Shenyang Aerospace University

In recent years, two-dimensional atomic-level thickness crystal materials have attracted widespread interest such as graphene, hexagonal boron nitride (h-BN), silicene, germanium, black phosphorus (BP), transition metal sulfides and so on.

Highlights The eco-friendly shaddock peel-derived carbon aerogels were prepared by a freeze-drying method. Multiple functions such as thermal insulation, compression resistance and microwave absorption can be integrated into one material-carbon aerogel. Novel computer simulation technology strategy was selected to simulate significant radar cross-sectional reduction values under real far field condition. . Abstract Eco-friendly electromagnetic wave absorbing materials with excellent thermal infrared stealth property, heat-insulating ability and compression resistance are highly attractive in practical applications. Meeting the aforesaid requirements simultaneously is a formidable challenge. Herein, ultra-light carbon aerogels were fabricated via fresh shaddock peel by facile freeze-drying method and calcination process, forming porous network architecture. With the heating platform temperature of 70 °C, the upper surface temperatures of the as-prepared carbon aerogel present a slow upward trend. The color of the sample surface in thermal infrared images is similar to that of the surroundings. With the maximum compressive stress of 2.435 kPa, the carbon aerogels can provide favorable endurance. The shaddock peel-based carbon aerogels possess the minimum reflection loss value ( RL min ) of − 29.50 dB in X band. Meanwhile, the effective absorption bandwidth covers 5.80 GHz at a relatively thin thickness of only 1.7 mm. With the detection theta of 0°, the maximum radar cross-sectional (RCS) reduction values of 16.28 dB m 2 can be achieved. Theoretical simulations of RCS have aroused extensive interest owing to their ingenious design and time-saving feature. This work paves the way for preparing multi-functional microwave absorbers derived from biomass raw materials under the guidance of RCS simulations.

Abstract Alloys with ultra-high strength and sufficient ductility are highly desired for modern engineering applications but difficult to develop. Here we report that, by a careful controlling alloy composition, thermomechanical process, and microstructural feature, a Co-Cr-Ni-based medium-entropy alloy (MEA) with a dual heterogeneous structure of both matrix and precipitates can be designed to provide an ultra-high tensile strength of 2.2 GPa and uniform elongation of 13% at ambient temperature, properties that are much improved over their counterparts without the heterogeneous structure. Electron microscopy characterizations reveal that the dual heterogeneous structures are composed of a heterogeneous matrix with both coarse grains (10∼30 μm) and ultra-fine grains (0.5∼2 μm), together with heterogeneous L1 2 -structured nanoprecipitates ranging from several to hundreds of nanometers. The heterogeneous L1 2 nanoprecipitates are fully coherent with the matrix, minimizing the elastic misfit strain of interfaces, relieving the stress concentration during deformation, and playing an active role in enhanced ductility.

Despite the recent advances on intelligent fault diagnosis of rolling element bearings, existing research works mostly assume training and testing data are drawn from the same distribution. However, due to variation of operating condition, domain shift phenomenon generally exists, which results in significant diagnosis performance deterioration. To address cross-domain problems, latest research works preferably apply domain adaptation techniques on marginal data distributions. However, it is usually assumed that sufficient testing data are available for training, that is not in accordance with most transfer tasks in real industries where only data in machine healthy condition can be collected in advance. This paper proposes a novel cross-domain fault diagnosis method based on deep generative neural networks. By artificially generating fake samples for domain adaptation, the proposed method is able to provide reliable cross-domain diagnosis results when testing data in machine fault conditions are not available for training. The experimental results suggest that the proposed method offers a promising approach for industrial applications.

