Search

Rank minimization is widely used to extract low-dimensional subspaces. As a convex relaxation of the rank minimization, the problem of nuclear norm minimization has been attracting widespread attention. However, the standard nuclear norm minimization usually results in overcompression of data in all subspaces and eliminates the discrimination information between different categories of data. To overcome these drawbacks, in this article, we introduce the label information into the nuclear norm minimization problem and propose a labeled-robust principal component analysis (L-RPCA) to realize nuclear norm minimization on multisubspace data. Compared with the standard nuclear norm minimization, our method can effectively utilize the discriminant information in multisubspace rank minimization and avoid excessive elimination of local information and multisubspace characteristics of the data. Then, an effective labeled-robust regression (L-RR) method is proposed to simultaneously recover the data and labels of the observed data. Experiments on real datasets show that our proposed methods are superior to other state-of-the-art methods.

Poor ductility is the primary concern of magnesium matrix composites (MMCs) inflicted by non-deformable ceramic particle reinforcements. Metal particles which melt at elevated temperature can be used as reinforcement to improve the deformation characteristics. Ti-6Al-4V particles reinforced AZ31 MMCs were produced through friction stir processing (FSP) which was carried out in a traditional vertical milling machine. The microstructural features as well as the response to external tensile load were explored. A homogenous distribution of Ti-6Al-4V was achieved at every part of the stir zone. There was no chemical decomposition of Ti-6Al-4V. Further, Ti-6Al-4V did not react with Al and Zn present in AZ31 alloy to form new compounds. A continuous strong interface was obtained around Ti-6Al-4V particle with the matrix. Ti-6Al-4V particles underwent breakage during processing due to severe plastic strain. There was a remarkable refinement of grains in the composite caused by dynamic recrystallization in addition to the pinning of smaller size broken particles. Dense dislocations were observed in the matrix because of plastic deformation and the associated strain misfit. Ti-6Al-4V particles improved the tensile behavior and assisted to obtain appreciable deformation before fracture. Brittle mode of failure was avoided.

Sharing private parking spaces during their idle time periods has shown great potential for addressing urban traffic congestion and illegitimate parking problems in smart cities. In this article, aiming to address the online parking-space sharing issue while ensuring the privacy of customer parking destination locations, we propose a novel destination privacy-preserving online parking sharing (DPOPS) incentive scheme. In particular, the online parking-space sharing problem is formalized as a social welfare maximization problem in a two-sided market, where parking-space providers (PSPs) and customers are regarded as sellers and buyers. Then, novel threshold value-based rules are designed to determine winners, payments, and reimbursement. Finally, winners are matched by solving a mixed-integer nonlinear programming problem, aiming to minimize the distance between customer's destination and allocated parking space. In addition, the location privacy of the customers' destinations is protected by the Laplace mechanism. We prove that DPOPS achieves several economically effective properties and approximate differential privacy. We analyze the upper bound of the efficiency loss of our scheme. Extensive evaluation results demonstrate that our scheme can not only achieve good performance regarding social welfare, PSP satisfaction ratio, privacy preservation, and computation overhead but also leads to shorter travel distances for customers comparing to the baseline scheme.

In the 21st century, climate change is a severe threat to the survival of the planet and humanity; to counter this threat, the significant role of renewable energy cannot be denied for sustainable development worldwide. The motive of this study is to examine the threshold level and asymmetric relationship between renewable energy and economic development in different regimes of the different income levels of 122 countries from 1995-2018 by using the novel technique Panel Smooth Transition Regression Model (PSTR). The results of the PSTR model suggest that the relationship between Renewable Energy and the Economic Growth of these selected countries is nonlinear. Furthermore, in most countries, there is a significant and positive relationship between renewable energy and economic growth in the high regime, except the higher-income nations has a significant and positive relationship in both low and high regimes. Overall full sample shows a significant and positive relationship between renewable energy and economic growth. Recommendations based on this study for these countries include: 1) energy-efficient resources should be adopted to increase renewable energy consumption. 2) The conventional energy consumption share should be reduced to improve the environmental quality for a sustainable future.

Due to the existing effects of intermittent jumps of unknown parameters during operation, effectively establishing transient and steady-state tracking performances in control systems with unknown intermittent actuator faults is very important. In this article, two prescribed performance adaptive neural control schemes based on command-filtered backstepping are developed for a class of uncertain strict-feedback nonlinear systems. Under the condition of system states being available for feedback, the state feedback control scheme is investigated. When the system states are not directly measured, a cascade high-gain observer is designed to reconstruct the system states, and in turn, the output feedback control scheme is presented. Since the projection operator and modified Lyapunov function are, respectively, used in the adaptive law design and stability analysis, it is proven that both schemes can not only ensure the boundedness of all closed-loop signals but also confine the tracking errors within prescribed arbitrarily small residual sets for all the time even if there exist the effects of intermittent jumps of unknown parameters. Thus, the prescribed system transient and steady-state performances in the sense of the tracking errors are established. Furthermore, we also prove that the tracking performance under output feedback is able to recover the tracking performance under state feedback as the observer gain decreases. Simulation studies are done to verify the effectiveness of the theoretical discussions.

Effect of sulfate-reducing bacteria on the corrosion behaviors of API-X80 pipeline steel has been studied in crevices under simulated disbonded coatings in red soil solution. The results show that there are amounts of SRB in the crevice under the simulated disbonded coating in the red soil solution during the whole period of experiment. The electrochemical impedance spectroscopy shows that SRB can enhance the corrosion of the steel in the crevice under the simulated disbonded coating, and the thermodynamic analyses indicate that SRB can accelerate the corrosion of steel using not only sulfate but also ferric hydroxide as electron acceptors.

Representation learning (RL) has gained increasingly considerable attention in intelligent fault diagnosis due to its great capability of automatically learning useful features. Most existing studies focus on developing various variants about RL through modifying loss function of original versions, whereas it is a challenging task. Using the promising sparse filtering as the basic module, this paper presents a simple and effective RL method called multi-scale and multi-pooling sparse filtering (MSMPSF). Instead of modifying loss function of sparse filtering, we simply introduce two fusion mechanisms into sparse filtering, i.e., multi-scale fusion and multi-pooling fusion. In detail, the former aims to learn different local features from the collected signals under multiple scales. The latter tries to fuse various local features using multiple poolings. With these two mechanisms, MSMPSF is capable of capturing complementary fault information hidden in raw signals of several scales and obtaining richly informative feature representations, hence it can perform better. The proposed method is evaluated through experiments on three datasets about gear and bearing. Extensive comparison results confirm that both of two mechanisms facilitate a significant improvement on diagnosis performance. Furthermore, our method receives very reliable and competitive results in terms of diagnosis accuracy and stability in comparison with existing related works.

