Search

Graphyne has attracted considerable interest and attention since its successful synthesis, due to its enormous potential for applications in the fields of electronics, energy, catalysis, information technology, etc. Although various methods for synthesizing graphyne have been explored, single-layer graphynes have not been successfully developed. Hexaethynylbenzene (HEB) is considered an ideal precursor molecule because it can undergo Glaser coupling reactions between molecules to synthesize single layer graphdiyne on single crystal metal surfaces via on-surface reactions. Unfortunately, this method fails to achieve the expected results, and the underlying mechanism is not clear. In this work, we employed a combination of ab initiomolecular dynamics (AIMD) and quantum mechanics (QM) methods to investigate the initial reaction mechanism of HEB molecules on a Au(111) surface. We revealed that HEB molecules undergo both intermolecular coupling and intramolecular cyclization on the Au(111) surface. The favorable pathways of these two types of reactions were then distinguished, confirming that the distance between the terminal carbon atoms of the ethynyl groups plays an important role in C–C coupling. The insights revealed from this work could facilitate the rational design of precursor molecules and deepen the understanding of the reaction processes.

Transmetalation represents an appealing strategy toward fabricating and tuning functional metal–organic polymers and frameworks for diverse applications. In particular, building two-dimensional metal–organic and organometallic networks affords versatile nanoarchitectures of potential interest for nanodevices and quantum technology. The controlled replacement of embedded metal centers holds promise for exploring versatile material varieties by serial modification and different functionalization. Herein, we introduce a protocol for the modification of a single-layer carbon–metal-based organometallic network via transmetalation. By integrating external Cu atoms into the alkynyl–Ag organometallic network constructed with 1,3,5-triethynylbenzene precursors, we successfully realized in situ its highly regular alkynyl–Cu counterpart on the Ag(111) surface. While maintaining a similar lattice periodicity and pore morphology to the original alkynyl–Ag sheet, the Cu-based network exhibits increased thermal stability, guaranteeing improved robustness for practical implementation.

To elucidate the dynamic mechanisms of unbalanced impellers in ultra-high head pump-turbines (PTs), this study employed a one- and three-dimensional coupled method to simulate the pump power-trip (PPT) process of an ultra-high head PT. The investigation revealed two novel pulsation frequency components, denoted asfDVandfINFT, associated with impeller forces. The pulsation intensities of these components were markedly higher than those of rotor-stator interaction frequency components in ultra-high head PTs. Notably, thefDVcomponents exhibited pulsations at 1–2 times the rated rotation frequency of the impeller, spanning the entire transition period. Meanwhile, thefINFTcomponents constituted a complex frequency band with various frequency values, primarily occurring near conditions (Q= 0,n= 0,M= 0, and dM/dt= 0). These two pulsation frequency components were predominantly linked to the unsteady evolution of dean vortices inside the volute and complex transitions of the flow pattern within the impeller, respectively. It is crucial to note that these unbalanced flow-induced impeller axial forces can elevate the risk of accidents where the rotor is subjected to significant upwind axial forces. These findings offer valuable insights into mitigating the risk of rotor lifting due to axial forces during PT events in ultra-high head PTs.

Micro-arc oxidation coatings were prepared on the surface of TC4 by adding different contents of rare-earth compound Pr(NO3)3to the aluminate and phosphate-based electrolyte. The results showed that the change in oxidation voltage increased at first and then decreased with the raised of Pr(NO3)3addition. At 0.05 g L−1, the oxidation voltage reached the maximum, the number of discharge micropores decreased, while the size increased, the thickness and hardness of the coatings reached the peak values. The coatings mainly included Al2TiO5, anatase, γ-Al2O3and a little of Pr2O3. Pr(NO3)3-doped coatings showed excellent corrosion resistance in 3.5% NaCl electrochemical workstation and oilfield simulation fluid containing 5% NaCl, and the corrosion resistance of the coatings reached the best at 0.05 g L−1.

The influence of various Fe contents (5, 10, 20, and 40 wt.%) on the microstructure and properties of powder metallurgy Cu–Fe alloy by spark plasma sintering, subjected to cold rolling and aging treatments, was explored. The results showed that, with increasing Fe content from 5 wt.% to 40 wt.%, the distribution of Fe phase evolved from the discrete spheroid to the continuous stagger, accompanying with its size increasing from 0.29 to 1.20 μm. Accordingly, the yield strength of the aged alloy increased from 411.5 to 788.8 MPa, whilst the electrical conductivity decreased from 62.5% (IACS) to 42.0% (IACS). Moreover, a rule of mixture was proposed to calculate the contribution of Cu matrix, primary Fe phase and secondary Fe phase to the yield strength of the alloy, which works well in the prediction of mechanical properties for Cu–Fe alloys with Fe content higher than10 wt.%.

Recognition and control of the laser welding states of aluminum profiles in China Railway High-speed trains is a challenge to current monitoring system, because the bottom-locking joints possess more complex welding states (Lack, Lock, Good, Fulland Overpenetration) than common butt joints. This work developed an online monitoring system based on a 3D-convolutional neural network (3DCNN), which utilized temporal information of both laser plume and weld pool images. The results indicated that multi-information fusion was necessary because the ability of single-information method to distinguish five welding states were insufficient. On this basis, there kinds of 3DCNN using different information-fusion methods were compared. The separated-channel 3DCNN with feature-level fusion was found to be more efficient than the original 3DCNN with data-level fusion. It realized the same level accuracy while decreasing the network params by 49.5 %. Additionally, the importance of different information was considered by trainable weight indexes and weight-fusion 3DCNN was proposed. The weight of weld pool information gradually decreased while the weight of laser plume increased. The weight-fusion 3DCNN was superior to the other two networks and the highest accuracy of 98.9 % was obtained. The processing speed of the whole online monitoring system with feedback control reached 10 Hz, which could meet the requirements in industrial applications. This work will provide more guidance on the application of multi-information fusion and deep learning methods in online monitoring system.

