Search

Unmanned aerial vehicles (UAVs) with an ultra-large aspect ratio and a joined-wing configuration exhibit characteristics such as minimal mass, remarkable flexibility, and a unified wing structure, thereby presenting a significant challenge in accurately assessing their dynamic behaviors in terrestrial environments. This study focuses on a specific UAV model characterized by an ultra-large aspect ratio and a joined-wing configuration, which underwent experimental scrutiny under free-free conditions, encompassing the entire aircraft. A rubber rope suspension mechanism was employed to support the UAV, and deliberate efforts were made to eliminate the additional stiffness effect, thus facilitating precise determination of the aircraft’s modal characteristics. Subsequent ground-based servoelasticity tests were conducted based on modal experimentation, and specialized filtering techniques were developed to enhance stability. Finally, an analysis of aeroservoelastic stability revealed a strong correlation between simulation and experimental results, with the UAV demonstrating high stability margins in aeroservoelasticity.

In recent years, collaborative classification of multimodal data, e.g., hyperspectral image (HSI) and light detection and ranging (LiDAR), has been widely used to improve remote sensing image classification accuracy. However, existing fusion approaches for HSI and LiDAR suffer from limitations. Fusing the heterogeneous features of HSI and LiDAR proved to be challenging, leading to incomplete utilization of information for category representation. In addition, during the extraction of spatial features from HSI, the spectral and spatial information are often disjointed. It leads to the difficulty of fully exploiting the rich spectral information in hyperspectral data. To address these issues, we proposed a multimodal data fusion framework specifically designed for HSI and LiDAR fusion classification, called modality fusion vision transformer. We have designed a stackable modality fusion block as the core of our model. Specifically, these blocks mainly consist of multimodal cross-attention modules and spectral self-attention modules. The proposed novel multimodal cross-attention module for feature fusion addresses the issue of insufficient fusion of heterogeneous features from HSI and LiDAR for category representation. Compared to other cross-attention methods, it reduces the alignment requirements between modal feature spaces during cross-modal fusion. The spectral self-attention module can preserve spatial features while exploiting the rich spectral information and participating in the process of extracting spatial features from HSI. Ultimately, we achieve overall classification accuracies of 99.91%, 99.59%, and 96.98% on three benchmark datasets respectively, surpassing all state-of-the-art methods, demonstrating the stability and effectiveness of our model.

Sulfur hexafluoride (SF6) is the most widely used gas insulation dielectric in power system, Currently, SF6 degradation technology based on non-thermal plasma faces issues with high content of difficult-to-treat byproducts like SO2F2. This study proposes improving the SF6 destruction and removal efficiency (DRE) through tandem DBD degradation and investigates the impact of adding H2 on the tandem DBD degradation of SF6 and its decomposition products. The results show that under conditions of 2% initial SF6 concentration, 1.5% H2, 150 mL/min gas flow rate, and 80W input power, the DRE in the first-stage tube is 73.2%, reaching 78.02% in the second-stage tube, and the DRE of SF6 in the dual-stage tandem reaches as high as 82.82%. In terms of decomposition products, there is no difference in the types of products formed in the two reaction tubes when H2 is added, and the main products are SO2F2, SO2, SOF2, and SOF4. The addition of H2 will inhibit the formation of S-O-F gas products, improves DRE and significantly regulates the distribution of degradation products, reducing the amount of SO2F2 generated in the second-stage tube and making SO2 the main product. Adding H2 with a concentration of 2.5 % can avoid the formation of SO2F2 in the second-stage tube. This study provides guidance for efficient SF6 degradation and product control.

Owing to the tolerance of circuit components in analogue circuits, feature overlap occurs between different component faults, especially in the case of early-stage faults. To address this issue, we present an improved model that utilizes the channel attention mechanism in conjunction with a fully convolutional network. The proposed model conducts end-to-end diagnosis by utilizing the response signals generated from the tested circuit. To diagnose different fault categories, we add a global average pooling layer to the last layer of the fully convolutional network to obtain the probabilities of various fault categories. Furthermore, we incorporating a channel attention, which allocates distinct weight coefficients to each channel to amplify crucial features while dampening less significant ones, within the network, which allocates distinct weight coefficients to each channel to amplify crucial features while dampening less significant ones. The proposed method is validated with four-opamp biquad high-pass filter circuit and leapfrog filter circuit. The results demonstrate that the proposed model achieves a fault diagnosis accuracy of 95.24% in Case 1,87.09% in Case 2. This method exhibits higher diagnostic accuracy and reliability than existing methods.

In response to the intricate interplay of uncertain factors within the airport system, which leads to challenges in predicting and optimizing departure operations in the airport, this study investigates a probabilistic prediction of operation times for departure flights, as well as an integrated stochastic optimized scheduling method considering multiple resources on the airport surface under the impact of these uncertain operation times. A probabilistic predictive model for departure operations is formulated, employing a combination of random forest regression and kernel density estimation. This model was used to obtain the prediction results of the probabilistic distribution of various departure operation times under uncertainty. Chance constraints are introduced to control the stability of the scheduling strategies including the push back time and takeoff time scheduling. Taking minimizing operational delays and maximizing the probability of satisfying the chance constraints as the optimization objectives, a comprehensive integrated stochastic scheduling model is developed for airport departure operations under uncertainty. Lastly, the method is validated and analyzed using Kunming Changshui International Airport as a case study. Experimental results firstly demonstrate that the proposed method is capable of effectively predict the probability distributions of estimated push back times and taxi times for departure flights. The continuous ranked probability scores of the estimated push back times and taxi times are 9.25 and 1.21 respectively, indicating that the estimated push back times of the departure flights is more uncertain than the taxi times. Then, results of the integrated airport surface departure scheduling model show that the model can comprehensively consider the minimization of flight delays and the satisfaction of the two chance constraints, and obtain the Pareto optimal solutions achieving balance between strategy efficiency and strategy stability.

Psychological stress represents a prevalent societal concern, particularly evident during crises such as epidemics, which exacerbate various forms of psychological strain. Prompt mitigation of psychological stress holds potential for enhancing quality of life and fostering societal advancement. In this study, drawing inspiration from the International Emotion Picture Library, we crafted four virtual reality (VR) scenes augmented with five emblematic animals, resulting in 20 VR scenarios for psychological stress assessment. Fifteen participants engaged in the test, which was preceded by the completion of the World Health Organization Quality of Life Questionnaire (WHOQOL-BREF). Throughout the test, EEG and heart rate data were meticulously recorded. Following each scenario, participants completed the Self-Assessment Manikin (SAM) scale, culminating in the formation of the VR psychological stress reduction dataset. We meticulously analyze the SAM scale. For EEG data, we devised a deep learning model leveraging a hybrid CNN-LSTM architecture for signal classification. The model was evaluated on both the publicly available DEAP dataset and our collected DSRP dataset, yielding superior performance. Leveraging the dataset amassed in this study, we scrutinize the alleviative impact of VR on human psychological stress, providing foundational insights for VR-based psychological stress reduction research. Moreover, the EEG classification model proposed herein enhances the categorization of EEG signals, thereby advancing the domain of emotion recognition grounded in EEG technology.