Reversible data hiding in encrypted images has attracted considerable attention from the communities of privacy security and protection. The success of the previous methods in this area has shown that a superior performance can be achieved by exploiting the redundancy within the image. Specifically, because the pixels in the local structures (like patches or regions) have a strong similarity, they can be heavily compressed, thus resulting in a large hiding room. In this paper, to better explore the correlation between neighbor pixels, we propose to consider the patch-level sparse representation when hiding the secret data. The widely used sparse coding technique has demonstrated that a patch can be linearly represented by some atoms in an over-complete dictionary. As the sparse coding is an approximation solution, the leading residual errors are encoded and self-embedded within the cover image. Furthermore, the learned dictionary is also embedded into the encrypted image. Thanks to the powerful representation of sparse coding, a large vacated room can be achieved, and thus the data hider can embed more secret messages in the encrypted image. Extensive experiments demonstrate that the proposed method significantly outperforms the state-of-the-art methods in terms of the embedding rate and the image quality.

Rotating machinery fault diagnosis problems have been well-addressed when sufficient supervised data of the tested machine are available using the latest data-driven methods. However, it is still challenging to develop effective diagnostic method with insufficient training data, which is highly demanded in real-industrial scenarios, since high-quality data are usually difficult and expensive to collect. Considering the underlying similarities of rotating machines, data mining on different but related equipments potentially benefit the diagnostic performance on the target machine. Therefore, a novel transfer learning method for diagnostics based on deep learning is proposed in this article, where the diagnostic knowledge learned from sufficient supervised data of multiple rotating machines is transferred to the target equipment with domain adversarial training. Different from the existing studies, a more generalized transfer learning problem with different label spaces of domains is investigated, and different fault severities are also considered in fault diagnostics. The experimental results on four datasets validate the effectiveness of the proposed method, and show it is feasible and promising to explore different datasets to improve diagnostic performance.

One of the challenges in hyperspectral image (HSI) classification is that there are limited labeled samples to train a classifier for very high-dimensional data. In practical applications, we often encounter an HSI domain (called target domain) with very few labeled data, while another HSI domain (called source domain) may have enough labeled data. Classes between the two domains may not be the same. This article attempts to use source class data to help classify the target classes, including the same and new unseen classes. To address this classification paradigm, a meta-learning paradigm for few-shot learning (FSL) is usually adopted. However, existing FSL methods do not account for domain shift between source and target domain. To solve the FSL problem under domain shift, a novel deep cross-domain few-shot learning (DCFSL) method is proposed. For the first time, DCFSL tackles FSL and domain adaptation issues in a unified framework. Specifically, a conditional adversarial domain adaptation strategy is utilized to overcome domain shift, which can achieve domain distribution alignment. In addition, FSL is executed in source and target classes at the same time, which can not only discover transferable knowledge in the source classes but also learn a discriminative embedding model to the target classes. Experiments conducted on four public HSI data sets demonstrate that DCFSL outperforms the existing FSL methods and deep learning methods for HSI classification. Our source code is available at <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/Li-ZK/DCFSL-2021</uri> .

MOTIVATION: Owing to its importance in both basic research (such as molecular evolution and protein attribute prediction) and practical application (such as timely modeling the 3D structures of proteins targeted for drug development), protein remote homology detection has attracted a great deal of interest. It is intriguing to note that the profile-based approach is promising and holds high potential in this regard. To further improve protein remote homology detection, a key step is how to find an optimal means to extract the evolutionary information into the profiles. RESULTS: Here, we propose a novel approach, the so-called profile-based protein representation, to extract the evolutionary information via the frequency profiles. The latter can be calculated from the multiple sequence alignments generated by PSI-BLAST. Three top performing sequence-based kernels (SVM-Ngram, SVM-pairwise and SVM-LA) were combined with the profile-based protein representation. Various tests were conducted on a SCOP benchmark dataset that contains 54 families and 23 superfamilies. The results showed that the new approach is promising, and can obviously improve the performance of the three kernels. Furthermore, our approach can also provide useful insights for studying the features of proteins in various families. It has not escaped our notice that the current approach can be easily combined with the existing sequence-based methods so as to improve their performance as well. AVAILABILITY AND IMPLEMENTATION: For users' convenience, the source code of generating the profile-based proteins and the multiple kernel learning was also provided at http://bioinformatics.hitsz.edu.cn/main/~binliu/remote/