The blood–brain barrier (BBB) and drug resistance present challenges for chemotherapy of glioblastoma (GBM). A microneedle (MN) patch with excellent biocompatibility and biodegradability was designed to bypass the BBB and release temozolomide (TMZ) and PLCG1-siRNA directly into the tumor site for synergistic treatment of GBM. The codelivery of TMZ and PLCG1-siRNA enhanced DNA damage and apoptosis. The potential mechanism behind this enhancement is to knockdown of PLCG1 expression, which positively regulates the expression of signal transducer and activator of transcription 3 genes, thereby preventing DNA repair and enhancing the sensitivity of GBM to TMZ. The MN patch enables long-term sustainable drug release through in situimplantation and increases local drug concentrations in diseased areas, significantly extending mouse survival time compared to other drug treatment groups. MN drug delivery provides a platform for the combination treatment of GBM and other central nervous system diseases.

Flexible moisture–electric generators (MEGs) capture chemical energy from atmospheric moisture for sustainable electricity, gaining attention in wearable electronics. However, challenges persist in the large-scale integration and miniaturization of MEGs for long-term, high-power output. Herein, a vertical heterogeneous phase-engineering MoS2nanosheet structure based silk and cotton were rationally designed and successfully applied to construct wearable MEGs for moisture–energy conversion. The prepared METs exhibit ∼0.8 V open-circuit voltage, ∼0.27 mA/cm2current density for >10 h, and >36.12 μW/cm2peak output power density, 3 orders higher than current standards. And the large-scale device realizes a current output of 0.145 A. An internal phase gradient between the 2H semiconductor MoS2in carbonized silks and 1T metallic MoS2in cotton fibers enables a phase-engineering-based heterogeneous electric double layer functioning as an equivalent parallel circuit, leading to enhanced high-power output. Owing to their facile customization for seamless adaptation to the human body, we envision exciting possibilities for these wearable METs as integrated self-power sources, enabling real-time monitoring of physiological parameters in wearable electronics.

Humanoid robots often struggle to express the intricate and authentic facial expressions characteristic of humans, potentially hampering user engagement. To address this challenge, we introduce a comprehensive two-stage methodology to empower our autonomous affective robot with the capacity to exhibit rich and natural facial expressions. In the initial stage, we present an innovative action unit (AU) driven facial expression disentangled synthesis method, enabling the generation of nuanced robot facial expression images guided by AUs. By harnessing facial AUs within a framework of weakly supervised learning, we effectively surmount the scarcity of paired training data (comprising source and target facial expression images). To preserve the integrity of AUs while mitigating identity interference, we leverage a latent facial attribute space to disentangle expression-related and expression-unrelated cues, employing solely the former for expression synthesis. In the subsequent phase, we actualize an affective robot endowed with multifaceted degrees of freedom for facial movements, facilitating the embodiment of the synthesized fine-grained facial expressions. We devise a specialized motor command mapping network that serves as a conduit between the generated expression images and the robot's realistic facial responses. By utilizing the physical motor positions as constraints, we refine the prediction of precise motor commands from the robot's generated facial expressions. This refinement process ensures that the robot's facial movements authentically express accurate and natural expressions. Finally, qualitative and quantitative evaluations on the benchmarking Emotionet dataset verify the effectiveness of the proposed generation method. Results on the self-developed affective robot indicate that our method achieves a promising generation of specific facial expressions with given AUs, significantly enhancing the affective human–robot interaction.

Optical monitoring of the position and alignment of objects with a precision of only a few nanometres is key in applications such as smart manufacturing and force sensing. Traditional optical nanometrology requires precise nanostructure fabrication, multibeam interference or complex postprocessing algorithms, sometimes hampering wider adoption of this technology. Here we show a simplified, yet robust, approach to achieve nanometric metrology down to 2 nm resolution that eliminates the need for any reference signal for interferometric measurements. We insert an erbium-doped quartz crystal absorber into a single Fabry–Pérot cavity with a length of 3 cm and then induce exceptional points by matching the optical loss with the intercavity coupling. We experimentally achieve a displacement response enhancement of 86 times compared with lossless methods, and theoretically argue that an enhancement of over 450 times, corresponding to subnanometre resolution, may be achievable. We also show a fivefold enhancement in the signal-to-noise ratio, thus demonstrating that non-Hermitian sensors can lead to improved performances over the Hermitian counterpart.

The increase in car ownership has improved the convenience of people’s travel, but it has also increased the potential risk of pedestrian-vehicle conflicts. In complex traffic scenarios, pedestrian crossing behavior may lead to frequent traffic accidents. It is crucial to accurately and timely anticipate pedestrian crossing intentions to avoid pedestrian-vehicle conflicts, improve driving safety and ensure pedestrian safety. We propose a novel dual-channel pedestrian crossing intention anticipation network (DPCIAN) to anticipate pedestrian crossing intentions. DPCIAN comprises the pedestrian action encoding module and the scene traffic object interaction relation module (STOIRM). Initially, the pedestrian action encoding module resorts to channel-refined Graph Convolutional networks (GCNs) to extract spatio-temporal action features from pedestrians’ skeletons, which improves the inflexible feature extraction of the adaptive GCN and the information redundancy of the multi-channel GCN. Afterward, the STOIRM introduces an interaction relation network to excavate scene object interaction features related to the target pedestrian’s crossing intention with better scene understanding. Finally, the adaptive average pooling layer is used to fuse the spatio-temporal and interaction features to achieve robust anticipation of pedestrian crossing intention. DPCIAN is evaluated on two public datasets, JAAD and PIE, with an accuracy of 89% and 91%, respectively. Both qualitative and quantitative evaluations indicate that DPCIAN can precisely anticipate pedestrians’ crossing intentions in complex traffic scenarios. Code is available at: https://github.com/wrysmile99/DPCIAN.

