In recent years, the applications of graph convolutional networks (GCNs) in hyperspectral image (HSI) classification have attracted much attention. However, hyperspectral classification faces problems such as complex noise effects, spectral variability, labelled training sample deficiency, and high spectral mixing between materials. Furthermore, the available GCN-based methods are computationally complex and cannot automatically adjust aggregate paths. To mitigate these issues, we propose a novel multiadaptive receptive field-based graph neural framework (MARP) for HSI classification. In our method, an adaptive receptive path aggregation (ARP) mechanism is proposed to suppress the impact of noise nodes on classification and automatically explore an adaptive receptive field, where a graph attention (GAT) neural network is introduced to learn the importance of different-sized neighbourhoods and a long short-term memory (LSTM) method is adopted to update the nodes and preserve the local convolutional features of the nodes. To address the problem that ARP may fall into a local optimum, we design a multiscale receptive mechanism. Extensive experimental results obtained on four public HSI datasets demonstrate that the proposed MARP method can mitigate oversmoothing and reduce computational complexity while achieving competitive performance when compared to several state-of-the-art methods.

As the world moves into the 21st century, the complex electromagnetic wave environment is receiving widespread attention due to its impact on human health, suggesting the critical importance of wearable absorbing materials. In this paper, graphene nonwoven (RGO/NW) composites were prepared by diffusely distributing graphene sheets in a polypropylene three-dimensional framework through Hummers’ method. Moreover, based on the Jaumann structural material design concept, the RGO/NW composite was designed as a multilayer microwave absorber, with self-recovery capability. It achieves effective absorption (reflection loss of −10 dB) in the 2~18 GHz electromagnetic wave frequency domain, exhibiting a larger bandwidth than that reported in the literature for absorbers of equivalent thickness. In addition, the rationally designed three-layer sample has an electromagnetic wave absorption of over 97% (reflection loss of −15 dB) of the bandwidth over 14 GHz. In addition, due to the physical and chemical stability of graphene and the deformation recovery ability of nonwoven fabric, the absorber also shows good deformation recovery ability and stable absorption performance. This broadband absorption and extreme environmental adaptability make this flexible absorber promising for various applications, especially for personnel wearable devices.

This study presents the efficient rigid-body inverse dynamics of a spatial parallel mechanism (PM) constrained directly by the base at two point-contact higher kinematic pairs (HKPs). The mechanism is designed to mimic the human masticatory system and is characterised by these constraints, from which parasitic motions and actuation redundancies are derived simultaneously. At the beginning, its constrained motions are analysed comprehensively. Then to seek efficient approaches to inverse dynamics and the influence of constraints at HKPs, three models are built using the Udwadia–Kalaba analytical mechanics method. In the first model, a dynamic model free of constraints from both the base and the chains to the end-effector is built first, and then these constraints are formulated to reach the complete dynamic model of the PM. In the second model, a dynamic model free of constraints only from the chains to the end-effector is built first, and then these constraints are imposed, achieving the complete dynamic model of the PM. Whilst in the third one, a dynamic model free of only direct constraints from the base to the end-effector is built first, and then these constraint forces are developed to arrive at the model of the PM. The results show that the second model is the most time-consuming. However, the first and third models can significantly reduce the computational complexity without any accuracy loss, which is even comparable to that of the PM’s counterpart free of direct constraints from the base to the end-effector.

针对工程机械结构原理与维修教学需要,分析虚拟仿真实验的基本原理,设计涵盖 7 种常用工程机械的虚拟维修训练系统.基于虚拟仿真技术的交互控制原理,完成了自主学习、虚拟拆装训练及考核评价3个主要功能模块的设计与开发.教学实践表明,基于虚拟仿真技术的工程机械虚拟维修训练系统互动性好、沉浸感强,可以有效改善工程机械结构原理与维修教学效果.

Hyperspectral image (HSI) classification has attracted wide attention in many fields. Applying Graph Neural Network (GNN) to HSI classification is one of the research frontiers, which has improved the HSI classification accuracy greatly. However, GNN-based methods have not been widely applied due to their time-consuming, inefficient information description as well as poor anti-noise robustness. To overcome the deficiencies, a novel multi-scale receptive fields graph attention neural network (MRGAT) is proposed for HSI classification in this paper. In this network, a superpixel segment method is adopted to abstract the original HSI local spatial features. A two-layer one-dimensional convolution neural network (1D CNN) spectral transformer mechanism, is designed to extract the spectral features of superpixels, with which the spectral features can be acquired automatically. Furthermore, graph edges are introduced into Graph Attention Network (GAT) to acquire the local semantic feature of the graph. Moreover, inspired by the transformer network, we design a novel multi-scale receptive field GAT to extract the local-global adjacent node-features and edges-features. Finally, a graph attention network and a softMax function are utilized for multi-receptive feature fusion and pixel-label predicting. On Pavia University, Salinas, and Houston 2013 datasets, the overall accuracies (OAs) of our MRGAT are 71.76%, 82.61%, and 63.82%, respectively. Moreover, the performances with limited labeled samples indicates that the MRGAT contains superior adaptability. Compared with the competitive classifiers, MRGAT achieves high classification efficiency verified by training time comparison experiment.

