Compact Joint Research Articles

Empowering lightweight video transformer via the kernel learning

AbstractVideo transformers achieve superior performance in video recognition. Despite the recent advances in video transformers, they still require substantial computation and memory resources. To cater for the computation efficiency, a kernel‐based video transformer is proposed, including: (1) a new formulation of the video transformer via the kernel learning is presented to better understand the individual components of it; (2) a lightweight Kernel‐based spatial–temporal multi‐head self‐attention block is explored to learn the compact joint spatial–temporal video feature; (3) an adaptive‐score position embedding method is conducted to promote the flexibility of video transformer. Experimental results on several action recognition datasets demonstrate the effectiveness of the proposed method. Only pretrained on ImageNet‐1K, the method achieves the preferable balance between computation and accuracy, while requiring 7 fewer parameters and 13 fewer floating point operations than other comparable methods.

Energy-Efficient MIL-53(Fe)/Sn3O4 Nanosheet Photocatalysts for Visible-Light Degradation of Toxic Organics in Wastewater

The fabrication of photocatalysts with preferable degradation ability is critical to purify wastewater. MIL-53(Fe), as a representative Fe-based metal–organic framework (Fe-MOF), has been widely used in photocatalysis due to its excellent intrinsic characteristics. However, the inadequate electron number in MIL-53(Fe) has restricted the ability of photodegradation of organic pollutants. Herein, desirable, energy-efficient, and environmentally friendly MIL-53(Fe)/Sn3O4 nanosheet photocatalysts were first constructed by the solvothermal method. The characterization results indicated that the MIL-53(Fe)/Sn3O4-5 (5 wt % Sn3O4) catalyst exhibited optimal photoelectric properties and maintained a relatively regular morphology. Moreover, the prepared MIL-53(Fe)/Sn3O4-5 exhibited preferable photodegradation ability for dyes and antibiotics under visible light, and the maximum photodegradation capacity for malachite green, methylene blue, tetracycline, ciprofloxacin, and ofloxacin could reach 96.9, 88.2, 70.3, 83.5, and 88.1%, respectively. The initial pH, photocatalyst dosage, and initial concentration were comprehensively investigated for degrading malachite green. In addition, we also explored the prepared catalyst for the removal of malachite green from different actual wastewater samples. After three cycles, the degradation efficiency of malachite green by MIL-53(Fe)/Sn3O4-5 decreased very little, exhibiting good reuse performance. The enhanced photodegradation efficiency of MIL-53(Fe)/Sn3O4-5 could be attributed to the construction of heterojunctions derived from a compact joint interface between Sn3O4 and Fe-MOF, which expedited the separation ability of photogenerated carriers and broadened the visible response range. The active species trapping experiments and electron spin-resonance results determined that many active species (h+, •OH, and •O2–) were involved in the photocatalytic process. Correspondingly, a convincible photodegradation mechanism between Sn3O4 and MIL-53(Fe) was proposed according to the experimental results. This study demonstrates that the fabricated Sn3O4 and MOF composites can provide a feasible solution for the purification of wastewater.

Active Vibration Control of a Large Flexible Appendage Using a Novel Joint Mechanism

With the evolution of space exploration, large flexible appendages have been developed in space structures. External perturbation or attitude maneuvering stimulates the vibration (low frequencies) of the aforementioned structures. In this research, a novel joint mechanism was developed in conjunction with active control to inhibit the low-frequency vibration of a large flexible appendage. A compact active joint, based on the idea of electromagnetic direct drive, was designed. The dynamic equations for the large flexible appendage system and the active joint were derived using the Lagrange function with the assumed-modes approach. Single- and multi-frequency excitations were simulated by two noncontact strategies for periodic vibration stimulation along the direction of rotation. The research results revealed that the interference signal had a primary frequency bandwidth of 0.07–0.63 Hz, and the vibration attenuation was prominent between 5.95 and 32.41 dB within the valid bandwidth. Effective inhibition of both the larger and the smaller amplitude vibrations at frequencies lower than 1 Hz could be realized using the proposed active joint without attachment of intelligent materials onto the flexible appendage surface.