In Unmanned Aerial Vehicle (UAV) systems, packet loss during sensor data transmission causes data missing, which reduces fault features in sensor signals and causes the accuracy of state monitoring to decrease. This study proposes a state monitoring algorithm combining a convolutional neural network (CNN) with a random forest (RF) for data missing scenarios. CNN algorithm is designed to extract the distributed fault information from the available signals and acquire the state features of the system. Random forest algorithm processes the state features and judges the system state. The integrating strategy utilizes the automatic feature extraction capability of CNN and the superior discrimination capability of an RF classifier to improve the state monitoring accuracy. The experimental results show that the accuracy of state monitoring in data missing condition reaches 92.74%. The comparative experiments verify the validity of the proposed algorithm.

Ethical concerns about how artificial intelligence (AI) impacts individuals and society are increasing rapidly, but few studies have systematically investigated the public awareness of AI ethics. This research collected and analyzed data from a public poll in Taiwan, an Asian region with a developed economy and specific social conditions, to identify societal views on AI ethics. The analysis of 84 AI ethics guidelines worldwide provided the survey framework covering five ethical principles: transparency, fairness, privacy, nonmaleficence, and accountability. The overarching goal was to determine the commonalities and differences in the ethical concerns of Taiwanese laypersons toward AI. Participants aged from 20 to 70 ( $N =1$ ,200) completed a computer-assisted random-digit-dial telephone survey, which utilized ethical scenarios to capture social views, and item validity was confirmed using focus-group interviews. Results found that respondents concerned about nonmaleficence the most, emphasizing that AI applications should not harm humans. Taiwanese people therefore tended to support strict AI technology regulation. It was particularly interesting that different patterns of public concern emerged about accountability, with the opinions on attributing responsibility to stakeholders varying with scenarios and the public’s backgrounds. Those with higher education levels tended to attribute more responsibility to the industry, whereas those who had only received elementary-school education attributed accountability to AI developers. For self-driving cars, accountability was attributed to AI developers, whereas for medical decision-making, the accountability was attributed to the hospitals. These findings may help to elucidate the associations between societal views and the ethical principles of AI worldwide.

Millimeter-wave (MMW) point clouds, characterized by their low resolution and high noise, limit the detection accuracy of point-based IA-SSD method due to the inadequate consideration of contextual information in MMW scenarios. Therefore, this paper proposes a three-dimensional (3D) MMW object detector, greatly augmenting the detection performance of the baseline model IA-SSD by the integration of the local-global context information. Central to our approach is the implementation of a multi-scale feature aggregation (MFA) module in the encoder stage of IA-SSD, which utilizes a self-attention mechanism to apprehend local contextual distinctions. This module is further applied to the centroid aggregation stage to enhance the capture of local context from foreground points. Complementarily, a global feature fusion module is devised to combine global contextual insights, drawing upon the localized information delineated by the MFA modules. This integrated framework significantly diminishes the false detection rate while concurrently elevating the detection precision for occluded objects. Relative to the IA-SSD baseline, the empirical evaluations validate the efficiency of our proposed model, demonstrating marked decreases in false positives and false negatives. Specifically, there is a 2.78% and 7.39% improvement in AP_R40_0.25 and AP_R40_0.5, respectively. When the intersection-over-union threshold is set as 0.25 and 0.5, the corresponding recall rate increases by 2.13% and 6.2%, respectively. Moreover, the inference speed reaches 32.3 frames per second(FPS), only a slight decrease of 2.9 FPS compared to the baseline model. These results demonstrate that the proposed detector significantly enhances detection performance without compromising on speed, marking a considerable advancement in the domain of 3D MMW object detection.

In this study, an innovative method has been proposed for resource allocation among contractors in large construction projects. This method is designed based on a combination of machine learning techniques, fuzzy theory, and auction modeling. Resource allocation in the context of large construction projects, where multiple contractors work simultaneously, can pose a complex problem. Developing an efficient method to address this issue can contribute to improving project performance in terms of cost and construction delays. We have presented a three-stage method for resource allocation in large construction projects. In the first stage, machine learning techniques are utilized to develop two distinct neural network models for predicting costs and delays for each contractor. These models utilize the Genetic Algorithm (GA) to optimize their parameters. In the second stage, a fuzzy model is used, which takes inputs from the neural network models and other contractor-specific features. This model prioritizes the needs of the contractors. Finally, an auction model is employed to fairly distribute the limited project resources between the contractors in need. The implementation results indicate that the proposed method for allocating resources among contractors in large construction projects have achieved MAE and RMSE values of 18.88 and 22.98, respectively, demonstrating a significant performance improvement compared to other proposed methods.

Stroke is one of the leading causes of death and disability worldwide. The ideal solution to the stroke problem is to prevent it in advance by controlling metabolic factors, atrial fibrillation, hypertension, smoking, Etc. However, unless the physiological indicators are abnormal, it is difficult for medical personnel to decide whether special precautions are necessary for a patient based solely on monitoring the potential patient. There was a great category imbalance between stroke and non-stroke patients, so this study tried to use various techniques to solve the problem of categorical unbalanced stroke prediction problem. Then, deep learning models were used to predict whether the patients would have a stroke. Finally, the classification experiment is carried out through transfer learning to observe whether the evaluation metrics are further improved. According to the experimental results, this study effectively reduced the false negative rate (FNR) and false positive rate (FPR) of stroke prediction and improved the overall accuracy of stroke prediction through the category imbalance treatment and deep learning method.

The damage imaging method based on Lamb wave beamforming has been widely used in the field of SHM. The DAS method has a high imaging efficiency, but its ability to suppress interference signals is weak, resulting in low imaging resolution and signal-to-noise ratio. Drawing inspiration from the adaptive weighted MVDR damage imaging method, this paper constructs a neural network based on FCNN, with the images generated by the MVDR method as the target. By training the model, the mapping relationship between delayed channel input data and adaptive weighting factors is established, thereby improving the resolution and signal-to-noise ratio of Lamb wave damage imaging and achieving rapid imaging of damage. To verify the effectiveness and imaging performance of the FCNN method, imaging of two types of damage in aluminum plates is conducted through simulation and experiments, and the imaging results are compared and analyzed with DAS and MVDR. The results show that the imaging quality and the quantitative indicators of the FCNN method have not yet reached the performance level of the MVDR, but compared with DAS, FCNN has a significantly narrower main lobe width and lower sidelobe level. Furthermore, its quantitative indicators such as API, SNR, and FWHM are better than DAS. The proposed adaptive Lamb wave beamforming method based on FCNN combines high resolution and signal-to-noise ratio, as well as the advantage of rapid imaging, providing reference and support for real-time SHM based on Lamb waves.

The vehicle routing problem with time windows (VRPTW) is a well-known NP-Hard combinatorial optimization problem, which is frequently encountered in transportation and logistics scenarios. When traditional algorithms solve this problem, there are some problems, such as slow convergence speed. In this paper, a Self-competitive Particle Swarm Optimization (ScPSO) algorithm is proposed, which regulates the learning direction of the next generation of particles based on their degree of self-competition. When the particle’s self-competition degree is low, it is compelled to learn from the individual optimal solution using the self-competition selection probability, thereby achieving rapid convergence. ScPSO uses the nonlinear inertia weight adaptive strategy to adjust the search preference in the search process, and it can balance the relationship between exploration and exploitation. Meanwhile, Random greedy heuristic selection and variable neighborhood search strategies are used in initial solution construction and constraint restriction. Finally, ScPSO is tested on 56 Solomon 100-customers benchmark problems and compared with four state-of-the-art VRPTW algorithms and best-known solutions reported on Solomon’s webpage.The running results of the ScPSO algorithm are better than four state-of-the-art VRPTW algorithms and best-known solutions reported on Solomon’s webpage. The experimental results show that ScPSO can solve VRPTW problems efficiently.