A spatial parallel mechanism (PM) that mimics the human masticatory system and then reproduces chewing behaviours has been proposed in a bio-inspired manner. Its end effector is directly constrained by the base at two point-contact higher kinematic pairs (HKPs), which imitate the temporomandibular joints (TMJs) in the masticatory system. The six chains play the role of the primary chewing muscles. The two direct constraints from the base simultaneously produce parasitic motions and actuation redundancy, rendering the inverse dynamics considerably complicated and time-consuming, which does not facilitate the model-based real-time motion and/or force control. As such, finding an efficient computational strategy is the aim of this study. Inspired by the procedure wherein the target PM is formed, the dynamic model of its counterpart free of these constraints can be utilised to build an efficient model. To validate the effectiveness of the proposed strategy, three dynamic methods, namely, Khalil-Ibrahim method, natural orthogonal complement, and Kane’s equations, are employed. Under each of them two models are built for comparison. In the first model, the method is applied directly to the target PM as a regular strategy. In the second model, a highly efficient strategy is proposed. The dynamic model of its counterpart is built first, and then the direct constraints from the base at HKPs are formulated, reaching the inverse dynamic model of the target PM. The results show that the second model is far superior to the first in alleviating the computational complexity without any accuracy loss. It is concluded that the target PM is designed based on its counterpart free of HKPs; then its dynamic model can also be built based on that of its counterpart as a highly efficient strategy.

Due to its impressive representation power, the graph convolutional network (GCN) has attracted increasing attention in the hyperspectral image (HSI) classification. However, the most of available GCN-based methods for HSI classification utilize superpixels as graph nodes, which ignore the pixel-wise spectral-spatial features. To overcome the issues, we propose a novel multi-feature fusion network (MFGCN), where two different convolutional networks, i.e., multi-scale GCN and multi-scale convolutional neural network (CNN), are utilized in two branches, separately. The multi-scale superpixel-based GCN can reduce the computing power requirements, deal with the problem of labeled deficiency, and refine the multi-scale spatial features from HSI. The multi-scale CNN can extract the multi-scale pixel-wise local features for HSI classification. Furthermore, we introduced a 1D CNN to extract the spectral features for superpixels (nodes), which is different from most existing methods. Finally, a concatenate operation is employed to fuse the complementary multi-scale features. In comparison with the state-of-the-art models on three datasets, the proposed method achieves superior experimental results and outperforms competitive methods.

Graph convolutional network (GCN) has shown potential in hyperspectral image (HSI) classification. However, GCN is a transductive learning method, which is difficult to aggregate the new node. The available GCN-based methods fail to understand the global and contextual information of the graph. To address this deficiency, a novel semisupervised network based on graph sample and aggregate-attention (SAGE-A) for HSIs’ classification is proposed. Different from the GCN-based method, SAGE-A adopts a multilevel graph sample and aggregate (graphSAGE) network, as it can flexibly aggregate the new neighbor node among arbitrarily structured non-Euclidean data and capture long-range contextual relations. Inspired by the convolution neural network (CNN) self-attention mechanism, the proposed network uses the graph attention mechanism to characterize the importance among spatially neighboring regions, so the deep contextual and global information of the graph can be learned automatically by focusing on important spatial targets. Extensive experimental results on different real hyperspectral data sets demonstrate the performances of our proposed method compared with the state-of-the-art methods.

Hyperspectral image (HSI) clustering is an extremely fundamental but challenging task with no labeled samples. Deep clustering methods have attracted increasing attention and have achieved remarkable success in HSI classification. However, most existing clustering methods are ineffective for large-scale HSI, due to their poor robustness, adaptability, and feature presentation. In this article, to address these issues, we introduce unsupervised self-correlated learning smoothy enhanced locality preserving graph convolution embedding clustering ( $\text{S}^{2}$ LGCC) for large-scale HSI. Specifically, the spectral-spatial transformation is introduced to transform the original HSI into a graph while preserving the local spectral features and spatial structures. After that, a locality preserving graph convolutional embedding encoder is designed to learn the hidden representation from the graph, in which the deep layer-wise graph convolutional network (LGAT) is proposed to preserve the adaptive layerwise locality features. In addition, the self-correlated learning smoothy module is developed to learn the smoothy information and the nonlocal relationship in the hidden representation space for clustering. Finally, a self-training strategy is proposed to cluster the graph node, in which a self-training clustering objective employs soft labels to supervise the clustering process. The proposed $\text{S}^{2}$ LGCC is jointly optimized by the fusion graph reconstruction loss and self-training clustering loss, and the two benefit each other. On Indian Pines (IP), Salinas, and UH2013 datasets, the overall accuracies (OAs) of our $\text{S}^{2}$ LGCC are 71.76%, 82.61%, and 63.82%, respectively.