GJTD-LR: A Trainable Grouped Joint Tensor Dictionary With Low-Rank Prior for Single Hyperspectral Image Super-Resolution

Reconstructing a high-resolution hyperspectral image (HR-HSI) by using a single low-resolution hyperspectral image (LR-HSI) is a significant technique for increasing the spatial resolution of HSIs and overcoming the physical limitation of the HSI sensor. Most single HSI super-resolution methods have achieved great success recently. However, owning to the difficulty of acquiring an HSI, the available training samples are relatively few, which will inevitably lead to relatively low performance. To address this issue, in the paper, we propose a novel single HSI super-resolution method by combining a trainable grouped joint tensor dictionary and a low-rank prior (GJTD-LR). First, we design a trainable grouped joint tensor dictionary, which can build an accurate mapping relationship between training HR-HSIs and their corresponding LR-HSIs with relatively few training samples. To be specific, the training HR-HSI and LR-HSI pairs are decomposed into a joint tensor dictionary and a set of sparse coefficients by using tensor-tensor product to fully preserve the spectral correlation. In addition, we apply a grouped strategy to divide the training images into several groups and learn a compact joint dictionary for each group. Second, a tensor low-rank model is forced into the reconstruction model to further capture the spatial correlation. At last, GJTD-LR is optimized by employing alternating direction method of multipliers (ADMM), soft threshold algorithm, singular value decomposition and fourier domain transform. The experimental results on both remote sensed HSIs and indoor HSIs show the superiority of GJTD-LR to some other traditional and advanced single HSI super-resolution methods.

Practical SM2-Based Multisignature Scheme with Applications to Vehicular Networks

In vehicular networks, the increasing value of transportation data and scale of connectivity also brings many security and privacy concerns. Peer authentication and message integrity are two vital security requirements to ensure safe transportation system. Because of the constrained resources of the units performing the cryptographic components, the proposed security-enhancing schemes should be lightweight and scalable. In this paper, we present a multisignature scheme derived from the SM2 signature which enables a group of parties to collaboratively sign a message and generate a compact joint signature at the end. Our scheme requires no preprocessing or interactions among the parties before signing, and its performance matches or surpasses known ones in terms of signing time, verification time, and signature size. Therefore, our scheme is also suitable for vehicular networks, with the goal to enhance security with small computation and storage cost.

Experimental investigations on the design and execution of shear loaded blind rivet joints in steel lightweight construction

AbstractThere is a barely manageable variety of blind rivets available on the market. This is because there is no universal product standard, which establishes the mechanical properties of blind rivets in a suitable manner. Whenever structural engineers dimension joints with blind rivets, they face a major challenge due to the lack of design values for the blind rivets themselves and the impractical application limits according to EN 1993‐1‐3. It is therefore necessary to determinate the design values of the blind rivets itself and for their applications in the field of steel lightweight construction in time and cost‐intensive experimental investigation. Furthermore, the regulations according to the ECCS, which allow the experimental‐based determination of design values for such applications, are insufficient, as they do not consider alternating stresses as well as compact joint dimensions.To close the gap of knowledge this paper presents results from systematic investigations on the shear and load‐bearing behavior of blind rivets and joints with blind rivets. Based on the investigations the authors suggest the extension of the current design and execution rules in order to allow a safe design by separate verification of the components and blind rivet.

Effect of Ankle Torque on the Ankle-Foot Orthosis Joint Design Sustainability.

The ankle joint of a powered ankle–foot orthosis (PAFO) is a prominent component, as it must withstand the dynamic loading conditions during its service time, while delivering all the functional requirements such as reducing the metabolic effort during walking, minimizing the stress on the user’s joint, and improving the gait stability of the impaired subjects. More often, the life of an AFO is limited by the performance of its joint; hence, a careful design consideration and material selection are required to increase the AFO’s service life. In the present work, a compact AFO joint was designed based on a worm gear mechanism with steel and brass counterparts due to the fact of its large torque transfer capability in a single stage, enabling a compact joint. Further, it provided an added advantage of self-locking due to the large friction that prevents backdrive, which is beneficial for drop-foot recovery. The design was verified using nonlinear finite element analysis for maximum torque situations at the ankle joint during normal walking. The results indicate stress levels within its design performance; however, it is recommended to select high-grade structural steel for the ankle shaft as the highest stresses in AFO were located on it.

Conditional out-of-distribution generation for unpaired data using transfer VAE.