Resting-state functional magnetic resonance imaging (rs-fMRI) has been widely used in the identification of mild cognitive impairment (MCI) research, MCI patients are relatively at a higher risk of progression to Alzheimer’s disease (AD). However, almost machine learning and deep learning methods are rarely analyzed from the perspective of spatial structure and temporal dimension. In order to make full use of rs-fMRI data, this study constructed a dynamic spatiotemporal graph neural network model, which mainly includes three modules: temporal block, spatial block, and graph pooling block. Our proposed model can extract the BOLD signal of the subject’s fMRI data and the spatial structure of functional connections between different brain regions, and improve the decision-making results of the model. In the study of AD, MCI and NC, the classification accuracy reached 83.78% outperforming previously reported, which manifested that our model could effectively learn spatiotemporal, and dynamic spatio-temporal method plays an important role in identifying different groups of subjects. In summary, this paper proposed an end-to-end dynamic spatio-temporal graph neural network model, which uses the information of the temporal dimension and spatial structure in rs-fMRI data, and achieves the improvement of the three classification performance among AD, MCI and NC.

This article discusses ‘Metaverse’ from a technical perspective, focusing on networked systems aspects. Based on a technical definition of the ‘Metaverse,’ we examine the current state and challenges in communication and networking within Metaverse systems. We describe the state-of-the-art in different enabling Metaverse technologies and provide a technical analysis of current Metaverse system architectures. We then detail the state-of-the-art and the gaps in four areas: communication performance, mobility, large-scale operation, and end system architecture. Based on our analysis, we formulate a vision for future Metaverse infrastructure, outlining goals, design concepts, and suggested research directions.

To realize the effective reuse of the existing model measurement process planning and to improve the efficiency of the preparation of the measurement program, this paper proposes an on-machine measurement model partial retrieval method based on geometric features. First, the B-Rep information and measurement procedures of the model are extracted, and combined with the analysis of the geometric features of the on-machine measurement, the model is transformed into an attribute adjacency labeling graph; next, the index tree is constructed by using hash coding, and the result set is obtained through the filtering-verification framework to realize the synthesis similarity evaluation of the geometrical level of the on-machine measurement model; lastly, different reuse methods of the measurement process planning are selected based on the model’s similarity value. Fast programming based on partial retrieval of on-machine measurement models is realized by directly selecting programs, calling macros, and assigning values to variables. The experimental results show that the method can realize the effective reuse of the existing model measurement process planning, which can improve the efficiency of the preparation of the measurement program and meet the actual needs of on-machine measurement.

Current control approaches for gross prosthetic arm movement mainly regulate movement over a continuous range of target poses. However, these methods suffer from output fluctuation caused by input signal variations during gross arm movements. Prosthesis control approaches with a finite number of discrete target poses can address this issue and reduce the complexity of the pose control process. However, it remains under-explored in the literature and suffers from the consequences of misclassifying the target poses. Here, we propose a novel Uncertainty-Aware Discrete-Target Prosthesis Control (UA-DPC) approach. This approach consists of (1) an uncertainty-aware classification scheme to reduce unintended pose switches caused by misclassifications, and (2) real-time trajectory planning that adjusts motion to be rapid or conservative based on low or high quantified uncertainty, respectively. By addressing the impact of misclassification, this approach facilitates more efficient and smooth movements. Human-in-the-loop experiments were conducted in a virtual reality environment with 12 non-disabled participants. The participants controlled a transhumeral prosthesis using three approaches: the proposed UA-DPC, a discrete-target approach based on a traditional off-the-shelf classifier, and a continuous-target approach. The results demonstrate the superior performance of UA-DPC, which provides more efficient task completion with fewer misclassification instances as well as smoother residual limb and prosthesis movement.

Through phase-amplitude analysis, this study investigated how low-frequency postural fluctuations interact with high-frequency scalp electroencephalography (EEG) amplitudes, shedding light on age-related mechanic differences in balance control during uneven surface navigation. Twenty young ( $24.1~\pm ~1.9$ years) and twenty older adults ( $66.2~\pm ~2.7$ years) stood on a training stabilometer with visual guidance, while their scalp EEG and stabilometer plate movements were monitored. In addition to analyzing the dynamics of the postural fluctuation phase, phase-amplitude coupling (PAC) for postural fluctuations below 2 Hz and within EEG sub-bands (theta: 4-7 Hz, alpha: 8-12 Hz, beta: 13-35 Hz) was calculated. The results indicated that older adults exhibited significantly larger postural fluctuation amplitudes(p ${p} =0.005$ ) than young adults. The PAC between postural fluctuation and theta EEG (FCz and bilateral temporal-parietal-occipital area), as well as that between postural fluctuation and alpha EEG oscillation, was lower in older adults than in young adults (p ${p}=0.006$ ), was higher in older adults than in young adults. In summary, the postural fluctuation phase and phase-amplitude coupling between postural fluctuation and EEG are sensitive indicators of the age-related decline in postural adjustments, reflecting less flexible motor state transitions and adaptive changes in error monitoring and visuospatial attention.

With the rapid advancements in deep learning technology, various deep learning-based object detection algorithms have found extensive applications in UAV-related tasks. However, motivated by the fact that current object detection algorithms for unimodal aerial remote sensing images fail to achieve around-the-clock object detection. To tackle this, we propose an around-the-clock object detection algorithm YOLOFIV that fuses infrared and visible features. First, we design a dual-stream backbone network based on the attention mechanism to adequately extract the features of both modalities. Moreover, the ECA attention mechanism is integrated into the feature enhancement network to amplify attention toward challenging detection scenarios. Finally, we improve the horizontal detection head to a rotating one to preserve object orientation. We evaluate the proposed method YOLOFIV on the widely used drone vehicle dataset, YOLOFIV achieves an accuracy of 64.71% (in terms of $\text{mean average precision}_{0.5}$), accuracy improvement of 8.32% over baseline bimodal model, similar performance to UACMD designed for ARSI object detection but with 92.35% reduction in parameter count, and 17.87 times speedup. The results show that our approach achieves round-the-clock object detection while maintaining a favorable accuracy-speed tradeoff.