为了探究成形法磨削齿轮的磨削温度生成机制与变化规律,构建了单位接触宽度齿面磨削力与成形法磨齿温度的理论模型,揭示了它们与砂轮特性、齿轮规格、磨削用量与材料特性的内在关系.成形磨齿温度随着砂轮磨刃密度、磨削速度、径向进给量的增大而增大,却随着砂轮直径、齿轮压力角增大而减小.齿轮的模数对齿廓磨削温度的影响较弱,而齿数对齿顶与齿根两处磨削温度的影响态势却截然相反.成形法磨削齿轮的砂轮特性系数τ越大,则磨削性能越差,并在一定程度上决定了轴向进给速度对磨削温度的影响规律.此外,通过正交试验回归模型分析法证实了成形法磨齿温度理论模型具有较好的实际适用性,且微晶刚玉砂轮的磨削性能优于白刚玉砂轮.

Honeycomb (HC) composites were fabricated by impregnating an aramid paper HC core with carbon nanotubes/carbon black/reduced graphene oxide (CNTs/CB/RGO) and polyurethane resin (PU). The sandwich HC (SHC) absorber containing HC composites with superior microwave-absorption properties were fabricated using the vacuum bagging method. Through the absorption performance of the SHC absorber, it can be concluded that the triple-layer SHC absorber has the best absorbing performance. The effective bandwidth (reflection loss < 10 dB) can be achieved in the entire frequency range of 2.2–18 GHz, and the minimum RL value is −35 dB. Furthermore, the compressive stress of the triple-layer SHC absorber reached 3.71 MPa, which is similar to the compressive stress of aluminum HC panels for aviation. Benefiting from the excellent integration of absorption and mechanical performance, the SHC has significant potential in the stealth-technology field.

The honeycomb structure absorbing material, which has both load-bearing and absorbing functions, has an irreplaceable position in the field of modern absorbing materials. In order to realize the effective permittivity calculation of honeycomb absorbers for further multi-structure design. Here, firstly, the Maxwell-Garnett (MG) formula is derived to summarize the conditions of use and the scope of application. Secondly, a new effective medium model is constructed based on the MG formula and characteristics of honeycomb structured wave absorbing composites. Finally, Based on the MG formula, the equivalent calculation of an example using the model shows that the obtained permittivity is basically consistent with the measured permittivity, which proves that the model can predict the effective permittivity of the honeycomb absorbing composite precisely. This is of great guidance and use for the future design of multilayer honeycomb structures.

推导了电阻点焊中接触电阻的计算模型,结合COMSOL仿真软件模拟了电容脉冲焊补设备对厚度为0.2 mm1Cr18Ni9Ti不锈钢的点焊.分析了不同电极压力下对应的熔核及设备参数,得到在该厚度下不锈钢最佳的焊补电极压力为50N,分析了不同电压下对应的熔极及设备参数,得到该厚度不锈钢的焊补电压为10 V,放电压差为3 V.模拟了上述参数下熔核边缘温度的冷却过程,得到的冷却时间为15 ms,对应放电控制开关的频率范围为1?65 Hz.

为了探究硅片器件精密磨削加工的切削特征与机理,运用三棱锥形状的金刚石磨粒以不同加载压力划刻单晶硅材料表面模拟磨削加工过程,分析了划痕形貌特征、切削力与切削深度的演变规律,阐释了单晶硅的微米级切削加工机理.单晶硅微破碎去除发生的临界条件为法向切削力80 mN,临界切削深度2.03μm;剥落去除发生的临界条件为法向切削力800 mN,切削深度5.65μm.切削深度、切削力比在不同切削机理条件下具备可区分的差异化特征.平均切削深度随加载压力的变化规律呈现出鲜明的自相似性特征.此外,还分别构建了塑性去除、微破碎去除、剥落去除三个阶段的切削力方程,更准确地描述了切削力与切削深度的密切关系.