While generative models have shown great success in sampling high-dimensional samples conditional on low-dimensional descriptors (stroke thickness in MNIST, hair color in CelebA, speaker identity in WaveNet), their generation out-of-distribution poses fundamental problems due to the difficulty of learning compact joint distribution across conditions. The canonical example of the conditional variational autoencoder (CVAE), for instance, does not explicitly relate conditions during training and, hence, has no explicit incentive of learning such a compact representation. We overcome the limitation of the CVAE by matching distributions across conditions using maximum mean discrepancy in the decoder layer that follows the bottleneck. This introduces a strong regularization both for reconstructing samples within the same condition and for transforming samples across conditions, resulting in much improved generalization. As this amount to solving a style-transfer problem, we refer to the model as transfer VAE (trVAE). Benchmarking trVAE on high-dimensional image and single-cell RNA-seq, we demonstrate higher robustness and higher accuracy than existing approaches. We also show qualitatively improved predictions by tackling previously problematic minority classes and multiple conditions in the context of cellular perturbation response to treatment and disease based on high-dimensional single-cell gene expression data. For generic tasks, we improve Pearson correlations of high-dimensional estimated means and variances with their ground truths from 0.89 to 0.97 and 0.75 to 0.87, respectively. We further demonstrate that trVAE learns cell-type-specific responses after perturbation and improves the prediction of most cell-type-specific genes by 65%. The trVAE implementation is available via github.com/theislab/trvae. The results of this article can be reproduced via github.com/theislab/trvae_reproducibility.

Underwater friction stir welding of Al-Mg alloy: Thermo-mechanical modeling and validation

The aim of this article is the in-depth analysis of material behavior during underwater friction stir welding (UFSW) of Al-Mg alloy. A thermo-mechanical model was employed to find the relation between physical changes during UFSW and mechanical properties of the final joint. To better understand the effects of the water environment, the results were compared with conventional friction stir welding (FSW) conditions. Thermal profile and surface heat flux results revealed that lower preheated area (PHA) at the leading edge (LE) decreased the total amount of frictional heat generation in the UFSWed joint. According to the simulation results, the strain rate in the stir zone (SZ) at the FSW sample was more than that of the UFSW specimen. Rapid cooling on submerged welding can result in finer grain structure at the UFSWed joint. The applied compression pressure on the workpiece at the trailing edge (TE) of the tool was more in the UFSW sample compared to the FSW, which led to higher compact joint line in the UFSWed specimen. These phenomena increased the joint line hardness and tensile properties at the UFSWed joint compared with the FSWed sample.

A new concept for developing a compact joint structure for reducing joint resistance between high-temperature superconductors (HTS) and low-temperature superconductors (LTS)

It is significantly difficult to develop a superconducting joint between REBCO and low-temperature superconductors (LTS) using a solder matrix replacement technique. This is because the REBCO superconducting layer is highly corrosive to Sn. In this work, TEM observations were first conducted on the reaction interface between the REBCO layer and Sn to reveal the reaction. Then, a new idea to create a compact low-resistance joint to reduce the joint resistance between REBCO and LTS was proposed. In this method, the REBCO tape is rolled in a metal boat, with the Cu stabilizer remaining around the tape. Then, an LTS wire, whose superconducting filament ends are coated with a superconducting solder, is set straight into a boat. Then, the boat is filled with a superconducting solder, so that the joint state between the LTS and superconducting solder matrix remains superconductive. However, the electrical joint between the REBCO tape and the solder matrix is resistive, even if the solder matrix is superconductive. Consequently, the overall joint resistance is determined by the boundary resistance between the REBCO tape and the superconducting solder matrix. However, to achieve a joint resistance below 10–10 Ω, a long joint length of more than 5 m, preferably more than 10 m, will be required. Considering the strain state of the REBCO layer when it is rolled in to a boat, the boat-type joint structure proposed in this work enables the joint size to be significantly compact, even if a length of more than 10 m is required. At present, a joint resistance of 0.7 nΩ was obtained in a field range of less than 0.4 T by using a boat with an inner size dimensions as follows: 50 mm length, 16 mm width, 7 mm height, radius of curvature of 8 mm, and tape length of 2 m.

Multi-channel Joint Sparse Learning Model for Non-rigid Three-dimensional Object Classification

Abstract In order to solve the issues of inadequate feature description and inefficient feature learning model existing in current classification methods, this article proposes a multi-channel joint sparse learning model for three-dimensional (3D) non-rigid object classification. First, the authors adopt a multi-level measurement of intrinsic properties to create complementary shape descriptors. Second, they build independent and informative bag of features (BoF) by embedding these shape descriptors into the visual vocabulary space. Third, a max-dependency and min-redundancy criterion is applied for optimal feature filtering on each BoF dictionary based on mutual information; meanwhile, each dictionary is learned and weighted according to its contribution to the classification task, and then a compact multi-channel joint sparse learning model is constructed. Finally, the authors train the joint sparse learning model followed by a Softmax classifier to implement efficient shape classification. The experimental results show that the proposed method has stronger feature representation ability and promotes greatly the discrimination of sparse coding coefficients. Thus, the promising classification performance and the powerful robustness can be obtained compared to the state-of-the-art methods.