With the rapid development of deep learning (DL), change detection (CD) in remote sensing (RS) image has achieved remarkable success. Nevertheless, as the image resolution improves, the visual features extracted by current methods have limited expression ability, and the networks generally suffer from spatial degradation, both of which lead to incomplete boundary detection and the undetection problem of small changed areas. At the same time, the registration errors in image pairs make remote sensing image change detection (RSCD) more challenging. To alleviate the aforementioned issues, this article proposes a transformer with feature interaction and fusion network (TFIFNet) for CD. To be specific, the proposed network utilizes the advantages of transformer in long-range dependence modeling first, which can learn feature representations with spatial-temporal information from a global perspective. Then, to alleviate the irrelevant changes caused by image registration errors, the bitemporal feature interaction module (BFIM) is proposed, which utilizes an attention mechanism to learn the bitemporal background distribution. Subsequently, an intertemporal joint-attention (JointAtt) is introduced to learn the consistency of bitemporal features for further refinement. Finally, to address the issue caused by spatial degradation during the process of network training, a triple feature fusion module (TFFM) is proposed. This module can learn spatial information from adjacent layer's features as additional spatial information. Extensive experimental studies show that the proposed network achieves the most advanced results on two CD benchmark datasets.

A hybrid wireless-wireline access network (HWWAN) is an integrated network infrastructure that combines wireless and wired technologies to offer users seamless and reliable connectivity. However, the existing resource management schemes for HWWANs have largely overlooked the stringent quality of service (QoS) requirements of latency-critical users. Moreover, the surge in the demands for ultra-broadband applications with extremely strict real-time requirements has resulted in an explosive growth of data traffic, rendering network resource allocation challenging for HWWANs. This paper proposes an efficient radio resource allocation scheme for HWWANs that effectively addresses the requirements of both latency-critical user equipment (UEs) and hybrid customer premises equipment (HCPEs) with throughput gap (TG) requirements. Specifically, we formulate a resource optimization problem to maximize the effective capacity (EC) of UEs while considering the TG requirements of HCPEs, and the availability of physical resource blocks (PRBs) at the base station. The problem is solved using convex optimization. Simulation results indicate the efficacy of the proposed scheme in improving the percentages of satisfied UEs and HCPEs, the sum EC of UEs, and the aggregate throughput of all the HCPEs.

Wide bandwidth requirements for multi-Gbps communications have prompted the global telecommunications industry to consider new mid-band spectrum allocations in the 4–8 GHz FR1(C) and 7–24 GHz FR3 bands, above the crowded bands below 6 GHz. Allocations in the lower and upper mid-band aim to balance coverage and capacity, but there is limited knowledge about the radio propagation characteristics in the 4–24 GHz frequency bands. Here we present the world’s first comprehensive indoor propagation measurement and channel modeling study at 6.75 GHz and 16.95 GHz in mid-band spectrum conducted at the NYU WIRELESS Research Center spanning distances from 11–97 m using 31 dBm EIRP transmit power with 15 and 20 dBi gain rotatable horn antennas at 6.75 GHz and 16.95 GHz, respectively. Analysis of the omnidirectional and directional propagation path loss using the close-in free space model with 1 m reference distance reveals a familiar waveguiding effect in indoor environments for line-of-sight (LOS). Compared to mmWave frequencies, the omnidirectional LOS and non-LOS (NLOS) path loss exponents (PLE) are similar, when using a close-in 1 m free space path loss reference distance model. Observations of the omnidirectional and directional RMS delay spread (DS) at FR1(C) and FR3 as compared to mmWave and sub-THz frequencies indicate decreasing RMS DS as the carrier frequency is increased. The RMS angular spreads (AS) at 6.75 GHz are found to be wider compared to 16.95 GHz, showing greater number of multipath components from a broader set of directions in the azimuthal spatial plane when compared to higher frequencies. This work also presents results from extensive material penetration loss measurements at 6.75 GHz and 16.95 GHz using co and cross polarized antenna configurations for ten common construction materials found inside buildings and on building perimeters, including concrete walls, low-emissivity glass, wood, doors, drywall, and whiteboard. Our findings show penetration loss increases with frequency for all of the ten materials and partitions tested, and suggest further investigation of 3GPP material penetration loss models for at least infrared reflective (IRR) glass and concrete may be necessary. The empirical data and resulting models for radio propagation and penetration loss presented in this paper provide critical information for future 5G and 6G wireless communications.

The fault characteristics of large-capacity wind farms affect the sensitivity of transmission-line protection, resulting in protection devices refusing to operate or mal-operate. A phase-to-ground fault-locating method with single-end transient information for wind farm transmission lines was proposed in this study. The proposed method employs a time-domain transient quantity integration equation matrix with the transmission line equivalent resistance, inductance, and fault path resistance to solve the fault location. The effectiveness of the scheme was evaluated using Real-Time Lab (RTLAB). The impact of factors such as the fault resistance, location, and distributed capacitance is considered. Under different fault locations, the distance measurement error of the proposed algorithm is within 3%, and there is no rejection or mal-operation. The results demonstrate that the proposed algorithm is characterized by high speed and reliability.

Chest X-rays have played an indispensable part in medical diagnosis for several decades. However, there is a scarcity of experts who can interpret these images to diagnose critical illnesses, which can lead to preventable fatalities. This paper introduces a novel Rapid AI Diagnosis at Edge using Ensemble Models for Radiology (RAIDER) designed to leverage the advantages of cross-geolocation meta-learning models. We can generate local machine learning models at individual locations and distribute them across other locations for diagnosing diseases at the edge or on-premises if required before they become worldwide pandemics, significantly enhancing the rapid or near-real-time identification of fast-spreading respiratory diseases through online learning. This novel approach allows for geo-distributed multi-fold model training, harnessing the unique strengths of diverse geographical data sources to improve diagnostic accuracy and speed by leveraging edge computing. Using the existing Convolutional Neural Network (CNN) models and distributed training at the edge, we can enhance the accuracy and cost-effectiveness of diagnosis. The proposed architecture allows for distributed training and independently verified performance metrics on the MIMIC-CXR and COVIDGR chest X-ray datasets with accuracy, sensitivity, specificity, F1-score and AUC of 97.80%, 97.06%, 98.48%, 96.51%, and 0.9739, respectively. Our proposed RAIDER architecture marks the first implementation of a collaborative framework that facilitates seamless interaction across different geographic locations and edge computing, enabling a more effective and efficient response to emerging health threats.

This paper introduces Waste Factor (W), also denoted as Waste Figure (WF) in dB, a promising new metric for quantifying energy efficiency in a wide range of circuits and systems applications, including data centers and Radio Access Networks (RANs). Creating and managing 5G and future 6G networks that are energy-efficient is of paramount importance as the wireless industry evolves to become a major consumer of energy. Also, the networks used to connect data centers and artificial intelligence (AI) computing engines with users for machine learning (ML) applications must become more power efficient. This paper illustrates the limitations of existing energy efficiency metrics that inadequately capture the intricate energy dynamics of RAN components. We show here that W provides a generalized analysis of power utilization and energy waste at both the component and system levels for any source-to-sink communication system. We delineate the methodology for applying W across various network configurations, including multiple-input single-output (MISO), single-input and multiple-output (SIMO), and multiple-input multiple-output (MIMO) systems, and demonstrate the effectiveness of W in identifying energy optimization opportunities. Our findings reveal that W not only offers nuanced insights into the energy consumption of RANs but also facilitates informed decision-making for network design and operational efficiency. Furthermore, we show how W can be integrated with other key performance indicators (KPIs) and key value indicators (KVIs) to guide the development of optimal strategies for enhancing network energy efficiency under different operational conditions. Additionally, we present simulation results for a distributed multi-user MIMO (MU-MIMO) system at 3.5, 17, and 28 GHz, demonstrating overall network power efficiency on a per square kilometer basis, and show how overall W decreases (e.g., energy efficiency increases over the entire network) with an increasing number of base stations and increasing carrier frequency. This paper shows that adopting W as a figure of merit (FoM) can enable the design of more sustainable next-generation wireless communication networks, paving the way for greener and more sustainable, energy-efficient 5G and 6G technologies.