Early detection of signet ring cell carcinoma (SRCC) can significantly improve patient survival rate. Pathological image analysis is applied as the golden standard for SRCC diagnosis. Automatic detection of pathological cells with deep learning methods can greatly reduce the burden of pathologists. Deep learning methods are commonly trained using large amounts of annotated data. However, due to the uneven distribution of medical resources and tedious manual examination procedure of high-resolution images, annotation data are usually insufficient and incomplete for deep learning model training. In this paper, we propose a new semi-supervised deep convolutional framework to address the data annotation problem for signet ring cell detection. Specifically, we propose a self-training strategy to generate pseudo bounding boxes based on Test Time Augmentation and modified Non-Maximum Suppression to re-train our detector. Our framework achieves 0.8774 in Valid Recall and 100.00 in FPs, winning the 1 st place in the signet ring cell detection task of the Digestive-System Pathological Detection and Segmentation Challenge 2019. Code has been made publicly available at: https://github.com/ooooverflow/DigestPath2019.

Intrinsic partial occupation of Cu induces strong phonon scattering for an extremely low lattice thermal conductivity in TmCuTe2, which leads this material to be a promising candidate for thermoelectric applications. This work focus on the electronic and phononic transport properties of TmCuTe2 with Ag-alloying on the Cu site. Such a Ag/Cu substitution is found to enable a decrease in Hall carrier concentration from 14 × 1019 to 8 × 1019 cm−3 for an evaluation of the charge transport based on a single parabolic band (SPB) model with acoustic scattering. In addition, Ag-substitution simultaneously introduces Cu/Ag point defects for extra phonon scattering, leading the lattice thermal conductivity to be as low as ∼0.25 W/m-K in a broad temperature. The optimization in carrier concentration and the reduction in lattice thermal conductivity contribute to a ∼40% improvement in average thermoelectric figure of merit (zT). The determination of band parameters enables a guidance for further advancements in this promising thermoelectric material.

The intelligent question answering system aims to provide quick and concise feedback on the questions of users. Although the performance of phrase-level and numerous attention models have been improved, the sentence components and position information are not emphasized enough. This article combines Ci-Lin and word2vec to divide all of the words in the question-answer pairs into groups according to the semantics and select one kernel word in each group. The remaining words are common words and realize the semantic mapping mechanism between kernel words and common words. With this Chinese semantic mapping mechanism, the common words in all questions and answers are replaced by the semantic kernel words to realize the normalization of the semantic representation. Meanwhile, based on the bi-directional LSTM model, this article introduces a method of the combination of semantic role labeling and positional context, dividing the sentence into multiple semantic segments according to semantic logic. The weight is given to the neighboring words in the same semantic segment and propose semantic role labeling position attention based on the bi-directional LSTM model (BLSTM-SRLP). The good performance of the BLSTM-SRLP model has been demonstrated in comparative experiments on the food safety field dataset (FS-QA).

Li-rich layered oxides have attracted intense attention for lithium-ion batteries, as provide substantial capacity from transition metal cation redox simultaneous with reversible oxygen-anion redox. However, unregulated irreversible oxygen-anion redox leads to critical issues such as voltage fade and oxygen release. Here, we report a feasible NiFe2O4 (NFO) surface-coating strategy to turn the nonbonding coordination of surface oxygen into metal-oxygen decoordination. In particular, the surface simplex M-O (M = Ni, Co, Mn from MO6 octahedra) and N-O (N = Ni, Fe from NO6 octahedra) bonds are reconstructed in the form of M-O-N bonds. By applying both in operando and ex situ technologies, we found this heterostructural interface traps surface lattice oxygen, as well as restrains cation migration in Li-rich layered oxide during electrochemical cycling. Therefore, surface lattice oxygen behavior is significantly sustained. More interestingly, we directly observe the surface oxygen redox decouple with cation migration. In addition, the NFO-coating blocks HF produced from electrolyte decomposition, resulting in reducing the dissolution of Mn. With this strategy, higher cycle stability (91.8% at 1 C after 200 cycles) and higher rate capability (109.4 mA g-1 at 1 C) were achieved in this work, compared with pristine Li-rich layered oxide. Our work offers potential for designing electrode materials utilizing oxygen redox chemistry.

In the past decade, single image super-resolution (SISR) based on convolutional neural networks (CNNs) has been represented remarkable performance. Powerful characterization of CNN is important for recent methods to learn an intricate non-linear mapping between high-resolution and corresponding low-resolution images. However, a deeper and wider network structure brings superior performance while increasing the number of network parameters and calculations so that it is difficult to handle the real-time information. Hence, it can be embedded in mobile devices with difficulty. Inspired by the above motivation, a lightweight network for the real-time SISR is proposed by stacking efficient cascading residual blocks, which consist of several concatenate effective modules with wide activation. To further improve the network performance, with the increase of a slight number of parameters, the proposed network cooperates with a lightweight residual efficient channel attention module to capture feature interaction between channels. Extensive experiments provide significant demonstrations that the proposed network obtains the superior trade-off between performance and parameters compared with other current methods. The lightweight trait of our method allows it to implement real-time image processing and can be embedded in mobile devices.

High-efficient heat dissipation is essential to maintain the best performance of the electron devices. A new electron thermal conductive route is first proposed and constructed for the graphene metal matrix composites in this work. Functionalization of graphene with conjugated 4-ethynylaniline (FGr) and deposition of FGr and Cu2+ ions onto the copper surface by pulsed-current co-electrodeposition are successively employed to prepare the FGr/Cu composite. The resulting thermal diffusivity of FGr/Cu is high as 1.444 cm2 s−1 at 100 °C, which is corresponding to the thermal conductivity of 497 W m−1 K−1 and higher than the reported values via single phonon thermal conductive route. FGr/Cu can maintain a superior and stable thermal conductive property at high temperature with a multiple of 1.61 and 1.31 at 100 °C and 150 °C compared to the Cu, respectively. The delocalized conjugated π bond in the graphene/Cu interface is successfully established through the bridging of graphene and metal with conjugated organic molecule possessing p orbits, which leads to the construction of a new electron thermal conductive route. This innovation presents an entirely new direction to develop graphene-based metal matrix composite with high-efficient heat dissipation.