The high-frequency pulse tube cryocooler (HPTC) represents a promising miniature cryocooling technology due to its compact structure and the absence of low-temperature moving components. However, limited to the non-ideal gas effect of 4He, the HPTC is hard to obtain high cooling performance in the liquid helium temperature range. 3He as the working fluid can effectively improve the cooling performance of the HPTC, but the high cost hinders its wide application. In consideration of both cooling performance and cost-effectiveness, this paper explores the feasibility of utilizing 3He-4He mixtures as the working fluid for HPTCs. Firstly, the experimental results of a developed HPTC based 4He are reported. With a total power consumption of 575 W, the lowest temperature of 3.26 K was observed. And the measured cooling power at 4.2 K was 20.8 mW. Then the theoretical utmost efficiency of the cryocooler was calculated in terms of the thermophysical properties of the working fluids, using 3He-4He mixtures with different compositions as the working fluids. The whole machine modeling of the HPTC was further carried out, and the influence of the working fluids with different components on the structural parameters such as double-inlet and inertance tube, and operating parameters such as pressure and frequency were analyzed. The calculated results show that the cooling power is expected to be increased to 36 mW and 53 mW if the equimolar 3He-4He mixture and pure 3He are used, respectively.

Forecasting pedestrian trajectories in dynamic scenes remains a critical problem in various applications, such as autonomous driving and socially aware robots. Such forecasting is challenging due to human-human and human-object interactions and future uncertainties caused by human randomness. Generative model-based methods handle future uncertainties by sampling a latent variable. However, few studies explored the generation of the latent variable. In this work, we propose the trajectory predictor with pseudo Oracle (TPPO), which is a generative model-based trajectory predictor. The first pseudo oracle is pedestrians’ moving directions, and the second one is the latent variable estimated from ground truth trajectories. A social attention module is used to aggregate neighbors’ interactions based on the correlation between pedestrians’ moving directions and future trajectories. This correlation is inspired by the fact that pedestrians’ future trajectories are often influenced by pedestrians in front. A latent variable predictor is proposed to estimate latent variable distributions from observed and ground-truth trajectories. Moreover, the gap between these two distributions is minimized during training. Therefore, the latent variable predictor can estimate the latent variable from observed trajectories to approximate that estimated from ground-truth trajectories. We compare the performance of TPPO with related methods on several public datasets. Results demonstrate that TPPO outperforms state-of-the-art methods with low average and final displacement errors. The ablation study shows that the prediction performance will not dramatically decrease as sampling times decline during tests.

It is critical for autonomous vehicles to accurately forecast the future trajectories of surrounding agents to avoid collisions. However, capturing the complex interactions between agents in complex urban scenes is challenging. As a result, complex interactions may impair trajectory prediction accuracy. A trajectory prediction network with an enhanced Graph Transformer (TP-EGT) is proposed to forecast the future trajectories of traffic-agents. A collision-aware Graph Transformer is introduced to capture the complex social interactions between traffic-agents. Following that, an additional interaction prediction task that could predict the interaction probabilities between agents is proposed to mitigate the over-smoothing issue of the Graph Transformer via a multi-task learning strategy. Afterward, the trajectory prediction performance is improved with additional interaction probabilities, which are beneficial for the decision-making and planning modules of autonomous vehicles. Quantitative and qualitative evaluations of TP-EGT on the ETH/UCY and ApolloScape databases demonstrate that the trajectory prediction accuracy of TP-EGT is comparable to the state-of-the-art baseline methods, and the predicted interaction probabilities can help autonomous vehicles comprehend the complex traffic scenes.

Batch processes often encounter insufficiency of modeling data and differences of data distributions between different tanks and batches, which pose great challenges for accurate data-driven soft sensor modeling. Transfer learning has become a promising solution for addressing such problems; however, the existing transfer learning soft sensors still have problems, such as not considering the quality of source domain samples, insufficient utilization of hidden feature information, and only considering single distribution alignment. To address these issues, a dynamic weighted multilayer domain adaptation network (DWMDAN) is proposed for online estimation of key variables in batch processes. The training of the model consists of two stages: Stage I, the feature extractor and the first domain discriminator are utilized for adversarial training to compute the importance weights of the source domain samples, and the whole model is updated by combining weighted regression with global domain adaptation. Stage II, a hybrid domain adaptation mechanism is further designed to fine-tune the target model based on importance-weighted information. Specifically, adversarial and distribution domain adaptation methods are combined to adapt the marginal and conditional distributions between domains at multiple network layers of the model, where an adaptive factor is introduced to dynamically adjust the distribution weights. The superiority and effectiveness of the proposed method are demonstrated by an industrial fed-batch chlortetracycline (CTC) fermentation process. The experimental results show that the proposed method can extract richer and more reliable domain-invariant features for tank-to-tank knowledge transfer, thus delivering more accurate quality predictions.