This paper proposes a multi-domain multi-level (MDML) orthogonal frequency division multiplexing (OFDM) optical encryption transmission scheme based on memristor Rulkov neuron chaos. In this scheme, the masking factors generated by the memristor Rulkov neuron chaos are used to encrypt the information of the digital modulation process, so as to improve the anti-malicious attack performance of the system. Among them, the memristor unit constructs a simple discrete map to capture the complex nonlinear neuronal behavior, and the generated masking factors encrypt the data in the digital modulation process. In addition, the proposed scheme introduces the encryption depth control parameters (EDCP), which can make up for the information damage caused by the complexity of encryption. The selection of EDCP can not only encrypt the data, but also change the distribution of the uniformly distributed constellation points position, reduce the average power of the constellation points, and improve the transmission performance of the fiber communication system. A 9.41 Gb/s OFDM signal transmission over 25 km standard single-mode fiber (SSMF) is experimentally demonstrated. The introduction of memristor Rulkov neurons makes the key space reach 10116. The introduction of the EDCP makes the key space expand 1036 times. When the bit error rate (BER) is 10−2, the receiving sensitivity of the EDCP with 0.2, 0.4, 0.8 is 6 dB higher than that of the EDCP with 0.9, 0.3, 0.1. The results show that the encryption scheme can effectively resist illegal attacks and improve the security performance of the system.

Multiple projects are increasingly managed in a decentralized environment, resulting in an inapplicable centralized resource-constrained multi-project scheduling problem (CRCMPSP). A decentralized resource-constrained multi-project scheduling problem (DRCMPSP) is proposed to ensure the timely delivery of products and services within budget. DRCMPSP often faces difficulty in effectively addressing resource conflict. Moreover, activity durations are hard to determine in advance due to various unpredictable events. Therefore, this paper investigates the performance of priority rules (PRs) for global resource allocation of stochastic DRCMPSP. A two-stage heuristic based on PR is then developed to resolve global resource contention under a stochastic environment. 25 PRs from the literature were selected and incorporated into our approach to explore their efficiency on global resource allocation in different global objectives, that is, the expected average project delay (EAPD) and expected total makespan (ETMS). Computational experiments based on the Multi-Project Scheduling Problem LIBrary (MPSPLIB) dataset show that more efficient PRs always favor activities that try to obtain a low delay for each project for EAPD, whereas for ETMS, more efficient PRs tend to favor the longest projects. Furthermore, the hybrid PRs based on more efficient PRs are designed to allocate global resources, and experimental results show that the average percentage of 3 hybrid PRs better than that of their components is 1.14% for EAPD and 0.81% for ETMS, respectively.

At present, the rapid development of big data technology and the Internet of Things has led to the generation of spatiotemporal data streams on a substantial scale. In this context, data holders often opt for cloud storage of their data to mitigate the pressure of local storage and computing overhead. However, this centralized storage mode renders the data susceptible to risks of tampering and leakage due to the absence of physical control over spatiotemporal big data. Therefore, this paper devises a blockchain-based spatiotemporal big data privacy protection scheme. Initially, to ensure data privacy, the scheme adopts a storage mode where the public and side chains are coordinated. Subsequently, the irreversible SHA256 algorithm is utilized to authenticate the access nodes. Finally, AES and RSA hybrid encryption are applied to encrypt the collected spatiotemporal big data, further enhancing the security of private data. The security performance of the system was evaluated by comparing the success rate of attacks on the blockchain under various circumstances. Additionally, the system throughput of this scheme was analyzed through simulation, revealing that the service life of this system extends to a minimum of 13.79 years when 100 nodes are used. In this context, the system throughput achieves 1607 transactions/s, each node hosts 1.2107 blocks, and querying specific spatiotemporal data takes less than 2 seconds.

Shoulder pain and injuries present significant challenges to researchers and clinicians for diagnosing underlying structural changes due to the complexity of the shoulder joint. Ultrasonography has been used for diagnosing shoulder impairment and can non-invasively assess structures and mechanical properties of the shoulder. However, the complexity of the shoulder structures often results in diagnostic difficulties and misdiagnosis. Although deep learning of artificial intelligence has been applied in various biomedical imaging, the adoption of deep learning techniques in the segmentation of musculoskeletal ultrasound images, especially the shoulder, is limited. This study addresses this gap by assessing the effectiveness of 3 deep learning models, U-Net, Mask R-CNN, and DeepLab V3+, for the segmentation of soft tissues from shoulder ultrasound images. We collected 721 images from the shoulder area of 17 healthy adults, including the anterior deltoid, medial deltoid, posterior deltoid, and supraspinatus. We employed a combination of three augmentation methods (elastic transform, horizontal flip, and shift scale rotate) to enhance the dataset. The mixed augmentation strategy resulted in U-Net outperforming Mask R-CNN and DeepLab V3+ with a mean Average Precision (mAP) of 78-81%, a mean Intersection over Union (mIoU) of 81-87%, and Recall and Precision values between 91-94% and 87-91%, respectively. The effective use of deep learning methods could assist clinicians on assessing shoulder structures from ultrasound images.

A miniature single-piston valve flow-matching water-hydraulic pump is described in this paper, which designs with a unique double-face cam mechanism to achieve the reciprocating motion of the piston and a new flow-matching valve to adapt to the requirements of the frequency of flow-matching, and the use of a brushless DC motor to regulate the speed and control the flow of the pump. In this paper, the mechanical structure and working principle of this new type of water-hydraulic pump are described, the mathematical model of the valve flow distribution process is established, the transfer functions of the suction and drain process are obtained, and the relationship between the structural parameters of the distribution valve and the transient output characteristics of this type of water-hydraulic pump is simulated, and the simulation results show that the quality of the valve spool is the key factor affecting the switching speed of the distribution valve. The transient pressure and transient flow rate of the pump are studied experimentally, which reveals the pulsation law of the output characteristics of this single-piston pump, and lays a theoretical foundation for the design of a double-piston pulsation-free hydraulic pump.

A refined bidirectional A $^{\ast }$ algorithm proposes to address the issue of imprecise path planning for unmanned mining vehicles navigating through complex open-pit mining terrains. To ensure the smooth traversal of these vehicles, a gradient factor incorporates into the cost function to circumvent obstacles along the pathway. Additionally, a weighted coefficient introduces into the heuristic function to fine-tune the combination of Euclidean and Manhattan distances, enhancing the accuracy of path distance measurement and ultimately leading to an optimal path. The enhanced algorithm adeptly simultaneously explores from both the starting and end points, significantly reducing search time and improving path planning efficiency. Utilizing the established map of the mining environment, an experiment for unmanned mining vehicle path planning devises, and a simulation test of the global path planning algorithm conducts using the MATLAB platform. The experimental results demonstrate that the refined bidirectional A $^{\ast }$ algorithm exhibits accelerated search speed and superior path planning effectiveness.