In this work, MgCo2O4@NiMn layered double hydroxide (LDH) core-shell structured nanocomposites on Ni foam (NF) are synthesized by facile hydrothermal and calcination methods. MgCo2O4/NF is synthesized first via a hydrothermal reaction and annealing treatment, and then utilized to prepare MgCo2O4@NiMn-LDH/NF core-shell structured nanocomposites via the second hydrothermal process. It is found that the MgCo2O4@NiMn-LDH/NF nanocomposite prepared from 6 h hydrothermal reaction (MC@NM-LDH-2) exhibits an excellent specific capacitance of 3757.2 F g−1 (at 1 A g−1). Moreover, a high capacitance retention (86.9% after 6000 cycles) and a low internal resistance (Rs) (0.565 Ω) can be achieved. Furthermore, an all-solid-state asymmetric supercapacitor (ASC) is assembled using MgCo2O4@NiMn-LDH/NF-2 as positive electrode and activated carbon (AC) as negative electrode. The as-fabricated MgCo2O4@NiMn-LDH/NF-2//AC ASC shows a high energy density of 62.33 Wh kg−1 at 750 W kg−1. Meanwhile, the MgCo2O4@NiMn-LDH/NF-2//AC ASC device possesses an outstanding cycling stability of 93.7% retention of the initial capacitance after 6000 cycles and three ASC devices connected in series can light up a LED bulb for 15 min. Our results manifest that these core-shell structure MgCo2O4@NiMn-LDH nanocomposites could envision huge potential application in energy storage devices.

The industrial control and automation systems have played an increasingly important role in critical manufacturing processes. In such systems, many Internet of Things devices continuously collect large number of streaming data for real-time processing. Verifiable data streaming (VDS) addresses such authenticity issue for streaming data, but most VDS schemes are not efficient and lightweight, do not support range querying, and cannot be used in practice. To improve the efficiency and achieve a verifiable range query in data streaming, we present here a new primitive, namely, a chameleon authentication tree with prefixes (PCAT), which is extended from the PBTree and chameleon authentication tree. Our scheme is not only lightweight but also supports dynamic expansion and verifiable range query in data streaming, making it more suitable for resource-constrained devices. We separate the PCAT's algorithms into the following phases: initialization, data appending, query, and verification. Our analyses prove that the PCAT satisfies all the security requirements of VDS. Moreover, an efficiency analysis and performance evaluation demonstrate that our scheme not only supports lightweight data streaming authentication but also has high efficiency, which means that the PCAT is easier to apply in the industrial control and automation systems.

The real-time and dissemination characteristics of network information make net-mediated public opinion become more and more important food safety early warning resources, but the data of petabyte (PB) scale growth also bring great difficulties to the research and judgment of network public opinion, especially how to extract the event role of network public opinion from these data and analyze the sentiment tendency of public opinion comment. First, this article takes the public opinion of food safety network as the research point, and a BLSTM-CRF model for automatically marking the role of event is proposed by combining BLSTM and conditional random field organically. Second, the Attention mechanism based on vocabulary in the field of food safety is introduced, the distance-related sequence semantic features are extracted by BLSTM, and the emotional classification of sequence semantic features is realized by using CNN. A kind of Att-BLSTM-CNN model for the analysis of public opinion and emotional tendency in the field of food safety is proposed. Finally, based on the time series, this article combines the role extraction of food safety events and the analysis of emotional tendency and constructs a net-mediated public opinion early warning model in the field of food safety according to the heat of the event and the emotional intensity of the public to food safety public opinion events.

The practical applications for the distinctive functions of metamagnetic Heusler alloys, such as magnetic shape memory effect, various caloric effects, etc., strongly depend on the phase transition temperatures. Here, flexible Heusler alloy Ni-Mn-Sn films have been deposited on mica substrates by pulsed laser deposition with a Ti buffer layer. Clear ferromagnetic (FM) transition followed by the martensitic transformation at around room temperature and exchange bias (EB) with a blocking temperature of 70 K are observed. Under the application of both tensile and compressive strains by bending the mica substrates, all the characteristic temperatures of Ni-Mn-Sn films, including the FM transition temperature, martensitic transformation temperature, and blocking temperature of EB, are significantly increased by about 10 K. Furthermore, EB field and coercivity are both strongly strengthened, which is mainly caused by the simultaneous enhancement of FM and anti-FM Mn-Mn coupling because of their shortened separations by strain and verified by the Monte Carlo simulation results. The strain controlling for structural and magnetic properties provides efficient manipulation for Heusler alloy-based magnetic devices.

This paper mainly focuses on the problems of attack detection and reconstruction of cyber physical systems (CPS) under dynamic load altering attacks (D-LAA). Taking the smart grid as an example, firstly, the power system subjected to the attack is modeled as a power CPS through system characteristics to provide conditions for attack detection. On this basis, we assume that two vulnerable loads are attacked simultaneously to investigate the impact of the attack on the system. Secondly, a robust sliding mode observer is designed to implement attack reconstruction with the residual signal generated by the observer. After that, an attack detection logic is applied to complete the attack detection by comparing the threshold with the residual. Finally, a three generators and six buses power system is given as an example to verify the feasibility of attack detection and reconstruction by using the real-time simulation platform built by StarSim and MT respectively. The simulation results are given in this paper.

Background The key challenge to constructing functional corticomuscular coupling (FCMC) is to accurately identify the direction and strength of the information flow between scalp electroencephalography (EEG) and surface electromyography (SEMG). Traditional TE and TDMI methods have difficulty in identifying the information interaction for short time series as they tend to rely on long and stable data, so we propose a time-delayed maximal information coefficient (TDMIC) method. With this method, we aim to investigate the directional specificity of bidirectional total and nonlinear information flow on FCMC, and to explore the neural mechanisms underlying motor dysfunction in stroke patients. Methods We introduced a time-delayed parameter in the maximal information coefficient to capture the direction of information interaction between two time series. We employed the linear and non-linear system model based on short data to verify the validity of our algorithm. We then used the TDMIC method to study the characteristics of total and nonlinear information flow in FCMC during a dorsiflexion task for healthy controls and stroke patients. Results The simulation results showed that the TDMIC method can better detect the direction of information interaction compared with TE and TDMI methods. For healthy controls, the beta band (14–30 Hz) had higher information flow in FCMC than the gamma band (31–45 Hz). Furthermore, the beta-band total and nonlinear information flow in the descending direction (EEG to EMG) was significantly higher than that in the ascending direction (EMG to EEG), whereas in the gamma band the ascending direction had significantly higher information flow than the descending direction. Additionally, we found that the strong bidirectional information flow mainly acted on Cz, C3, CP3, P3 and CPz. Compared to controls, both the beta-and gamma-band bidirectional total and nonlinear information flows of the stroke group were significantly weaker. There is no significant difference in the direction of beta- and gamma-band information flow in stroke group. Conclusions The proposed method could effectively identify the information interaction between short time series. According to our experiment, the beta band mainly passes downward motor control information while the gamma band features upward sensory feedback information delivery. Our observation demonstrate that the center and contralateral sensorimotor cortex play a major role in lower limb motor control. The study further demonstrates that brain damage caused by stroke disrupts the bidirectional information interaction between cortex and effector muscles in the sensorimotor system, leading to motor dysfunction.