Data-driven machinery fault diagnosis methods have been successfully developed in the past decades. However, the cross-domain diagnostic problems have not been well addressed, where the training and testing data are collected under different operating conditions. Recently, domain adaptation approaches have been popularly used to bridge this gap, which extract domain-invariant features for diagnostics. Despite the effectiveness, most existing methods assume the label spaces of training and testing data are identical that indicates the fault mode sets are the same in different scenarios. In practice, new fault modes usually occur in testing, which makes the conventional methods focusing on marginal distribution alignment less effective. In order to address this problem, a deep learning-based open-set domain adaptation method is proposed in this study. Adversarial learning is introduced to extract generalized features, and an instance-level weighted mechanism is proposed to reflect the similarities of testing samples with known health states. The unknown fault mode can be effectively identified, and the known states can be also recognized. Entropy minimization scheme is further adopted to improve generalization. Experiments on two practical rotating machinery datasets validate the proposed method. The results suggest the proposed method is promising for open-set domain adaptation problems, which largely enhances the applicability of data-driven approaches in the real industries.

In the past years, the practical cross-domain machinery fault diagnosis problems have been attracting growing attention, where the training and testing data are collected from different operating conditions. The recent advances in closed-set domain adaptation have well addressed the basic problem where the fault mode sets are identical in the source and target domains. While some attempts have also been made on the partial and open-set domain adaptations, no prior information of the target-domain fault modes can be usually available in the real industries, that forms a challenging problem in transfer learning. This article proposes a universal domain adaptation method for fault diagnosis, where no explicit assumption is made on the target label set. A hybrid approach with source class-wise and target instance-wise weighting mechanism is proposed for selective adaptation. By using additional outlier identifier, the proposed method can automatically recognize the unknown fault modes while achieving class-level alignments for the shared health states, without knowing the target label set. Experiments on two rotating machine datasets validate the proposed method, which is promising for practical applications under strong data uncertainties.

Abstract Graphene oxide is extensively compounded with polymers toward a wide variety of applications. Less studied are few‐layer or multi‐layer highly crystalline graphene, both of which are herein named as graphene platelets. This article aims to provide the most recent advancements of graphene platelets and their polymer composites. A first focus lies on cost‐effective fabrication strategies of graphene platelets – intercalation and exfoliation – which work in a relative mass scale, e.g., 5.3 g h −1 . As no heavy oxidization is involved, the platelets have high crystalline integrity, e.g., C:O ratio over 8.0, with thicknesses 2–4 nm and lateral dimension up to a few micrometers. Through carefully selecting the solvent for dispersion and the molecules for surface modification, graphene platelets can be liquid‐processable, enabling them to be printed, coated, or compounded with various polymers. A purpose‐designed experiment is undertaken to unravel the effect of reasonable ultrasonication time on the platelet thickness. Typical polymer/graphene platelet composites are critically examined for their preparation, structure, and applications such as thermal management and flexible/stretchable electronic devices. Perspectives on the limitations, current challenges, and future prospects for graphene platelets and their polymer composites are provided.

The universal application of wearable strain sensors in various situations for human-activity monitoring is considerably limited by the contradiction between high sensitivity and broad working range. There still remains a huge challenge to design sensors featuring simultaneous broad working range and high sensitivity. Herein, a typical bilayer-conductive structure Ti3C2Tx MXene/carbon nanotubes (CNTs)/thermoplastic polyurethane (TPU) composite film was developed by a simple and scalable vacuum filtration process utilizing a porous electrospun thermoplastic polyurethane (TPU) mat as a skeleton. The MXene/CNTs/TPU strain sensor is composed of two parts: a brittle densely stacked MXene upper lamella and a flexible MXene/CNT-decorated fibrous network lower layer. Benefiting from the synergetic effect of the two parts along with hydrogen-bonding interactions between the porous TPU fiber mat and MXene sheets, the MXene/CNTs/TPU strain sensor possesses both a broad working range (up to 330%) and high sensitivity (maximum gauge factor of 2911) as well as superb long-term durability (2600 cycles under the strain of 50%). Finally, the sensor can be successfully employed for human movement monitoring, from tiny facial expressions, respiration, and pulse beat to large-scale finger and elbow bending, demonstrating a promising and attractive application for wearable devices and human–machine interaction.