Stress corrosion cracking for dissimilar capacitive discharge welding joint with varied surface preparation and inputted energy

In this article is presented dissimilar welding joint that joins steel and brass bars. Due to its varied melting point the welding process of dissimilar metal has a higher degree of difficulties. Naturally, welding process embeds residual stress and provides voids at least in micro scale. The voids then act as a trigger to develop initial cracks which will grow under external load. Once, the welding joint is exposed to corrosive environment, it will be susceptible to Stress-Corrosion-Cracking (SCC) failure. The quality of the welded joints is considered based on SCC resilience. Two parameters of the Capacitive Discharge Welding Joint which have prominent effects were evaluated: surface preparation and inputted energy. With a proper surface preparation it is hoped optimum welded joints can be obtained. The experiment results show that the oc1 surface preparation, i.e. the sharpened steel interface while the brass is kept on perpendicular edge, provides better performance under SCC load. Highest inputted energy (120J) always provides the longest duration for joint exposed to SCC load before the failure. The high inputted energy gives high heat to melt the interface which in turn produces compact joint at the interface.

Facile In-Situ Synthesis of Freestanding 3D Nanoporous Cu@Cu2O Hierarchical Nanoplate Arrays as Binder-Free Integrated Anodes for High-Capacity, Long-Life Li-Ion Batteries

Freestanding 3D nanoporous Cu-supported Cu2O hierarchical nanoplate arrays (3D NPC@Cu2O HNPAs) have been prepared in situ by facile one-step oxidation-assisted electrochemical dealloying, in which Cu2O HNPAs are characteristic of large-sized (hundreds of nm) 2D nanoplate arrays firmly embedded in small-sized (tens of nm) counterparts. The unique 3D nanocomposites as anodes for Li-ion batteries (LIBs) display superior Li storage properties involving ultrahigh specific capacity, long cycle life and excellent rate capability, which deliver a reversible capacity as high as 3.0 mAh cm−2 with 71.4 pct capacity retention after 450 long cycles at 2 mA cm−2. Even when the current density reaches 5 mA cm−2, an ultrahigh reversible capacity of 3.4 mAh cm−2 still can be achieved smoothly without obvious capacity decay after 250 cycles. It is totally comparable to or even exceeds the current level of a commercial graphite anode. The outstanding electrochemical performance can be largely ascribed to the unique 3D electrode structure comprising HNPAs and NP substrate, the large contact area between active material and electrolyte, in situ growth of active material upon the porous substrate, a compact joint of small-sized intermediate nanolayers and favorable mass transfer among vertical hierarchical nanoplates, indicative of a quite promising candidate as a binder-free integrated anode toward practical application of advanced LIBs.

A Compact Contactless Coaxial VHF Rotary Joint for Surveillance Radar

A new design of coaxial rotary joint (RJ) made compact by applying “folded” choke quarter-wave impedance transformer is proposed for very high-frequency surveillance radar. The final prototype, which measures L $388\,\,\text {mm} \times \text {D}\,\,106$ mm, demonstrated 0.04-dB insertion loss and 28-dB return loss within the operating frequency range of 150–170 MHz. The RJ is able to handle peak power of 15 kW during experiments. By avoiding reliability issues inherent in contacting RJ, the lifetime and stability of the whole radar system would be remarkably enhanced.

10 kA Joints for HTS Roebel Cables

Future high temperature superconductor (HTS) high field magnets using multitape HTS cables need 10-kA low-resistance connections. The connections are needed between the poles of the magnets and at the terminals in a wide-operating temperature range, from 1.9–85 K. The EuCARD-WP10 Future Magnets collaboration aims at testing HTS-based Roebel cables in an accelerator magnet. Usually, low temperature superconductor (LTS) cables are jointed inside a relatively short soldered block. Powering tests at CERN have highlighted excess heating of a joint following classical LTS joint design. The HTS Roebel cables are assembled from REBCO-coated conductor tapes in a transposed configuration. Due to this, the tapes surface the cable at an angle with the cable axis. A low-resistance joint requires a sufficiently large interface area for each tape. Within one twist pitch length, each tape is located at the surface of the cable over a relatively small non-constant area. This geometry prevents making a well-controlled joint in a compact length along the cable. This paper presents a compact joint configuration for the Roebel cable overcoming these practical challenges. A new joint called fin-block is designed. The joint resistance is estimated computationally. Finally, the test results as a function of current and temperature are presented.