Aspect-based sentiment analysis (ABSA) is a fine-grained sentiment analysis task designed to predict the sentiment polarity of each aspect term in a text. Recent research mainly uses neural networks to model text and utilizes attention mechanisms to interact for associate aspect terms and context to obtain more effective feature representation. However, the general attention mechanism is easy to lose the original information. Besides, in the multi-aspect text, the sentiment information of other aspect terms interferes with the sentiment analysis of the current aspect term likely. In this paper, we propose two models named RA-CNN and RAO-CNN for ABSA tasks. In RA-CNN, we apply CNN to model the aspect term and utilize a specially designed residual attention mechanism to interact with the text. Based on the RA-CNN, RAO-CNN adds other aspect terms module, which can reduce interference of sentiment information related to other aspect terms in the multi-aspect text. To verify the proposed models’ effectiveness, we conduct a large number of experiments and comparisons on seven public datasets. Experimental results show that our proposed models are useful and achieve state-of-the-art results.

Fast pyrolysis of the mixed plastics (PE: PP: PS: PVC = 0.5: 0.3: 0.15: 0.05) with and without hydrothermal dechlorination pretreatment were investigated in a fixed bed reactor to determine the effects of pyrolysis parameters (temperature, carrier gas flow rate and particle size) on the distribution and the quality of the products. The results showed that high pyrolysis temperature, high carrier gas flow rate and small particle size promoted the migration of organic chlorine in condensable products to inorganic chlorine in gas products. And the optimal conditions for fast pyrolysis of mixed plastics to produce oil were the pyrolysis temperature of 500 °C, carrier gas flow rate of 40 ml/min and particle size of 0.1–0.15 mm with the maximum oil yield of 67.88 wt%. Hydrothermal pretreatment of mixed plastics has shown an excellent dechlorination efficiency of 99.9 %. Furthermore, the total yield of oil and wax was increased by 7.06 wt%, meanwhile the content of C5-C9 components in the pyrolysis oil showed an upward trend when compared with fast pyrolysis without pretreatment. Besides, the concentration of methane achieved 50.68 % due to the possible weakening effect of hydrothermal pretreatment on C C bond energy of β-position.

Microwave (MW) treatment was reported as an efficient method to prepare porous carbon-based electrode material with high capacity for a high-performance supercapacitor. Herein, the superiority of the mechanism of MW treatment was tried to be revealed by preparing a N and O co-doped porous activated carbon (MWAC) based on a heteroatom-rich corn cob and compared to that of the conventional heating method. The resulting MWAC has a large specific surface area of 2508 m²/g with hierarchical micro- and mesoporous structures and abundant heteroatom doping of O and N. In a three-electrode system, the MWAC shows a superior specific capacitance of 560 F/g at a discharge current density of 0.5 A/g in a 6 mol/L KOH aqueous electrolyte. The symmetric supercapacitor fabricated by MWAC electrodes in a 6 mol/L KOH electrolyte shows excellent capacity retention of 96.8%, even after 10 000 galvanostatic charge–discharge cycles at the current density of 1 A/g. The aqueous (6 mol/L KOH) symmetric supercapacitor delivers a high energy density of 9.24 Wh/kg at a power density of 250 W/kg. The results prove that the MW heating shows more superiority in retaining inherent heteroatoms and creating hierarchical micro- and mesoporous structures of activated carbon for supercapacitors with high performance.

Stimuli-responsive hydrogels are flourishing; however, a multiresponsive luminescent hydrogel actuator in which the stimuli do not interfere with each other is still difficult to produce. Here, a smart luminescent bilayer hydrogel actuator with shape memory and reversible luminescence switch behaviors under the co-stimulation of pH and heating is reported. Under alkaline condition, chitosan-containing layer generates physical microcrystals to fix the temporary shape of the actuator while lanthanide emits bright luminescence. Upon acid treatment, the actuator recovers to its original shape and the luminescence is quenched since the "antenna effect" between organic ligand and lanthanide ions is disrupted. It is to be noticed that this multiresponsive cycle can be repeated several times without interfering each other. This work is expected to have potential application prospects in the fields of biomimetic soft robots, sensors, and camouflage.

A comprehensive analysis of the intake valve opening duration (IVOD) effects on the performance of an Atkinson cycle engine is conducted in this work using numerical simulation and experimental validation. Through one-dimensional simulation, the relationship between the range of IVOD and the compression ratios is firstly investigated under the constraint of compression pressure. Two representative IVOD, 295 and 314°CA, are then respectively applied to the performance simulation and experiment of a practical Atkinson cycle engine. The simulation shows the combination of a late intake valve opening timing (IVO) angle and a late exhaust valve opening timing (EVO) angle is profitable for improving the fuel economy under part load operating conditions (i.e. 2000 rpm@2 bar and 3000 rpm@3 bar). The experimental results present the Atkinson cycle engine under both IVOD scenarios considerably improves the brake specific fuel consumption (BSFC) and reduces the pumping mean effective pressure (PMEP) compared to those of the original Otto cycle engine. Meanwhile, the comparison between two IVOD scenarios show that the shorter IVOD leads to an improvement of indicated thermal efficiency, especially at lower load. Considering fuel economy, a shorter IVOD is more favorable at part load for the Atkinson cycle engine. Two main contributions of this work are to numerically quantify the IVOD range for the Atkinson cycle engine under part load, and to experimentally validate the effectiveness of simulation. The findings of this work are expected to support the design of Atkinson cycle engines and provide a guideline of IVOD optimization under part load.