A dual-band sleeve monopole antenna with the substrate-integrated coaxial line (SICL) technology is proposed in this paper. Both the radiation structure and the feeding network are modified from the SICL structure. The antenna provides dual-band operation for LoRa (Long Range Radio) applications covering 433 MHz and 868 MHz bands. This antenna operates as a quarter-wavelength monopole with practical omnidirectional radiation performance for each band. The antenna exhibits a gain of 2.0 dBi at 433 MHz with more than 90% radiation efficiency, and a gain of 1.7 dBi at 868 MHz with more than 90% radiation efficiency. The measured −10 dB impedance bandwidths are from $393 \sim 450$ MHz and $826 \sim 931$ MHz, respectively. A good agreement is observed between the measured and the simulated results, which demonstrates that the proposed scheme can be a possible candidate for LoRa applications.

Limited by the capacity and computing ability of platform payload, researchers often face the challenge of balancing lightness and performance of models in synthetic aperture radar (SAR) ship detection, especially for models based on deep learning. Nonetheless, traditional lightweight methods, such as reduced convolutional layers and pruning, can easily lead to missed detections in models. Researchers have introduced knowledge distillation algorithms to address the issue of poor performance of lightweight models. However, the improvement effect of algorithms is limited due to shortcomings such as noise interference in the background and improper distillation strategies, especially for small ship detection with complex backgrounds. Aiming to address the limited performance improvement of distillation algorithms and missing detections of small ships in distillation models, we propose a multiscale feature enhancement and foreground-scene feature distillation algorithm for SAR ship detection. Specifically, in order to improve distillation efficiency, the feature learning distillation module is proposed to improve the quality of distillation knowledge by separating foreground and scene distillation. Then, the ship feature representation enhancement module utilizes a feature map decoupling and attention-based multiscale fusion algorithm to enhance student model's learning of small ship features and reduce missing detection. To validate the performance of the proposed method, we conducted experiments on SAR ship detection dataset (SSDD) and high-resolution SAR images dataset (HRSID) datasets and compared with several advanced methods. The results indicate that the models using our algorithm achieved significant improvements in average precision (AP). For instance, on the SSDD dataset, RetinaNet, Cascade R-CNN, and RepPoints based on ResNet18 achieved AP scores of 95.5%, 95.4%, and 95.9% respectively, surpassing the baseline by 3.9%, 3.1%, and 1.9%.

The Shuttle-Based Storage and Retrieval System (SBS/RS) faces challenges of low efficiency due to the constraints of single outbound or inbound operations. To overcome this limitation, a scheme enabling simultaneous outbound and inbound operations has been proposed. This involved developing a physical model that incorporates costs related to outbound cargo urgency, shelf stability, time, and warehouse busyness. The model was solved using a Genetic Algorithm with priority selection, adaptive operators, and a decay factor (GA-DF). Experimental results, validated across various environments, demonstrate that the proposed GA-DF algorithm achieves 50% higher efficiency compared to the IOSA algorithm when the shelf occupancy rate is 50% and multiple cost environments are considered. Additionally, GA-DF outperforms the Simulated Annealing algorithm, traditional Genetic Algorithm, and IOSA algorithm in optimizing storage and retrieval locations, significantly enhancing system optimization. This provides a crucial reference for optimizing such systems, particularly in dynamic and complex warehousing environments. The GA-DF algorithm’s applicability and advantages have been widely recognized through further validation, highlighting its potential to drive improvements in warehousing system efficiency and optimization strategies.

An interleaved Buck-Boost-Zeta converter with the coupled inductor multiplier cell and zero input current ripple is proposed in this paper, which is suitable for high step-up applications such as the photovoltaic generation system. By integrating the Buck-Boost converter with the Zeta converter and adding the coupling inductor voltage multiplier cell and the auxiliary input inductor, the proposed converter achieves very high voltage gain and very low switch voltage stress. At the same time, very low input current ripple and continuous output current can be achieved, which is friendly to photovoltaic arrays and DC-bus. The leakage inductance energy is absorbed by the passive clamp circuit, the zero-current-switching (ZCS) turn-on of the switch is achieved and the reverse recovery problem of the diode is alleviated. The operation principles and performances of the proposed converter are analyzed in detail. The proposed converter can further improve the voltage gain by adding different voltage multiplier cells and their general structures are given. A prototype with 32V input and 400V output was built to verify the theoretical analysis.

In recent years, deep learning algorithms represented by the Transformer model have gained widespread recognition in the application of time series anomaly detection. Based on the Boosting algorithm, this paper proposes a novel data-driven improved Saxformer framework for multivariate time series anomaly detection. This framework enhances the Transformer model’s ability to capture local feature information and improves both the robustness to noise and the flexibility in practical applications. A method using Boosting classifiers for fine control is designed to improve the Transformer output layer. By outputting text letter by letter, the method aims to balance the model’s creativity and conservatism, suppressing significant fluctuations in predictions and thereby enhancing detection accuracy. Experimental results show that the proposed model outperforms existing mainstream algorithms on multiple evaluation metrics. The F1 score, Precision, and Recall reach 0.882, 0.946, and 0.825, respectively. Compared with the best-performing baseline model, the F1 score is enhanced by 1.4%, demonstrating high accuracy in time series anomaly detection tasks.

In recent years, there have been numerous failures in submarine power cables caused by seawater erosion. Vortex-induced vibration is the main cause of fatigue damage or breakage in submarine power cables. In this paper, model similarity theory was introduced into finite element modeling of submarine power cable for the first time, which dramatically reduced the amount of model calculation. The resonance frequency, amplitude and corresponding flow rate were determined and the stress distribution was analyzed of submarine power cable with different length-to-diameter ratios by acoustics modal and harmonic response. A relationship equation between fatigue life and length-to-diameter ratio was established. The results show that the resonance frequency ranges from 2.651 to 6.633 Hz and the resonance amplitude was 0.64 times the cable diameter when the 110 kV submarine power cable had a length-to-diameter ratio between 70 and 110. The fatigue life was a power function of different length-to-diameter ratios. This research overcomes the difficulties of conducting physical tests and full-scale modeling of vortex-induced vibration of submarine power cable. It provides a viable method and data support for the characterization and fatigue analysis of vortex-induced vibration of submarine power cable.

Due to the distinctive fault characteristics of wind power converters in large-capacity wind farms (WFs) with power converter-based systems, protective relays for the connected transmission lines can malfunction, which poses a serious challenge to the stable and reliable power transmission. This article focuses on addressing this issue by proposing a phase-to-phase fault protection method that utilizes single-end transient information for wind farm transmission lines. Conventional solutions often encounter difficulties such as the negative influence of fault resistance and the uncertainty of remote parameters in single-ended data locations. The proposed method employs time-domain differential equations to accurately solve the fault location problem. It takes into account the equivalent resistance and inductance of the transmission line, as well as the fault path resistance. This helps minimize the adverse effects of fault resistance and parameter uncertainties, making the method insensitive to fault impedance and system parameters. The effectiveness of the proposed scheme is comprehensively evaluated using the Real-Time Lab (RTLAB). Various factors like fault resistances, locations, and distributed capacitance are considered. The results clearly demonstrate that the proposed algorithm not only operates at high speed but also exhibits remarkable reliability, offering a promising solution for enhancing the protection of wind farm transmission lines.