Design and Development of a Compact High-Torque Robotic Actuator for Space Mechanisms

Space robots require compact joint drive systems (JDSs), typically comprising of actuator, transmission, joint elements that can deliver high torques through stiff mechanical ports. Today's conventional space drive systems are made from off-the-shelf actuators and multistage transmissions that generally involve three to six stages. This current practice has certain benefits such as short development time due to the availability of mechanical components. However, it lacks a system-level integration that accounts for the actuator structure, size and output force, transmission structure, gear-ratio, and strength, and often leads to long and bulky assemblies with large number of parts. This paper presents a new robotic hardware that integrates the robot's JDS into one compact device that is optimized for its size and maximum torque density. This is done by designing the robotic joint using a special transmission which, when numerically optimized, can produce unlimited gear-ratios using only two stages. The design is computerized to obtain all the solutions that satisfy its kinematic relationships within a given actuator diameter. Compared to existing robotic actuators, the proposed design could lead to shorter assemblies with significantly lower number of parts for the same output torque. The theoretical results demonstrates the potential of an example device, for which a proof-of-concept plastic mockup was fabricated, that could deliver more than 200 N·m of torque in a package as small as a human elbow joint. The proposed technology could have strong technological implications in other industries such as powered prosthetics and rehabilitation equipment.

Multiple self-exothermic reactions for room-temperature aluminum bonding via instantaneous melting

We propose a room-temperature multiple-reaction bonding method for aluminum, by combining the thermite reaction of Al-CuO powder and self-propagating formation reaction of Ti-B powder with no external heat source. The joint microstructure consists in various composites (Al2O3, TiB2 and Al-Cu intermetallic), and their distribution affects the joint bonding properties. The multiple reaction leads to a compact joint layer with a very low porosity, in contrast to joints formed only by a formation reaction. This observation is explained by the inflow of molten aluminum metal into the joint layer during the joint formation, a process which is central in the formation of a sound joint. Furthermore, bonding between different composites in the joint is achieved by lattice defects at their interface. Analysis of shear tests show that fracture occurs in regions at the metal base/joint interface where Al2O3 particles were segregated. It is concluded that appropriate reactant design plays a significant role in controlling the microstructure, and thus, the bonding properties of the joint.

Design of compact joint transform correlator

To improve the quality of images taken with spatial cameras, we measure sub-pixel shifts between two adjacent images, using a joint...

로봇의 회전관절을 위한 케이블 로드를 갖는 유압 구동기 설계 및 구현

This paper presents a cable-driven hydraulic actuator named Cable Rod Hydraulic Actuator (CRHA). The cable actuating system is attractive for designing a compact joint in robotic applications since it can be installed remotely from the joint. Recently, cable rods have been used in pneumatic area for inertia reduction. However, designing cable rods in hydraulics is challenging because it is difficult to achieve flexibility and endurance simultaneously under high pressure conditions. In this paper, the cable rod, which consists of a cable and jacket, is proposed to meet both requirements. To design the CRHA, we determined the design parameters, such as cylinder size, and selected the cable rod’s material by friction and leakage test. Finally, comparisons experiments about step and frequency responses with conventional hydraulic actuators were performed to assess feasibility for robotic joints, and the results show that the proposed system has good bandwidth and fast response as robotic joints.

Thermal Stability of Conductor Joint for JT-60SA Central Solenoid

Broader Approach activities accompany the ITER project, and these activities have been carried out between the EU and Japan. The type of the joint between the conductors of the JT-60 Super Advanced (JT-60SA) central solenoid (CS) is a butt joint. The butt joint is a compact joint, but it has limited stability because of poor helium cooling. In this paper, the butt joint of the JT-60SA CS is analyzed, for the temperature rise at the butt joint, by using the FEM simulation software COMSOL Multiphysics. As a result, when the plasma disruption occurs, the butt joint of the CS in the JT-60SA has enough high T CS margin even if the mass flow decreases from 6.0 to 4.0 g/s. These analyzed results become the fundamental data to evaluate the thermal stability of the coils and to protect coil systems from quench.