针对离散制造车间实时监控困难、调控能力差、管理不透明等问题,采用面向对象的方法,提出一种基于数字孪生的离散制造车间可视化实时监控方法.首先,搭建了基于数字孪生的离散制造车间可视化实时监控方法体系架构,明确了其关键实现流程;然后,分别围绕4个关键技术:基于AutomationML与OPC UA的数据建模及传输方法、事件驱动的虚实映射方法、基于复杂事件处理的车间逻辑建模方法、信息可视化及推送,详细阐述了该可视化实时监控的实现方法.最后,以某航天产品机加车间为应用案例,结合实际生产过程和开发的原型系统,验证了该监控方法的有效性.

为了研究蜗杆砂轮展成法磨削齿轮的加工系统中不同因素对齿轮精度的影响规律,采用空间坐标系转换法构建了蜗杆砂轮磨削齿轮系统的运动学模型,并模拟分析了砂轮磨粒磨削齿面的运动空间轨迹.研究表明,砂轮廓形径向误差对齿轮精度影响较小.齿廓偏差与螺旋线偏差主要受砂轮廓形切向误差、安装夹角误差与机床传动比误差的影响,而齿距偏差主要受安装夹角误差、机床传动比误差的影响.此外,齿轮精度随工装误差、砂轮廓形误差、机床传动精度等影响因素误差值的增大而变差.

The application of graph convolutional networks (GCNs) to hyperspectral image (HSI) classification is a heavily researched topic. However, GCNs are based on spectral filters, which are computationally costly and fail to suppress noise effectively. In addition, the current GCN-based methods are prone to oversmoothing (the representation of each node tends to be congruent) problems. To circumvent these problems, a novel semi-supervised locality-preserving dense graph neural network (GNN) with autoregressive moving average (ARMA) filters and context-aware learning (DARMA-CAL) is proposed for HSI classification. In this work, we introduce the ARMA filter instead of a spectral filter to apply to GNNs. The ARMA filter can better capture the global graph structure and is more robust to noise. More importantly, the ARMA filter can simplify calculations compared with the spectral filter. In addition, we show that the ARMA filter can be approximated by a recursive method. Furthermore, we propose a dense structure, which not only implements the ARMA filter in the structure, but is also locality-preserving. Finally, we design a layerwise context-aware learning mechanism to extract the useful local information generated by each layer of the dense ARMA network. The experimental results on three real HSI datasets show that DARMA-CAL outperforms the compared state-of-the-art methods.

结合COMSOL软件中的薄层阻抗接口,采用提出的接触电阻计算模型模拟了焊补设备对1Cr1 8Ni9Ti不锈钢的电容储能点焊过程,得到的焊点熔核半径为148.9 μm,基体部位熔核高15.5 μm,该结果与实测得到的熔核半径153 μm、基体熔核高16.6 μm相差较小,证明了仿真方法的正确性.分析了动态电阻和接触电阻率的变化.动态电阻的变化因电容储能电源的放电特性可分为三个不同阶段.对于接触电阻率,在电极/补层接触区域熔核半径119 μm处会发生突变,表现为绝缘,在熔核区域则因熔点温度下屈服强度较小导致接触电阻率变化不明显.研究了动态电阻对模拟结果的影响,发现若不考虑动态电阻对电容放电的影响,熔核尺寸明显减小,得出了在电容储能电源仿真中不能因时间较短就将动态电阻设为定值的结论.本研究可为接触电阻的处理提供一种解决办法,对电容储能焊补设备的仿真提供借鉴.

准确的订单剩余完工时间预测有助于动态调整生产计划、优化制造过程,以满足订单产品按时交付的需求.订单剩余完工时间受到车间物料、设备、在制品等各类生产要素的综合影响,相关数据具有典型的大量、多维、高冗余的特点,有效的特征选择能够获得更高的预测精度.在构建候选特征集的基础上,提出了一种基于自组织映射(SOM)网络特征加权模糊C均值(FWFCM)的特征选择算法.通过构建SOM网络初始化FWFCM的聚类中心,减少后者对初始聚类中心的依赖;基于互信息计算特征权重,实现导向性特征聚类,根据聚类结果选择特征代表,构成高质量关键特征子集.以某机加工车间的生产数据为例,通过与其他4种特征选择算法的对比分析,验证了所提算法的有效性.

为了研究不同粗糙度表面、载荷与摩擦频率对20CrMnTi钢齿轮磨削表面的摩擦磨损性能影响规律和机理,分别在干接触状态与润滑接触状态下开展了往复滑动摩擦磨损试验.研究结果表明:摩擦系数随着接触表面粗糙度和载荷的增大而增大,但随着摩擦频率的增大而减小;磨损深度随着表面粗糙度、载荷和摩擦频率的增大而增大.因此,在干接触与润滑接触两种状态下,不同粗糙度表面、载荷与摩擦频率对摩擦学特性的影响机制均存在显著的差异.