The present study investigated the inhomogeneity of microstructure and properties of hybrid laser-MIG welded 10-mm-thick 316L/AH36 joints. The study focused on the inhomogeneity of the laser zone, hybrid zone, and transition zone. The temperature field was simulated using ABAQUS and showed a higher cooling rate in the laser zone. As a result, the microstructure in the laser zone was dominated by martensite while that in the hybrid zone was composed of 84 % austenite and 16 % ferrite. The laser zone showed poorer impact toughness and corrosion resistance than the hybrid zone. However, the corrosion resistance of the whole weld was the worst. Because the inhomogeneous microstructure formed a galvanic couple that accelerated the corrosion process. This work indicated that the energy ratio (the ratio of laser to arc energy) should be considered in the welding of dissimilar thick steel to reduce the inhomogeneity in the two zones and obtain a desirable property balance among hardness, toughness, and corrosion resistance.

The oral hexavalent live human-bovine reassortant rotavirus vaccine (RV6) developed by Wuhan Institute of Biological Products Co., Ltd (WIBP) has finished a randomized, placebo-controlled phase III clinical trial in four provinces of China in 2021. The trail demonstrated that RV6 has a high vaccine efficacy against the prevalent strains and is safe for use in infants. During the phase III clinical trial (2019–2021), 200 rotavirus-positive fecal samples from children with RV gastroenteritis (RVGE) were further studied. Using reverse transcription-polymerase chain reaction and high-throughput sequencing, VP7and VP4sequences were obtained and their genetic characteristics, as well as the differences in antigenic epitopes of VP7, were analyzed in detail. Seven rotavirus genotypes were identified. The predominant rotavirus genotype was G9P [8] (77.0%), followed by prevalent strains G8P [8] (8.0%), G3P [8] (3.5%), G3P [9] (1.5%), G1P [8] (1.0%), G2P [4] (1.0%), and G4P [6] (1.0%). The amino acid sequence identities of G1, G2, G3, G4, G8, and G9 genotypes of isolates compared to the vaccine strains were 98.8%, 98.2%–99.7%, 88.4%–99.4%, 98.2%, 94.2%–100%, and 93.9%–100%, respectively. Notably, the vaccine strains exhibited high similarity in amino acid sequence, with only minor differences in antigenic epitopes compared to the Chinese endemic strains. This supports the potential application of the vaccine in preventing diseases caused by rotaviruses.

Topological photonics is rapidly expanding. However, discovering three-dimensional topological electromagnetic systems can be more challenging than electronic systems for two reasons. First, the vectorial nature of electromagnetic waves results in complicated band dispersions, and simple tight-binding-type predictions usually fail. Second, topological electromagnetic surface modes inside the light cone have very low quality factors (Qfactors). Here, we propose the concept of scalar topological photonics to address these challenges. Our approach is experimentally validated by employing a nested meta-crystal configuration using connected coaxial waveguides. They exhibit scalar-wave-like band dispersions, making the search for photonic topological phases an easier task. Their surface states have skyrmion-like electric field distributions, resulting in a whole, bright surface state band inside the light cone continuum. As such, the topological surface states in our three-dimensional nested crystals can be exposed to air, making such systems well-suited for practical applications.

This article proposes a cognitive frequency diversity array (FDA) multiple-input multiple-output (MIMO) (FDA-MIMO) radar network (CFDA-MIMON) target discrimination and tracking algorithm under main-lobe deceptive trajectory interference. This algorithm can be divided into two stages: search-discrimination-initialization and cognitive target tracking. The first stage mainly uses the distribution characteristics of target and main-lobe deceptive trajectory interference on different radars to discriminate the target and interference in the radar network. A target state initialization strategy driven by G-pair measurements with large time interval is proposed for target tracking. In the second stage, a cognitive target tracking algorithm of single FDA-MIMO radar system is designed based on the criteria of maximum Capon power spectrum. In addition, in order to avoid the divergence problem of the extended Kalman filter, a target state correction method based on auxiliary particles is proposed. Finally, the fusion strategy of multiple radar trajectories is used to alleviate the estimation error of the single radar. Numerical experiments verify that the proposed CFDA-MIMON target discrimination and tracking algorithm can distinguish the target from main-lobe deceptive trajectory interference without prior information of the target, and output the target's trajectory.

AbstractAiming at the intercoupling of multiple physics and the time-varying characteristics of boundary conditions in the microwave heating process, how to optimize the temperature uniformity of materials with a cooperative-mode stirrer is focused in this paper. First, the formation of standing waves can be effectively avoided by introducing the mode stirrer, thereby improving the heating efficiency of the material. Description of pattern stirrer motion processes using implicit functions and level set methods. Second, in order to optimize the temperature uniformity of the material, the stirrer is made intelligent, and the electromagnetic field distribution can be optimized by using different stirrer positions and directions, and the temperature uniformity is improved through intelligent coordination of stirrers in different positions. Finally, the finite element method is used to perform numerical simulation calculations on integer and continuous variables. The results show that the proposed method can improve the temperature uniformity of the conventional microwave heating model by 28.2%–81.44% and 23.2%∼63.91% at each level and lead hammer section respectively, and can improve the heating efficiency by 14.8%∼36.4%. According to its performance when heating different shapes and different materials, the proposed method is efficient and feasible.

The effects of welding layer arrangement on the microstructure and mechanical properties of 5083/6005A aluminium alloy gas metal arc welded (GMAW) joints were studied. High-quality butt joints were obtained under different welding layer arrangements, including two layers and two passes, two layers and three passes, and three layers and three passes. The results indicated that the softening zone located in the heat-affected zone (HAZ) of the 6005A aluminium alloy side was the weakest part of the 5083/6005A alloy joints. The highest level of recrystallization and precipitate dissolution was observed in the HAZ of the two-layer and two-pass joints; however, the lowest level was verified in the three-layer and three-pass joints. With the increase in the number of welding layers and passes, grain boundary liquation in the partially melted zone (PMZ) and strength loss of the softened zone were improved. The tensile strength of the two-layer and two-pass joints was 161 MPa, while that of the three-layer and three-pass joints increased to 217 MPa.