Assessing plantar pressure is crucial for fabricating diabetic insoles and preventing diabetic foot ulcers (DFUs), which are caused by increased plantar pressure. However, the commonly used methods for assessing plantar pressure distribution involve professional sensor-based equipment and expertise, which are costly and time-consuming. Given the qualitative association between ink footprint images and plantar pressure, this study proposes using the footprint images to predict the quantitative values of dynamic plantar pressure in barefoot and 4 different insole conditions (including Nora Lunalastik EVA, Nora Lunalight A fresh, Pe-Lite, and PORON® Medical 4708) based on a multilayer perceptron (MLP) neural network model. To provide more precise insole material recommendations for specific foot regions for better plantar pressure distribution, the plantar of the foot is divided into 5 regions: the toes, metatarsal heads, medial midfoot, lateral midfoot, and heel. Patch-based MLP with localization embedding is introduced to learn the correspondence between ink density and plantar pressure information. Ground-truth data collected from 52 diabetes patients is constructed as a dataset named diabetes-footprint-to-pressure and used to train and validate the model. The mean absolute error (MAE) of the models for the barefoot and 4 insole conditions is 5.51% (33.06 kPa), 3.99% (23.94 kPa), 4.85% (29.10 kPa), 4.25% (25.50 kPa), and 3.57% (21.42 kPa) of the sensing range, respectively. Compared to traditional methods for plantar pressure assessment, this approach streamlines the process of acquiring the overall and regional dynamic plantar pressure with barefoot and 4 different insole materials. Clinicians can quickly provide recommendations on the type of insole material for individual patients.

Wind turbine blades are susceptible to various types of damage caused by environmental factors, external forces, and other influences. Achieving efficient and accurate detection of damage defects in wind turbine blades is essential to extend the service life of wind turbines. To this end, this paper proposes a cascaded adaptive hybrid attention network (CAHAN). Firstly, a high-resolution wind turbine blade damage defects image dataset is constructed using the dual aggregation transformation (DAT) model and the contrast enhancement algorithm. Secondly, a global contextual attention pyramid (GCAP) module is proposed to effectively integrate global and local feature information. Then, a cascaded adaptive hybrid attention (CAHA) module is developed to design a PANet feature fusion network, which can effectively filter out redundant global feature information. In addition, the dynamic head (DH) modules covering scale, space, and task awareness are applied to wind turbine blade defects detection, greatly enhancing the model’s capability to detect blade damage defects. Finally, the proposed method is tested using wind turbine blade defect images collected from open web downloads. The results show that the CAHAN model achieves an overall detection accuracy of 97.9% and boasts a fast detection speed of 149.25 frames per second (FPS), which has superior overall performance compared to other object detection models. This method is useful for defect detection of wind turbine blade damage.

Power loads usually have multiple cycles, and the traditional forecasting methods only use the historical loads under various types of cycles together as input features to construct the forecasting model, ignoring the deep features under multiple cycles. The simultaneous inputs will lead to the cycles overlapping and influencing each other, which will be difficult to deal with when building the model. Therefore, this paper proposes a GRO-Bagging day-ahead power curve forecasting model based on multi-cycle feature extraction. First, the multi-periodicity of the power load is analyzed. The one-dimensional time series is converted to two-dimensional according to the multiple cycles of power load. Then, the multi-periodic feature extraction is performed by a multi-size convolutional feature extraction layer with parallel selected based on the data characteristics and modeling mechanism, and Bootstrap Aggregating (Bagging) method is used to construct different prediction models for the datasets containing different periodic features; finally, Gold Rush Optimization (GRO) algorithm is introduced and improved by using the Tent chaotic mapping and elite strategy, and the improved optimization algorithm is used for the weight optimization of the model, the error feedback mechanism is introduced to achieve the weight dynamic Updating. To prove the superiority of the proposed model, a series of comparison experiments and ablation experiments are carried out on real datasets, and the results show that the proposed method has higher prediction accuracy, and prove that the multi-period feature extraction and dynamic weighting methods have a positive and active effect on load prediction.

In the context of implementing carbon quota trading policies and the rapid development of e-commerce, blockchain technology, with its traceable and tamper-proof characteristics, reduces product costs and corporate financing costs by promoting information sharing. It effectively solves the trust problem caused by information asymmetry. This study considers a supply chain consisting of a manufacturer with financial constraints and a retailer with dual-channel sales. We construct decision models for bank financing, zero-interest early payment financing, and in-house factoring financing under both benchmark conditions and blockchain technology conditions. The study explores the impact of blockchain technology on the optimal emission reduction and financing decisions of the supply chain under these three methods. The results show that regardless of whether carbon emissions exceed standards, there is a linear relationship between the manufacturer’s profit and carbon emission reduction efficiency. Additionally, there is a U-shaped relationship between the retailer’s profit and online market share. There exists a reasonable range for carbon reduction efficiency and online market share such that introducing blockchain technology can effectively increase profits for all parties involved in the green supply chain. Zero-interest early payment financing and in-house factoring financing have critical points related to production cost, while bank financing and in-house factoring financing have critical points related to carbon efficiency. As the impact of blockchain technology on carbon price increases, decision-makers will shift from bank financing to in-house factoring financing.

This paper studies the time-dependent cold chain logistics vehicle routing problem considering both traffic congestion and carbon emissions. A cold chain logistics model with time-dependent green vehicle paths with time windows (TDGVRPTW) was developed to fulfil the demands of green logistics and to take comprehensive account of consideration should be given to factors such as road congestion and carbon emissions. The objective of the model is to minimise total costs, which include carbon emission costs, penalty costs, fuel consumption costs, fixed costs, damage costs and refrigeration costs. Two-phase hybrid search algorithm was developed to solve this model. During the initial stage of the algorithm, a dual-population ant colony optimization (DACO) algorithm sharing the optimal individual is employed to optimize the distribution route of the vehicle. During the second phase, an adaptive golden section search (AGSS) algorithm is used to optimise the departure time of the vehicle from the distribution centre to avoid traffic congestion time periods. To validate the effectiveness of the suggested two-phase hybrid search algorithm, it is applied to the improved Solomon benchmark test set. The experimental findings demonstrate that the two-phase hybrid search algorithm can reasonably plan the driving routes and departure times for each vehicle, effectively avoiding peak traffic congestion periods in the city, and reducing the overall delivery cost.

In the digital age, the rapid growth of web information has made it increasingly challenging for individuals and organizations to effectively explore and extract valuable insights from the vast amount of information available. This paper presents a novel approach to automated web text summarization that combines advanced natural language processing techniques with recent breakthroughs in deep learning. we propose a dual-faceted technique that leverages extensive pre-training on a broad dataset outside the domain, followed by a unique refined tuning process. We introduce a carefully curated dataset that captures the heterogeneous nature of web articles and propose an innovative pre-training and tuning approach that establishes a new state-of-the-art in news summarization. Through extensive experiments and rigorous comparisons against existing models, we demonstrate the superiority of our method, particularly highlighting the crucial role of the refined tuning process in achieving these results. Through rigorous experimentation against state-of-the-art models, we demonstrate the superior performance of our approach, highlighting the significance of their refined tuning process in achieving these results.