The first multiphysical invisible sensor is theoretically and experimentally presented. An ultrathin, homogeneous, and isotropic shell is designed to simultaneously manipulate heat flux and DC current and eliminate the multiphysical perturbation, while maintaining the receiving and transmitting properties of the sensor.

A morphing aircraft can adapt its configuration to suit different types of tasks, which is also an important requirement of Unmanned Aerial Vehicles (UAV). The successful development of an unmanned morphing aircraft involves three steps that determine its ability and intelligent: configuration design, dynamic modeling and flight control. This study conducts a comprehensive survey of morphing aircraft. First, the methods to design the configuration of a morphing aircraft are presented and analyzed. Then, the nonlinear dynamic characteristics and aerodynamic interference caused by a morphing wing are described. Subsequently, the dynamic modeling and flight control methods for solving the flight control problems are summarized with respect to these features. Finally, the general as well as special challenges ahead of the development of intelligent morphing aircraft are discussed. The findings can provide a theoretical and technical reference for designing future morphing aircraft or morphing-wing UAVs.

As viewpoint issue is becoming a bottleneck for human motion analysis and its application, in recent years, researchers have been devoted to view-invariant human motion analysis and have achieved inspiring progress. The challenge here is to find a methodology that can recognize human motion patterns to reach increasingly sophisticated levels of human behavior description. This paper provides a comprehensive survey of this significant research with the emphasis on view-invariant representation, and recognition of poses and actions. In order to help readers understand the integrated process of visual analysis of human motion, this paper presents recent development in three major issues involved in a general human motion analysis system, namely, human detection, view-invariant pose representation and estimation, and behavior understanding. Public available standard datasets are recommended. The concluding discussion assesses the progress so far, and outlines some research challenges and future directions, and solution to what is essential to achieve the goals of human motion analysis.

In the recent years, data-driven machinery fault diagnostic methods have been successfully developed, and the tasks where the training and testing data are from the same distribution have been well addressed. However, due to sensor malfunctions, the training and testing data can be collected at different places of machines, resulting in the feature space with significant distribution discrepancy. This challenging issue has received less attention in the current literature, and the existing approaches generally fail in such scenarios. This article proposes a domain adaptation method for machinery fault diagnostics based on deep learning. Adversarial training is introduced for marginal domain fusion, and unsupervised parallel data are explored to achieve conditional distribution alignments with respect to different machine health conditions. Experiments on two rotating machinery datasets are carried out for validations. The results suggest the proposed method is promising to address the fault diagnostic tasks with data from different places of machines, further enhancing applicability of data-driven methods in real industries.

Chemical changes of Chinese lignite upon drying in superheated steam, microwave, and hot air have been studied in this paper using the Fourier transform infrared (FTIR) spectroscopy technique. The infrared (IR) spectra of raw and dried samples were curve-fitted to a series of bands in aliphatic hydrogen (3000–2800 cm–1) and carbonyl absorption (1850–1500 cm–1) zones. It has been found that aliphatic hydrogen absorbance decreased slightly with an increasing temperature during superheated steam drying, while absorption of carboxyl (COOH) and carbonyl (C═O) groups decreased drastically, indicative of the loss of oxygen functionalities with an increasing drying temperature. During steam drying, aromatic carbon and aromatic ring stretch absorption remained relatively unchanged up to 250 °C and decreased significantly thereafter because of some pyrolysis reactions that took place at higher drying temperatures. Microwave heating of lignite resulted in a significant decrease in the concentration of oxygen-containing functional groups. Aromatic carbon remained relatively unchanged under microwave drying conditions, while aliphatic hydrogen decreased slightly. The aromaticity of coal calculated from curve-fitted spectra of deconvoluted peaks showed a progressive increase with an increasing drying intensity under both steam and microwave drying conditions. Under air drying conditions, aliphatic hydrogen absorbance decreased drastically at 250 °C, while aromatic carbon remained unchanged. It was observed that oxidation in air mainly took place on aliphatic hydrogen sites, especially on methylene groups. Changes of carboxyl and carbonyl groups during air-dried samples showed a different trend compared to those dried in steam and microwave, increasing gradually up to 150 °C and then a sharp increase at 200 °C. The absorption of these groups decreased significantly at an increased air temperature up to 250 °C.