The continuous degradation of carbonate electrolytes and the dissolution of transition metal cations due to parasitic reactions on the cathode-electrolyte interphase (CEI) block the practical application of LiNi0.5Mn1.5O4-based lithium-ion batteries (LNMO-based LIBs) at a high voltage. cis-1,2,3,6-Tetrahydrophthalic anhydride (CTA) has been used as a functional additive in a carbonate baseline electrolyte (BE) for constructing the CEI film to enhance the cyclic stability of LNMO-based LIBs. The LNMO/Li cell with CTA exhibits a preponderant capacity retention of 83.3% compared with those of propionic anhydride (PA) (46.5%) and BE (13.6%) after 500 cycles at the current density of 1 C from 3.5 to 4.9 V. Additionally, the LNMO/graphite full cell with CTA still has a higher capacity retention of 95.46% even after 300 cycles at 1 C. By characterizations, it is reasonably demonstrated that CTA was oxidated to participate in the construction of a CEI film. An unsaturated aromatic group was introduced into the composition of the CEI film along with CTA in the formation process of the CEI film, which further improved the antioxidative activity of the CEI film under the influence of field-effect. Specifically, the CEI film obtains appreciable stability because of its higher antioxidative activity under the influence of field-effect. The stabilized CEI can significantly suppress the parasitic reactions of electrolytes, decrease the consumption of active-Li+, and protect the LNMO cathode structure, thereby enhancing the cyclic compatibility of LNMO-based LIBs with the carbonate electrolytes from 3.5 to 4.9 V.

Novel CaAl2xB2O4+3x (x = 0.25, 0.5, 0.75, and 1) ceramics were prepared via solid-state reaction method. The investigation is concentrated on sintering behavior, phase composition and dielectric properties of CaAl2xB2O4+3x ceramics with various ratios of Al2O3 content. The optimal sintering temperature and the content of CaAl2B2O7 are highly related to the variation of x value. CaAl2xB2O4+3x ceramics possess excellent microwave dielectric properties: er = 5.8 ± 0.05, Q × f = 63,338 ± 2690 GHz (@13.57 GHz), and τf=-29 ± 2 ppm/℃ when sintered at 940 ℃ for x = 0.5. Comparing the properties of CaAl2xB2O4+3x ceramics with reported borate ceramics, the ceramics prepared in this study with low dielectric loss and low sintering temperature have promising prospects in LTCC technology.

A MoSi2 coating was prepared on the Mo1 wire mesh via pack cementation method, and its failure mechanisms under isothermal oxidation and hot-fire test using a hydroxylammonium nitrate based monopropellant were comparatively studied. Under isothermal oxidation at 1300 °C and 1400 °C, degradation of MoSi2 into Mo5Si3 caused failure of the coating, and interdiffusion made a much larger effect relative to oxidation. However, the MoSi2 coating failed because of the synergy of oxidation, ablation, and interdiffusion under hot-fire test. Besides, dissolution of mullite into SiO2 and ablation of high velocity flame contributed to the failure of the coating as well.

Sustainable transportation is one of the solutions to global warming because road transport is responsible for a considerable part of the total carbon emissions. Electric vehicle (EV) technology is still developing, and most of the available brands market them as either environmentally friendly or operationally economical compared to internal combustion engine based traditional vehicles. However, the adoption rates of electric vehicles across the globe are meager. Customer engagement (CEN) is one of the emerging concepts for gaining and retaining users of a product. This research analyzes CEN in EVs based on more engaging features and technologies, i.e., smart connectivity features. Unfortunately, this vital research area remains unexplored in academia. The role of customer experience and customer brand value is also analyzed in relation to smart connectivity features and CEN. The study uses partial least square-based structural equation modeling to analyze the collected data through a web-based survey. The results show that smart connectivity features have a significant positive impact on CEN in EVs. Smart connectivity features also have a significant indirect effect on CEN through customer experience. Further, customer brand value has a direct impact on customer experience, and customer brand value also has a significant indirect effect on CEN in EVs. These results have significant theoretical and managerial implications.

New development on the modeling approaches is generally required to fusing the sensor data to the computational model in order to build up digital twins for the manufacturing processes. In this paper, we propose a modeling approach for building a digital twin for friction stir welding (FSW) based on a sensor-based numerical simulation. In the proposed approach, a novel iterative cyber-physics fusion algorithm is proposed to calculate the in-process 3D temperature field during FSW by fusing the measured temperature data to a ‘moving heat source’ analysis. Two FSW cases of commercial aluminum alloys AA2024 and AA6061 are analyzed to test the proposed modeling approach. It is demonstrated that the real-time calculating of the in-process 3D temperature field during FSW is highly feasible. When a timestep of 2.0 s is adopted, the proposed approach shows good reliability, as the average error between the measured and the calculated temperatures is less than 4 °C. More importantly, it is found that the calculation time for the whole welding process is shown to be less than the physical welding time. This allows that the sensor-based numerical simulation and the physical FSW process can be synchronized with a second-level temporal precision, which provides a viable cyber-physics fusion approach for digital twin modeling for the FSW process.

In this paper, we report the finding and basic characterization of a new multiferroic material in the form of the Y2FeAlO6 (YFAO) double perovskite with orthorhombic structure. In order to study the structural and multiferroic properties in detail, we prepared the polycrystalline YFeO3 (YFO) and Y2FeAlO6 ceramic samples through traditional high temperature solid-state reaction method. XRD analysis confirmed that these two samples are single-phase and all the peaks in the pattern can be indexed with orthorhombic symmetry of the Pnma space group. Compared to YFO, YFAO is more symmetric, and has a more stable crystalline structure and smaller grain size. The Fe ion in YFO and YFAO are in the mixed oxidation state of Fe2+ and Fe3+, while the relative amount of Fe2+ in YFAO is higher than that of YFO. Al3+ with filled 2p orbital and grain refinement may be the main reasons for the greatly enhanced multiferroicity of YFAO. The magnetic properties of YFO and YFAO are mainly dominant antiferromagnetism and weak ferromagnetism, and the exchange bias effect disappears in YFAO. The coexistence of ferroelectricity and ferromagnetic orders at room temperature in YFAO promises new opportunities and improvements in next generation applications for spintronic, information storage, and sensors etc.