Recently, solid oxide-ion conductors have been receiving considerable attention owing to their potential applications in solid oxide fuel cells, oxygen sensors and so on. An innovative solid oxide-ion conductor, lanthanum–molybdenum oxide (La2Mo2O9), presents a reversible phase transformation at around 580°C from the low-temperature form α-La2Mo2O9to the high-temperature form β-La2Mo2O9, leading to variation in the ionic conductivity. This paper reviews the research progress of lanthanum–molybdenum oxide and its doping systems; the structure and phase transition of the material, the conductivity of the oxide ion and the ionic conductivity and chemical stability of the material in a reducing atmosphere and at high temperatures are discussed. The research progress of the lanthanum–molybdenum oxide electrolyte was reviewed from four aspects – structure, conduction mechanism, preparation method and conductivity – and its future prospects assessed.

Stem cell-based transplantation is a promising therapeutic approach for intervertebral disc degeneration (IDD). Current limitations of stem cells include with their insufficient cell source, poor proliferation capacity, low nucleus pulposus (NP)-specific differentiation potential, and inability to avoid pyroptosis caused by the acidic IDD microenvironment after transplantation. To address these challenges, embryo-derived long-term expandable nucleus pulposus progenitor cells (NPPCs) and esterase-responsive ibuprofen nano-micelles (PEG-PIB) were prepared for synergistic transplantation. In this study, we propose a biomaterial pre-modification cell strategy; the PEG-PIB were endocytosed to pre-modify the NPPCs with adaptability in harsh IDD microenvironment through inhibiting pyroptosis. The results indicated that the PEG-PIB pre-modified NPPCs exhibited inhibition of pyroptosis in vitro; their further synergistic transplantation yielded effective functional recovery, histological regeneration, and inhibition of pyroptosis during IDD regeneration. Herein, we offer a novel biomaterial pre-modification cell strategy for synergistic transplantation with promising therapeutic effects in IDD regeneration.

Facial expressions are generally recognized based on handcrafted and deep-learning-based features extracted from RGB facial images. However, such recognition methods suffer from illumination/pose variations. In particular, they fail to recognize these expressions with weak emotion intensities. In this work, we propose a cross-modality attention-based convolutional neural network (CM-CNN) for facial expression recognition. We extract expression-related features from complementary facial images (gray-scale, local binary pattern, and depth images) to handle the illumination/pose variations and to capture appearance details that describe expressions with weak emotion intensities. Rather than directly concatenating the complementary features, we propose a novel cross-modality attention fusion network to enhance spatial correlations between any two types of facial images. Finally, the CM-CNN is optimized with an improved focal loss, which pays more attention to facial expressions with weak emotion intensities. The average classification accuracies on VT-KFER, BU-3DFE(P1), BU-3DFE(P2), and Bosphorus are 93.86%, 88.91%, 87.28%, and 85.16%, respectively. Evaluations on these databases demonstrate that our approach is competitive to state-of-the-art algorithms.

Pedestrian trajectory prediction is essential for self-driving vehicles, social robots, and intelligent monitoring applications. Diverse trajectory data is critical for high-accuracy trajectory prediction. However, the trajectory data is captured in scattered scenes, which can cause the problem of data island. Furthermore, artificial aggregation of trajectory data suffers from the risk of data leakage, ignoring the rule of privacy protection. We propose a multi-scene federated trajectory prediction (Fed-TP) method to solve the above problems. As our key contribution, a destination-oriented LSTM (Long Short-Term Memory)-based trajectory prediction (DO-TP) network is proposed in each scene to forecast future trajectories in an encoder-decoder manner. The independent training using trajectory data in each scene can prevent data leakage and achieves high privacy security. As another key contribution, a federated learning framework is introduced to break the scene limitation by conducting distributed collaborative training. The performance of different federated learning methods is compared on public datasets, including ETH, UCY, and Stanford Drone Dataset (SDD). Compared with FedAvg and FedProx, FedAtt is more suitable for pedestrian trajectory prediction. Experimental results demonstrate that the proposed method has better data privacy security than directly training on multiple scenes and superior prediction performance than training on a single scene.

In the process of prediction and health management (PHM) about remaining useful life (RUL) of aero-engine, the state evaluation of aero-engine by constructing health index (HI) plays a very important role. At present, in the research of aero-engine state evaluation, there are still many problems: the analysis of the impact between operating conditions and sensor measurement data is insufficient, the standard for sensor data sequence selection is not established, and the HI construction is not reasonable. Aiming at these problems, firstly, a data reconstruction method is used to standardize the sensor data sequences, so as to eliminate the coupling effect between operating conditions and sensor measurement data. Secondly, different from the screening of sensor data sequences by observing the trend manually, a monotonicity standard is proposed. Then, a HI calculation method based on weighted Hellinger distance is proposed, which can construct HI more reasonably and better characterize the health condition of aero-engine. Finally, based on the standard data set of aero-engine degradation process provided by NASA in 2008, a typical HI construction method based on linear regression mapping and a similarity-based RUL prediction method are selected for comparison to verify performance of this method. The results show that the proposed method has better evaluation performance, and the HI constructed by this method is also better in RUL prediction.

Aiming at the problem of optimising the temperature field distribution of a multi microwave source agent system, in this work, group consistency is used to explore the multi microwave source agent collaborative tracking strategy to improve both the uniformity of the temperature distribution and the heating efficiency. First, to coordinate the inputting power state information of the multi microwave source agents, a consistency algorithm of non-fully connected communication topology is adopted. Second, according to the efficiency state information of the microwave source agents in different distribution positions of the cavity, the agent group consistency strategy and the group consistency strategy with expected power deviation are used to adjust the power distribution of each agent. Finally, the feasibility and effectiveness of the proposed strategy are verified by using heating materials with different specifications. The experimental results show that compared with the traditional multisource heating method the proposed temperature distribution optimization strategy based on group consistency can improve the temperature uniformity and heating efficiency.

