研究主要识别了旅游客流网络形成的外生机制,而对网络内生机制缺乏研究,文章旨在更全面地揭示景区间旅游客流网络形成的内-外生机制。文章选择四川省为案例地,基于旅游社交平台的游记数据构建了景区间旅游客流的有向加权网络,通过引入加权指数随机图模型,定量模拟并识别了影响该网络形成的内-外生机制。1)四川省景区间旅游客流网络为稀疏网络、无标度网络,但不具备小世界特性;核心的景区节点和网络连边大多集中在成都市。2)该网络的形成是内-外生机制协同作用的结果,内生机制表现为网络内生结构效应,外生机制包括景区属性效应和外部情境效应。3)作为内生机制的互惠机制和择优链接机制促进了该网络中联系的形成;外生机制方面,在线评论分数、景区等级同质性、景区类型同质性、地理邻近性、组织邻近性均会显著影响该网络中联系的形成。