A facile synthetic route was developed to prepare a surface-grafted brush layer of poly(vinyl ethers) (PVEs) directly by a radical mechanism, with the "naked" Li+ acting as a catalyst. Density functional theory calculations suggested that complexation of naked Li+ to VEs significantly reduced the highest unoccupied molecular orbital-lowest unoccupied molecular orbital (HOMO-LUMO) energy gap from 5.08 to 0.68 eV, providing a better prospect for electron transfer. The structure, morphology, and surface properties of grafted polymer layers were characterized using attenuated total reflection Fourier transform infrared spectroscopy, Raman spectroscopy, X-ray photoelectron spectroscopy, atomic force microscopy, and dynamic water contact angle (DCA). Moreover, ellipsometry data indicated that the thickness of the polymer brushes was in the range of 20-60 nm, which corresponds to the grafting densities of 0.65-1.15 chain/nm2, and DCA decreased from 84.4 to 45.3°. Most importantly, no hydrolysis was observed for the modified surface after 30 days of exposure to phosphate-buffered saline solution, 0.1 mol/L NaOH(eq) and 0.1 mol/L HCl(eq), demonstrating excellent hydrolysis resistance with long service life. In addition, as a proof of concept, the side hydroxyl groups of grafted PVEs provide active sites for efficient fixation of bioactive molecules, e.g., glycosaminoglycan and serum protein.

In this paper, the typical fault estimation and dynamic analysis are presented for a leader-follower unmanned aerial vehicle (UAV) formation system with external disturbances. Firstly, a dynamic model with proportional navigation guidance (PNG) control of the UAV formation is built. Then, an intermediate observer design method is adopted to estimate the system states and faults simultaneously. Based on the graph theory, the topology relationship between each node in the UAV formation has been also analyzed. The estimator and the system error have been created. Moreover, the typical faults, including the components failure, airframe damage, communication failure, formation collision, and environmental impact, are also discussed for the UAV system. Based on the fault-tolerant strategy, five familiar fault models are proposed from the perspectives of fault estimation, dynamical disturbances, and formation cooperative control. With an analysis of the results of states and faults estimation, the actuator faults can be estimated precisely with component failure and wind disturbances. Furthermore, the basic dynamic characteristics of the UAV formation are discussed. Besides, a comparison of two cases related to the wind disturbance has been accomplished to verify the performance of the fault estimator and controller. The results illustrate the credibility and applicability of the fault estimation and dynamic control strategies for the UAV system which are proposed in this paper. Finally, an extension about the UAV formation prognostic health management system is expounded from the point of view of the fault-tolerant control, dynamic modeling, and multifault estimation.

Titanium alloy Ti-6Al-4V particles were used as reinforcement for AZ31B magnesium metal matrix composites (Mg MMCs) which were synthesized via friction stir processing (FSP). Ti-6Al-4V particulates were packed in the machined grooves at first and subsequently processed by means of a traditional machine used for vertical milling. The processing speed and the repeated passes were considered as the major processing factors to vary during processing apart from the content of Ti-6Al-4V particles. Optical and scanning electron microscopes were used to record the evolved microstructure of the composites. Changes in Ti-6Al-4V particle distribution across the stir zone resulted to form several kinds of microscopic regions. The distribution of particles became inhomogeneous as the traverse speed was advanced to higher level. Conversely, the distribution became homogeneous with every pass. The role of selected process variables on the tensile behavior was analyzed. Mg MMCs that were prepared at lower traverse speed (30 mm/min) and higher number of passes (5) showed higher properties during tensile test. The micrographs of the fracture surfaces were studied to understand the mode of fracture.

Microstructure evolution of columnar dendrites during isothermal coarsening is simulated by a three-dimensional (3-D) cellular automaton (CA) model. The 3-D CA model incorporates the thermodynamics, and kinetics aspects influenced by the factors of temperature, composition, and interfacial weighted mean curvature. After quantitative validation through grid independence test and the liquid pool migration problem, the 3-D CA model is applied to simulate the dendrite coarsening phenomena caused by the simultaneous melting and solidification for a SCN-2.0 wt.% ACE alloy in an isothermal environment. Several typical features of dendrite coarsening occurring in the 3-D space are reproduced by the CA model, including melting of tertiary and small secondary branches, interdendritic groove base advancement, and coalescence of neighboring dendrite arms. The simulated dendritic microstructure compares well with the in situ experimental observation. Besides, the simulated specific surface area of the solid/liquid interface decreases with time following a power-law relation with an exponent of 1/3. The simulated distribution curve of the interfacial weighted mean curvatures becomes narrower with time, and the peak gradually shifts to zero, showing that the S/L interface is flattened during the isothermal dendrite coarsening. The average thickness of the primary dendrite arm and the secondary dendrite arm spacing (λ2) are found to increase with time, and the third power of λ2 shows a nearly linear relationship with time. The CA model not only visualizes morphological evolution of the dendrite arms, but also provides insight into the complex interaction among local melting/solidification, solute diffusion, and variation of weighted mean curvature in the 3-D space.

We report on the fabrication of circular cladding waveguides with cross-section diameters of 60−120 μm in Pr:YAG crystal by applying femtosecond laser inscription. The fabricated waveguides present 2D guidance on the cross-section and fairly low propagation losses. Multiple high-order guiding modes are observed in waveguides with different diameters. Corresponding simulation results reveal the origin of a specific kind of guiding modes. Confocal micro-Raman (μ-Raman) experiments demonstrate the modification effects in femtosecond laser affected areas and ascertain the refractive index induced guiding mechanism. In addition, luminescence emission properties of Pr3+ ions at waveguide volume are well preserved during the femtosecond laser inscription process, which may result in a potential high-power visible waveguide laser.

Electric vehicle (EV) has become one of the most critical components in the smart grid with the applications of the Internet-of-Things (IoT) technologies. Real-time charging control is pivotal to ensure the efficient operation of EVs. However, the charging control performance is limited by the uncertainty of the environment. On the other hand, it is challenging to determine a charging control strategy that is able to optimize multiple objectives simultaneously. In this article, we formulate the EV charging control model as a Markov decision process (MDP) by constructing state, action, transition function, and reward. Then, we propose a deep-reinforcement-learning-based approach: charging control deep deterministic policy gradient (CDDPG) to learn the optimal charging control strategy for satisfying the user’s requirement of battery energy while minimizing the user’s charging expense. We utilize the long short-term memory (LSTM) network that extracts the information of previous energy price to determine the current charging control strategy. Moreover, Gaussian noise is added to the output of the actor network to prevent the agent from sticking into the nonoptimal strategy. In addition, we address the limitation of sparse rewards by using two replay buffers, of which one is used to store the rewards during the charging phase and another is used to store the rewards after charging is completed. The simulation results prove that the CDDPG-based approach outperforms the deep- $Q$ -learning-based approach (DQL) and the deep-deterministic-policy-gradient-based approach (DDPG) in satisfying the user’s requirement for the battery energy and reducing the charging cost.