Pedestrian trajectory prediction in dynamic scenes remains a challenging and critical problem in numerous applications, such as self-driving cars and socially aware robots. Challenges concentrate on capturing pedestrians’ motion patterns and social interactions, as well as handling the future uncertainties. Recent studies focus on modeling pedestrians’ motion patterns with recurrent neural networks, capturing social interactions with pooling- or graph-based methods, and handling future uncertainties by using the random Gaussian noise as the latent variable. However, they do not integrate specific obstacle avoidance experiences (OAEs) that may improve prediction performance. For example, pedestrians’ future trajectories are always influenced by others in front. Here, we propose the Graph-based Trajectory Predictor with Pseudo-Oracle (GTPPO), an encoder–decoder-based method conditioned on pedestrians’ future behaviors. Pedestrians’ motion patterns are encoded with a long short-term memory unit, which introduces temporal attention to highlight specific time steps. Their interactions are captured by a graph-based attention mechanism, which draws OAE into the data-driven learning process of graph attention. Future uncertainties are handled by generating multimodal outputs with an informative latent variable. Such a variable is generated by a novel pseudo-oracle predictor, which minimizes the knowledge gap between historical and ground-truth trajectories. Finally, the GTPPO is evaluated on ETH, UCY, and Stanford Drone datasets, and the results demonstrate state-of-the-art performance. Besides, the qualitative evaluations show successful cases of handling sudden motion changes in the future. Such findings indicate that GTPPO can peek into the future. [ABSTRACT FROM AUTHOR]

AbstractIn mathematical models of microwave heating with infinite-dimensional characteristics, it is difficult to use traditional numerical methods to improve computational efficiency. In this work, we propose a fast and accurate method to calculate the temperature distribution of materials under microwave heating. First, we analysed the relationship between the choice of model order and the solution accuracy by downscaling the infinite-dimensional heat conduction partial differential equation (PDE) model into a finite-dimensional ordinary differential equation (ODE) model. Additionally, the effect of different boundary conditions on the global temperature distribution was analysed. Second, the equilibrium conversion matrix was calculated using the singular value decomposition (SVD) truncation method under homogeneous boundary conditions. Using this matrix, a lower-dimensional microwave heating ODE model was further obtained. Finally, the numerical simulation results showed that the root mean square error (RMSE) was only 0.07 and the maximum relative error was only −0.85%. The computation time of the equilibrium conversion matrix was 2.12 ∼ 3.00 ms, and the model calculation time was reduced by 97.78%. We compared the calculated temperature rise curves with those obtained using the conventional COMSOL model. The SVD truncation method achieved an efficient and accurate solution for the microwave heating model.

The Al butt-lap joints are widely applied in China Railway High-speed (CRH) trains body joints. However, the gaps caused by the thermal deformation in Al butt-lap joints lead to the lack penetration or over penetration. To predict the penetration state of laser welds in Al butt-lap joints with different gaps, this work investigated the correlation between the plasma plume morphology and penetration state monitored by high-speed imaging system and the vertical gap and horizontal gap were utilized as intermediary. The ensemble experience mode decomposition (EEMD) was used to obtain the frequency features of plasma plume morphology. A support vector machine (SVM) model combined with EEMD was then established to classify the different penetration state, which adopted the original signals at time-domain and the mode decomposition signals at frequency-domain as input. The EEMD-SVM accuracy of 97.98% was the highest among the EMD-SVM of 86.44% and the SVM of 58.15%. The EEMD-SVM obtained both high Accuracyand Recallwhen the quantity of positive and negative samples were utmost different. The classification of the EEMD-SVM model was the most superior among the EMD-SVM model and SVM model. This proposed method provided a novel and accurate approach to perform process monitoring and penetration defects detection during laser welding of butt-lap joints.

Gas-coupled Stirling-type pulse tube cryocooler (SPTC) is currently the most compact and simplest configuration among all types of cryocoolers, but it is challenging to achieve a very low temperature. This paper investigates a gas-coupled SPTC which is capable of directly achieving a temperature of around 4 K. Theoretical analyses were performed based on SAGE to study the effects of employing one or more multi-bypass structures on apparent cooling performance, and internal working parameters. The simulation results indicate that the function of the multi-bypass is similar to that of a multi-stage gas-coupled structure, producing a pre-cooling effect on the lower-temperature section by increasing the acoustic power and the enthalpy flow in the pulse tube of the higher-temperature section. The cooperation of two multi-bypass structures can promote a higher enhancement of the cooling performance, but it is difficult to achieve the same cooling performance of a completely multi-stage gas-coupled SPTC due to weak phase-shifting capability and excessive reduction of the mass flow. Based on the model, the developed prototype has achieved a no-load temperature of 4.4 K, which shows the great potential of using a gas-coupled SPTC to obtain a cooling temperature below 4 K.

Owing to the rapid increase in the scale of grid connections of uncertain energy sources, such as wind and solar, the regulation capacity of grids has been challenged, and the development of large-scale energy storage technologies has become urgent. Hence, hydraulic compressed air energy storage technology has been proposed, which combines the advantages of pumped storage and compressed air energy storage technologies. This technology offers promising applications and thus has garnered considerable attention in the energy storage field. Herein, research achievements in hydraulic compressed air energy storage technology are reviewed. The operating principle and performance of this technology applied to six systems are summarized. The application prospects in power generation, grids, and microgrid systems are discussed. The technical bottlenecks encountered in the further development of this technology are analyzed. Finally, some problems that must be solved are identified. This review provides support for the further development of hydraulic compressed air energy storage technology.