Computed tomography (CT) imaging is widely applied to identify tumors and internal injuries in the body. CT image retargeting involves reducing the size of less important areas (like normal organs) either horizontally or vertically, while maintaining the size of key areas (such as affected organs) in the CT image. This retargeting process can significantly improve the way CT images are displayed, which can be advantageous for further educational tasks, e.g., optimally displaying CT images to medical students. This paper focuses on effectively merging multiple channels of perceptual visual features for retargeting CT images with intricate spatial layouts. The key objective is to develop an active learning model that identifies where human attention is directed in a scenery. To capture both semantically and visually important elements in each CT image, we use the BING objectness descriptor, which quickly and accurately localize objects or their parts across multiple scales. Herein, multi-channel low-level features describe each object-aware patch. In an effort to characterize how humans perceive important scenic elements, we propose a locality-preserved and interactive active learning (LIAL) approach. This method sequentially generates gaze shift paths (GSP) for each CT image. Due to the high dimensionality of such GSP feature, a distribution-preserved feature selection (DPFS) is designed to acquire a few highly discriminative GSP features. Last, these refined GSP features are combined by a Gaussian mixture model (GMM) to retarget CT images. Extensive empirical studies demonstrate the power of the designed method.

Recent advancements in sequence-structure co-design have demonstrated the efficacy of integrating pre-trained language models (PLMs) with Graph Neural Network (GNN)-based modules for antigen-specific antibody design. A significant advantage of PLMs is their ability to directly predict the Complementarity-Determining Region (CDR), or reduce reliance on structure data by providing a good initialization for the downstream GNN modules. However, while the performance scaling law suggests that training larger models can lead to improved performance, scaling up PLMs for downstream modules faces limitations due to constraints on the size of datasets used for training and fine-tuning. To address the limitations of larger PLMs trained on constrained datasets, we introduce AbGraftBERT. Initially, inspired by the depth up-scaling approach, we provide a Self-Grafting strategy by dividing a PLM into two components: the root layers (first 10 layers, with parameters frozen) and the crown layers (last 10 layers, with parameters frozen), then interconnected these layers through a full-connection layer inspired by the model reprogramming strategy. Furthermore, we introduce Cross-Embedding Attentional Perturbation (CEAP) which perturbs the embeddings passed to the downstream GNN, enabling the model to enhance its focus on critical amino acids and improve robustness. Our experiments demonstrate that our model surpasses current state-of-the-art models in antibody-related tasks, including sequence and structure prediction, antigen-binding antibody design and optimization of Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) neutralization. The source code of the AbGraftBERT model is available at https://github.com/azusakou/AbGraftBERT

This study proposes a novel approach to address the challenges posed by the increasing number of renewable sources in power grids, particularly focusing on networked microgrids. The proposed method integrates digital twin technology and blockchain technology to provide power support in networked microgrids. The method comprises three main components: 1) Blockchain Network: A blockchain network is established, utilizing a modified proof of stake (PoS) consensus mechanism. This network is responsible for maintaining hourly records and tracking system changes, including faults. Fault events are recorded in real-time, and a bidding mechanism is employed to determine which microgrids interconnect and provide power support to mitigate power imbalances. 2) Digital Twins: Digital twins are created for each distributed generation and energy storage unit within the microgrids. These digital twins play a crucial role in the proposed PoS consensus mechanism, providing an additional layer of validation for transaction records. 3) Long short-term memory (LSTM)-based Controller: An LSTM-based controller is implemented to ensure stable operation of the microgrids, even during fault events and interconnections with other microgrids. The controller is trained to maintain stable voltage levels within the system. The digital twins and LSTM-based controllers are optimized using Bayesian Optimization to determine the best parameters and hyperparameters. Simulation results demonstrate that the proposed method effectively records hourly system details and fault events. Additionally, power hardware-in-the-loop simulations confirm that the LSTM-based controller can maintain stable voltage levels and outperforms other controllers in terms of performance.

Non-uniform low-light images, characterized by complex lighting conditions, pose a significant challenge in image restoration and enhancement, particularly under backlit scenarios where the high-brightness background obscures the foreground information. Existing methods struggle with the diverse illumination factors across different scenes. To overcome this, we introduce a novel deep reconstruction network designed specifically for enhancing non-uniform low-light images. This network leverages the feature representation of uniform natural images, employing the dark channel of the absorption light scattering model (ALSM) to estimate and enhance facial features. Furthermore, the reconstruction process is optimized through the posterior constraint of uniform image characteristics, leading to superior detail enhancement. Our approach has been rigorously tested on both synthetic and real-world images, demonstrating its effectiveness in addressing the complexities of non-uniform low-light image enhancement. The results illustrate a notable enhancement, boosting PSNR by 3.37dB and SSIM by 0.0579 compared to recent methods, with a particular focus on enhancing facial feature details. The code link is https://github.com/zbysygdsghh/face-relight.

With the advancement of technology in vehicle-road collaboration and autonomous driving, new commercial applications have surfaced. These include autonomous ride-hailing vehicles and unmanned delivery vehicles. As a result of the challenges presented by commercial applications, dispatching systems are moving towards being maintenance-free, centralized, multitasking, and real-time. Yet, most existing dispatching systems have been designed for single-task purposes and cannot tackle multitasking issues. Moreover, traditional optimization algorithms make it difficult to achieve timeliness in real-time changing traffic conditions. Therefore, this paper innovatively proposes a task allocation method based on Multi-Agent Reinforcement Learning (MARL). Firstly, this study introduces a classification model of task relationships through the binary assumption model of geographical areas and vehicles. Secondly, the study matches the classification model’s task cost state transition process with the Markov Decision Process, constructing a Multi-Agent Reinforcement Learning framework. Finally, the study constructs a simulation environment suitable for reinforcement learning based on Simulation of Urban Mobility (SUMO). Simulation results indicate that the task allocation system based on MARL can effectively improve the system’s overall efficiency by determining the order of task allocation and the matching relationships between tasks.

Accurate carbon price prediction can help save energy and reduce emissions worldwide. Thus, this paper proposes a model that combines swarm intelligence algorithms with deep learning to predict carbon prices. In this model, we collect news related to carbon trading, construct a dictionary of carbon financial sentiment, and determine the emotional value of the carbon news. Secondly, The Harris Hawks Optimization (HHO) algorithm is improved by updating the escape energy and introducing the inertia weight. Then, the LSTM is optimized using the improved Harris Hawks Optimization (IHHO) algorithm. Finally, technical and emotional data on carbon price as multiple source input values are integrated, and the MS-IHHO-LSTM prediction model is established. The results show that the MAPE of IHHO-LSTM is 1.89%, 30.48%, and 10.30% better than that of HHO-LSTM in Hubei, Shanghai, and Shenzhen Carbon Exchanges, respectively. Similarly, MS-IHHO-LSTM showed a lower MAPE than IHHO-LSTM by 27.79%, 29.82%, and 6.33% in the corresponding regions. The results of the experiment indicate that: 1) Using IHHO to optimize LSTM hyperparameters can avoid falling into local optimal and improve prediction accuracy; 2) Incorporating emotional values can further enhance the model’s performance. The MS-IHHO-LSTM prediction model facilitates low-carbon investment, technological innovation, and green production, enabling enterprises to support environmental sustainability.