The reliability of products is a critical issue as it has high economic impacts, especially in current competitive markets. In modern applications, the complex and high-dimensional data of products are collected that can be used for reliability analysis and failure prediction. The existing reliability approaches, however, cannot efficiently model complex covariates and their effects on the time-to-failure of products. In this article, we propose a novel deep learning-based reliability approach to model the complex relationship between the covariates and product failure. To estimate model parameters, neither the traditional deep learning parameter estimation method nor the maximum likelihood estimation method is applicable. To overcome this difficulty, a new model parameter estimation method is developed based on the partial likelihood framework. Furthermore, as there are often only a limited number of samples for real-world reliability problems, a new penalized partial likelihood estimation method is developed for this special circumstance. The developed method is capable of estimating model parameters for censored reliability data. A simulation study is conducted to verify the developed methods. The proposed method is justified by a real-world case study of the reliability analysis of materials. The case study shows that the proposed model outperforms the existing ones.

Key nodes identification is an important way to analyze and understand the characteristics, structure, and functions of the complex network. In this paper, links in complex networks are taken as the basic unit, and the overlapping community in complex networks is obtained through the clustering analysis of links. Then, the importance of the node is judged according to the number of associations containing the node and the weight value of the association in the network, because the nodes Shared between communities have more influence on the functional structure of the network. Finally, the method is applied to rank the importance of nodes in IEEE standard 300 node system, and the results are verified by network conductivity. The comparison with the results of the Betweenness algorithm, HITS algorithm and Pagepank algorithm shows that the method presented in this paper can effectively identify the key nodes from the complex network.

The classification tasks for numerical or categorical data have been well developed. However, the data collected in the real world are frequently the mixed type containing numerical and categorical values, and how to classify the mixed data quickly and efficiently is a critical yet challenging task. Existing classification models for mixed data usually treat the mixed data processing and subsequent classification as two independent phases, without considering their compatibility. By fusing the mixed data processing into a classification algorithm, this paper proposes an extended version of RBF-ELM (Radial Basis Function-Extreme Learning Machine), a Mixed Data RBF-ELM method (MD-RBF-ELM for short), which can achieve direct, fast, and efficient classification for mixed data. Specifically, a distance metric method for mixed data is firstly designed to calculate the distances between the input data and the RBF centers, and then these distances are used to train the network structure and weights of MD-RBF-ELM, thereby realizing the fusion of data processing with model learning. In addition, to alleviate the problem of MD-RBF-ELM’s unstable performance caused by randomly selecting the RBF centers, we propose an improved density peak clustering algorithm and use it to select the optimal RBF centers automatically and adaptively. Extensive experimental results on 34 data sets demonstrate that MD-RBF-ELM significantly enhances the classification performance (increasing 2.37% for F1-score, up to 14/34 for the number of best results, and reaching 2.4/8 for the averaged ranks), compared with seven state-of-the-art competitors.

This study presents the formation mechanism of a layered structure consisting of coherent cellular and columnar solidification microstructure accompanied with the macrosegregation in aluminum dissimilar-filler welding. The filler-depleted zone resulted from the incomplete mixing of the solute; moreover, it solidifies before the surrounding mixed zone, wherein the subsequent dendritic growth is hindered by the pre-solidified microstructure. This solidification sequence leads to a layered solidification microstructure and segregation patterns.

Millions of patients worldwide are implanted with permanent pacemakers for the treatment of cardiac arrhythmias and conduction disorders. The increased use of these devices has established a growing clinical need to mitigate associated complications. Pacemaker leads, in particular, present the primary risks in most implants. While wireless power transfer holds great promise in eliminating implantable device leads, anatomical constraints limit efficient wireless transmission over the necessary operational range. We thereby developed a transmitter-centered control system for wireless power transfer with sufficient power for continuous cardiac pacing. Device safety was validated using a computational model of the system within an MRI-based anatomical model. The pacer was then fabricated to meet the acute constraints of the anterior cardiac vein (ACV) to enable intravascular deployment while maintaining power efficiency. Our computational model revealed the wireless system to operate at > 50 times below the tissue energy absorption safety criteria. We further demonstrated the capacity for ex vivo pacing of pig hearts at 60 beats per minute (BPM) and in vivo pacing at 120 BPM following pacer deployment in the ACV. This work thus established the capacity for wireless intravascular pacing with the potential to eliminate complications associated with current lead-based deep tissue implants.

Three new emitters, namely 10,10'-(quinoline-2,8-diyl)bis(10H-phenoxazine) (Fene), 10,10'-(quinoline-2,8-diyl)bis(10H-phenothiazine) (Fens) and 10,10'-(quinoline-2,8-diyl)bis(9,9-dimethyl-9,10-dihydroacridine) (Yad), featuring quinoline as a new electron acceptor have been designed and conveniently synthesized. These emitters possessed small singlet–triplet splitting energy (ΔEst) and twisted structures, which not only endowed them show thermally activated delayed fluorescence (TADF) properties but also afforded a remarkable aggregation-induced emission (AIE) feature. Moreover, they also showed aggregation-induced delayed fluorescence (AIDF) property and good photoluminescence (PL) property, which are the ideal emitters for non-doped organic light-emitting diodes (OLEDs). Furthermore, high-performance non-doped OLEDs based on Fene, Fens and Yad were achieved, and excellent maximum external quantum efficiencies (EQEmax) of 14.9%, 13.1% and 17.4%, respectively, were obtained. It was also found that all devices exhibited relatively low turn-on voltages ranging from 3.0 V to 3.2 V probably due to their twisted conformation and the AIDF properties. These results demonstrated the quinoline-based emitters could have a promising application in non-doped OLEDs.

Failures due to high-cycle fatigue have led to a high cost in aerospace engineering over the past few decades. In this paper, the experimental results of the fatigue behavior of compressor blade specimen subjected to resonance and the effects of a damping hard coating on relieving the fatigue progress are presented. The crack initiation and propagation processes were observed under resonance of the first bending mode by using the resonant frequencies as the indicator. Significant nonlinear features were observed in the spectrum of the blade with a fatigue crack. The finite element model considering the breathing crack was established with nonlinear contact based on the crack localization and size, which was obtained by ultrasonic phased array technology. The simulation results of the vibration behavior of the cracked blade were obtained and consistent with the experimental results. A NiCrAlY coating was deposited on the blade, and increases in the fatigue life were observed under the same condition. The results of this paper can help to better understand the fatigue of a compressor blade subjected to resonance and provide a preference for the application of a damping hard coating on compressor blades.