Over the past two decades, superhydrophobic surfaces that are easily created have aroused considerable attention for their superior performances in various applications at room temperature. Nowadays, there is a growing demand in special fields for the development of surfaces that can resist wetting by high-temperature molten droplets (>1200 °C) using facile design and fabrication strategies. Herein, bioinspired directional structures (BDSs) were prepared on Y2O3-stabilized ZrO2(YSZ) surfaces using femtosecond laser ablation. Benefiting from the anisotropic energy barriers, the BDSs featured with no additional modifiers showed a remarkable increase from 9.2° to 60° in the contact angle of CaO–MgO–Al2O3–SiO2(CMAS) melt and a 70.1% reduction in the spreading area of CMAS at 1250 °C, compared with polished super-CMAS-melt-philic YSZ surfaces. Moreover, the BDSs demonstrated exceptional wetting inhibition even at 1 400 °C, with an increase from 3.3° to 31.3° in contact angle and a 67.9% decrease in spreading area. This work provides valuable insight and a facile preparation strategy for effectively inhibiting the wetting of molten droplets on super-melt-philic surfaces at extremely high temperatures.

Ceramic coatings were prepared on 6061 aluminum alloy in electrolytes containing La2O3particles using the micro-arc oxidation technique. The main work focuses on the microstructure, phase composition, elemental distribution and corrosion resistance of micro-arc oxidation coatings. The results showed that the addition of La2O3increased the oxidation voltage and promoted the formation rate. Scanning electron microscopy observations indicated that the quantity of discharge micropores decreased and the thickness increased, and thus, the microhardness increased. The X-ray diffraction results showed that the coatings mainly consisted of γ-Al2O3, α-Al2O3and SiO2. Electrochemical corrosion tests showed that the corrosion resistance of the coated sample with La2O3increased greatly because the addition of La2O3significantly reduced the micropores of the coating surface and made the coating thicker, smoother and harder. Hence, the addition of La2O3can optimize the structure of the micro-arc oxidation coating and improve its corrosion resistance.

Advancement in human-machine interface (HMI) is indispensable for development of virtual/augmented reality, robotics control, and artificial intelligence (AI). However, there lack a simple facile electronics to achieve accurate finger motion recognition and remote robot control through simple signal processing. Herein, a wearable patterned single electrode-based triboelectric nanogenerator (PS-TENG) based on regenerated silk film (rSF) is developed for peak-encoded interface, which can accurately identify the handwritten numbers or letters. The patterned single electrode, composed of nine units with five different patterns, is successfully constructed by ink-jet printing Ag inks on the rSF. When handwriting on the PS-TENG, the generated output voltage with different peak amplitude can be transformed into a peak spectrum with spatial information. Handwriting numbers or letters can be accurately identified through simple digitization data processing of the peak-encoded spectrum without complex algorithms. Furthermore, the PS-TENG as peak-encoded interface enables precise game play, accurate pattern lock, and wireless robot control. Importantly, based on relative amplitude of output signals, the peak-encoded code is regardless of factors including absolute amplitude of output signals, sliding speed and environmental humidity. Consequently, the PS-TENG, characterized by self-powered feature, simple structure and straightforward signal processing, demonstrates remarkable identification accuracy of handwritten numbers and letters through decoding unique and stable peak spectrum, which would have great potential in the field of virtual control panels, HMI and AI.

ABSTRACTChronic stress refers to nonspecific systemic reactions under the over-stimulation of different external and internal factors for a long time. Previous studies confirmed that chronic psychological stress had a negative effect on almost all tissues and organs. We intended to further identify potential gene targets related to the pathogenesis of chronic stress-induced consequences involved in different diseases. In our study, mice in the model group lived under the condition of chronic unpredictable mild stress (CUMS) until they expressed behaviors like depression which were supposed to undergo chronic stress. We applied high-throughput RNA sequencing to assess mRNA expression and obtained transcription profiles in lung tissue from CUMS mice and control mice for analysis. In view of the prediction of high-throughput RNA sequences and bioinformatics software, and mRNA regulatory network was constructed. First, we conducted differentially expressed genes (DEGs) and obtained 282 DEGs between CUMS (group A) and the control model (group B). Then, we conducted functional and pathway enrichment analyses. In general, the function of upregulated regulated DEGs is related to immune and inflammatory responses. PPI network identified several essential genes, of which ten hub genes were related to the T cell receptor signaling pathway. qRT-PCR results verified the regulatory network of mRNA. The expressions of CD28, CD3e, and CD247 increased in mice with CUMS compared with that in control. This illustrated immune pathways are related to the pathological molecular mechanism of chronic stress and may provide information for identifying potential biomarkers and early detection of chronic stress.

Understanding agents’ motion behaviors under complex scenes is crucial for intelligent autonomous moving systems (like delivery robots and self-driving cars). It is challenging duo to the inherent uncertain of future trajectories and the large variation in the scene layout. However, most recent approaches ignored or underutilized the scenario information. In this work, a Multi-Granularity Scenarios Understanding framework, MGSU, is proposed to explore the scene layout from different granularity. MGSU can be divided into three modules: (1) A coarse-grained fusion module uses the cross-attention to fuse the observed trajectory with the semantic information of the scene. (2) The inverse reinforcement learning module generates optimal path strategy through grid-based policy sampling and outputs multiple scene paths. (3) The fine-grained fusion module integrates the observed trajectory with the scene paths to generate multiple future trajectories. To fully explore the scene information and improve the efficiency, we present a novel scene-fusion Transformer, whose encoder is used to extract scene features and the decoder is used to fuse scene and trajectory features to generate future trajectories. Compared with the current state-of-the-art methods, our method decreases the ADE errors by 4.3% and 3.3% by gradually integrating different granularity of scene information on SDD and NuScenes, respectively. The visualized trajectories demonstrate that our method can accurately predict future trajectories after fusing scene information.