In the past years, deep learning-based machinery fault diagnosis methods have been successfully developed, and the basic diagnostic problems have been well addressed where the training and testing data are collected under the same operating conditions. When the training and testing data are from different distributions, domain adaptation approaches have been introduced. However, the existing methods generally assume the availability of the target-domain data in all the health conditions during training, which is not in accordance with the real industrial scenarios. This article proposes a deep learning-based fault diagnosis method to address the partial domain adaptation problems, where the unsupervised target-domain training data do not cover the full machine health state label space. The conditional data alignment and unsupervised prediction consistency schemes are proposed to achieve partial domain adaptation. The experimental results on two rotating machinery datasets suggest the proposed method offers a promising tool for this practical industrial problem.

Vehicular computation offloading is a well-received strategy to execute delay-sensitive and/or compute-intensive tasks of legacy vehicles. The response time of vehicular computation offloading can be shortened by using mobile edge computing that offers strong computing power, driving these computation tasks closer to end users. However, the quality of communication is hard to guarantee due to the obstruction of dense buildings or lack of infrastructure in some zones. Unmanned Aerial Vehicles (UAVs), therefore, have become one of the means to establish communication links for the two ends owing to its characteristics of ignoring terrain and flexible deployment. To make a sensible decision of computation offloading, nevertheless vehicles need to gather offloading-related global information, in which Software-Defined Networking (SDN) has shown its advances in data collection and centralized management. In this paper, thus, we propose an SDN-enabled UAV-assisted vehicular computation offloading optimization framework to minimize the system cost of vehicle computing tasks. In our framework, the UAV and the Mobile Edge Computing (MEC) server can work on behalf of the vehicle users to execute the delay-sensitive and compute-intensive tasks. The UAV, in a meanwhile, can also be deployed as a relay node to assist in forwarding computation tasks to the MEC server. We formulate the offloading decision-making problem as a multi-players computation offloading sequential game, and design the UAV-assisted Vehicular computation Cost Optimization (UVCO) algorithm to solve this problem. Simulation results demonstrate that our proposed algorithm can make the offloading decision to minimize the Average System Cost (ASC).

SDVN is a promising architecture to extend the computation resources which break through the limitations of current vehicular networks. It is possible to learn new networking schemes by observing the surrounding environment in SDVN. However, within SDVN, the construction and application of such schemes still lack proper consideration in data collection, prediction, verification, and validation before applying these schemes in the real network, which is due to the limited knowledge of the physical environment. Intelligent Digital Twin (IDT) was initially designed for realizing intelligent manufacturing by virtualizing and learning the data of the physical space in cyberspace. Hence, bringing IDT to networking can provide additional valuable functionalities to meet the above considerations by constructing a virtual intelligent network space, aiming to realize the iterative update of the networking schemes in an adaptive way. In this article, we introduce a new network architecture, IDT-SDVN, by maximizing the advantages of SDVNs. We present the challenges and open issues of IDT-SDVNs. A case study is presented to demonstrate the effectiveness of SDVNs. The experimental results show that significant improvement of performance is achieved for vehicular networking with the proposed IDT-SDVNs.