Chip-scale molecular clocks (CSMCs) perform frequency stabilization by referencing to the rotational spectra of polar gaseous molecules. With, potentially, the “atomic” clock grade stability, cm3-level volume, and $4.3\times 10^{-11}$ under an averaging time of $\tau =10^{3}$ s while consuming 70.4-mW dc power.

Carbon nanotubes (CNTs) are regarded as the next generation of one-dimensional super-strong nanomaterials due to their extraordinary mechanical properties. Recently, much attention has been dedicated into super-strong CNT-based fibers resulting from twisting operation. However, the roles of a promising compression operation on the mechanical enhancement mechanisms of CNT fibers (CNFs) are not clear. Especially, the combined twisting-compression effect has not been known. In this work, with using coarse-grained molecular dynamic simulation method, the tensile mechanical properties and mechanical enhancements of randomly-distributed CNT fibers (RDCNFs) were studied. It was found that the mechanical performances of such RDCNFs could be significantly improved by increasing CNT length or decreasing initial bending degree of CNTs. Furthermore, both specific strength and specific modulus can be effectively tuned by twisting and compressing operations, where self-locking and unhitching mechanisms of CNT knots play critical roles. On the other hand, following theoretical guidance, a super-strong CNF with the maximum specific strength up to 296 mN/tex (∼6.8 GPa) was designed and fabricated. This work provides a theoretically and experimentally support to design super-strong CNT-based fibers and lays a foundation for their future applications.

Enhancing the mechanical strength of highly conductive pure metals usually causes significant reduction in their electrical conductivity. For example, introducing phase/matrix interfaces or more grain boundaries, are common and effective methods to strengthen metals. But it simultaneously increases the electron scattering at the interface, thus reducing the electrical conductivity. In this study, we demonstrate that pure aluminum (Al)/carbon nanotubes (CNTs) nanocomposites prepared by friction stir processing have successfully broken through these limitations. The yield strength and tensile strength of Al/CNTs nanocomposites have improved by 104.7% and 51.8% compared to pure Al, while the electrical conductivity remained comparable to that of pure Al. To explore the potential mechanisms, the interface between CNTs and Al was examined and characterized by transmission electron microscopy (TEM) and Raman spectroscopy. Little interfacial reaction compounds were present and no visible physical gaps were observed at CNTs and Al interfaces. We defined it as a clean and tightly bonded interface. Although the quantity of phase interface has increased, the electrical conductivity of the nanocomposite remains approximately unchanged. We attribute the preserved electrical conductivity to the clean and tightly bonded CNTs/Al interface in the nanocomposite.

Developing low-cost and highly active hydrolysis catalytic materials for the dehydrogenation of hydrogen-rich chemicals is a promising strategy to store and easily release hydrogen for fuel cell applications. In this study, new bimetallic hybrid nanoplates containing Co3O4 and CuCoO2 were developed via a facile and low-cost approach for catalytically producing hydrogen from ammonia borane (NH3BH3). The as-prepared efficient Co3O4–CuCoO2 material was employed as a catalyst for NH3BH3 hydrolysis, and it exhibited a high catalytic performance with a turnover frequency (TOF) of 65.0 molhydrogen molcat.−1 min−1 at room temperature. The results revealed that the hybrid Co3O4–CuCoO2 catalysts exhibited much improved catalytic activity towards the hydrolysis reaction in contrast to the individual Co3O4 and CuCoO2 catalysts. A kinetic study showed that the Co3O4–CuCoO2 catalytic material had superior activity with an apparent activation energy (Ea) of 20.5 kJ mol−1, which was found to be one of the lowest Ea for NH3BH3 hydrolysis ever reported in the literature. This work provides a promising approach for the synthesis of noble metal-free catalysts exhibiting high catalytic activity towards NH3BH3 hydrolysis.

Excessively high brake temperature of hydrodynamic retarders may lead to brake fading and failure, resulting in a decrease in brake effectiveness. However, the temperature performance modeling of hydrodynamic retarders is a challenge because of the non-linear characteristics of the system. In this study, a temperature model based on an artificial neural network is constructed to predict the temperature performance of a hydrodynamic retarder in constant-torque braking process. The model is developed from a back-propagation neural network trained with the Levenberg-Marquardt algorithm. Before the application of the neural network, computational fluid dynamics is used to obtain the controllable region where experimental tests were performed to collect data for neural network training and validation. The linear regression method is adopted to check the quality of the training. It is shown that the constructed back-propagation neural network model is within 98% accuracy. Furthermore, the temperature model of the hydrodynamic retarder, which consists of the back-propagation neural network and thermal balance models, is simulated for 1500N $\cdot \text{m}$ , 2000N $\cdot \text{m}$ , and 2500N $\cdot \text{m}$ constant-torque braking processes. The simulation results are in agreement with experimental data within 2.87% error. The proposed temperature model can predict the temperature of the hydrodynamic retarder accurately and provide theoretical guidance for brake control strategy and thermal management.

Long service time and excessive vibration of aero-engine will cause failure of clamps, which usually fix pipelines on the outside of an aero-engine. In this paper, the nonlinear vibration characteristics caused by loosening clamps in aero-engine pipeline are determined. According to the real parameters of an aero-engine clamp-pipeline test rig, the clamp is simplified into a nonlinear spring-damping. Thus, a clamp-pipeline nonlinear model based on cubic stiffness is established. The multi-scale method is used to obtain the analytical solution of the nonlinear model. Based on the real test data, the model parameters of the clamp-pipeline nonlinear system are identified, and then the nonlinear vibration characteristics of the system are obtained. Finally, using the vibration transmissibility method, the influence of the severity of clamp loosening on the nonlinear vibration characteristics of the system is obtained.

Bilateral filter has demonstrated its effectiveness in many traditional methods for image restoration tasks. In this letter, we incorporate the idea of bilateral grid processing in a CNN framework and propose a bilateral stereo super-resolution network (BSSRnet). Specifically, we use a parallax-attention module to incorporate information from left and right views to learn content-aware bilateral filters. Then, these bilateral filters are used to recover missing details at different spatial locations while preserving stereo consistency. Our network is fully differentiable and is robust to both content and disparity variations. Comparative results show that our BSSRnet achieves state-of-the-art performance on the Flickr1024, Middlebury, KITTI 2012 and KITTI 2015 datasets. Source code is available at.