ABSTRACTTo probe into the potential mechanism of microRNA (miR)-98-5p inhibiting the biological progress of cervical cancer (CC) cells via regulating PI3K/Akt pathway. Reverse transcription quantitative polymerase chain reaction was applied to detect miR-98-5p expression in CC tissues and cell lines; Cell counting kit-8 and Edu analysis were performed for checking cell proliferation, flow cytometry for cell apoptosis, transwell for cell invasion and migration, Western blot for proliferation-related proteins Ki67 and Proliferating cell nuclear antigen expression, apoptosis-related proteins Bcl-2 and Bax expression, epithelial–mesenchymal transition (EMT)-related proteins Snail, matrix metalloproteinase-3, E-cadherin and N-cadherin expression, as well as PI3K/Akt pathway-related proteins PTEN, PI3K as well as Akt expression levels, and the nude mouse tumor xenograft experiment was applied to verify in vivo. The result clarified, miR-98-5p was reduced in CC. Overexpression miR-98-5p could inhibit CC cell proliferation, invasion, migration and EMT, whereas promoted its apoptosis, but silencing miR-98-5p was opposite. Overexpression miR-98-5p could depress the activation of PI3K/Akt pathway in CC in vivoand in vitro. MiR-98-5p targeted CBX5. In short, miR-98-5p is able to be used as a potential target for treating CC in future research.

The present study investigated the influence of energy ratio on microstructure and mechanical properties in the transition zone of hybrid laser-MIG welded 316L/AH36 joints. The region between the laser zone and hybrid zone was defined as the transition zone. The ratio of laser to arc energy was defined as energy ratio. Similar microstructure types including ferrite, austenite, and martensite were observed in the transition zone under each energy ratio. However, obvious difference in microstructure distribution and crystallographic characteristic in transition zone was found in varying energy ratio. With the energy ratio decreasing from 2:1 to 1:2, the fraction of martensite was decreased from 83.1% to 65.6%, the fraction of high angle grain boundaries was increased from 44.8% to 61.3% and the average grain size was increased from 3.62 to 17.20 μm. The phenomenon was attributed to the improvement of cooling rate and reduction of filler metal. Therefore, the highest average microhardness value of 445 HV in the fusion zone was obtained under the highest energy ratio compared to that of 223 HV under the lowest energy ratio. Besides, the low-temperature impact absorption energy was increased from 41 to 90 J and the tensile fracture mode was changed from cleavage fracture mode to ductile fracture mode as the energy ratio decreased from 2:1 to 1:2.

In the last few decades, topological phase1–11has emerged as a new classification of matter states beyond the Ginzburg–Landau symmetry-breaking paradigm. The underlying global invariant is usually well characterized by integers, such as Chern numbers or winding numbers—the Abelian charges12–15. Very recently, researchers proposed the notion of non-Abelian topological charges16–19, which possess non-commutative and fruitful braiding structures with multiple (more than one) bandgaps tangled together. Here we experimentally observe the non-Abelian topological charges in a time-reversal and inversion-symmetric transmission line network. The quaternion-valued non-Abelian topological charges are clearly mapped onto an eigenstate-frame sphere. Moreover, we find a non-Abelian quotient relation that provides a global perspective on the distribution of edge/domain-wall states. Our work opens the door towards characterization and manipulation of non-Abelian topological charges, which may lead to interesting observables such as trajectory-dependent Dirac/Weyl node collisions in two-dimensional systems16,17,20, admissible nodal line configurations in three dimensions16,19,20, and may provide insight into certain strongly correlated phases of twisted bilayer graphene21.

We use broadband records from a dense seismic network deployed in and around the Qaidam Basin in northwestern China to analyze the crustal phases and investigate the depth of the Conrad and Moho discontinuities as well as the P‐wave velocity. Waveform cross‐correlation is used to assist in the identification of the crustal phases and in determining their arrival times. Depth of the Conrad discontinuity is determined by fitting the travel times of Conrad‐diffracted P‐waves using a two‐layer model. The depth of the Conrad discontinuity under the eastern part of the basin is shallower than the western part, which can be attributed to different crustal shortening mechanisms. The upper crust shortening in the western part of the basin leads to thickening of the upper crust, while multiple thrust faults result in the rise of the Conrad discontinuity in the east. These two different mechanisms determine the depth change of the Conrad discontinuity in the basin from the west to the east, which is supported by the results in this study. Broadband records from a dense seismic network were used to analyze the crustal phases and investigate the depth of the Conrad and Moho discontinuities in the Qaidam Basin.The depth of the Conrad discontinuity under the eastern part of the basin is shallower than the western part, which can be attributed to different crustal shortening mechanisms.The depth of the Moho discontinuity under the Qilian orogenic belt and Bayan Har block is deeper than under the Qaidam Basin, which is consistent with the previous results. Broadband records from a dense seismic network were used to analyze the crustal phases and investigate the depth of the Conrad and Moho discontinuities in the Qaidam Basin. The depth of the Conrad discontinuity under the eastern part of the basin is shallower than the western part, which can be attributed to different crustal shortening mechanisms. The depth of the Moho discontinuity under the Qilian orogenic belt and Bayan Har block is deeper than under the Qaidam Basin, which is consistent with the previous results.