Occluded person Re-identification (Re-ID) is to identify a particular person when the person’s body parts are occluded. However, challenges remain in enhancing effective information representation and suppressing background clutter when considering occlusion scenes. In this paper, we propose a novel Attention Map-Driven Network (AMD-Net) for occluded person Re-ID. In AMD-Net, human parsing labels are introduced to supervise the generation of partial attention maps, while we suggest a Spatial-frequency Interaction Module (SIM) to complement the higher-order semantic information from the frequency domain. Furthermore, we propose a Taylor-inspired Feature Filter (TFF) for mitigating background disturbance and extracting fine-grained features. Moreover, we also design a part-soft triplet loss, which is robust to non-discriminative body partial features. Experimental results on Occluded-Duke, Occluded-Reid, Market-1501, and Duke-MTMC datasets show that our method outperforms existing state-of-the-art methods. The code is available at: https://github.com/ISCLab-Bistu/SA-ReID.

In this paper, based on thin film integrated passive device (TF-IPD) technology, compact ultra-wideband (UWB) bandpass filter (BPF) and dual-band bandpass filter (DBBPF) with equal inductance configuration are proposed. The configuration of equal inductance in a circuit contributes to reducing variables, simplifying the physical structure, and shortening the debugging time in the TF-IPD design process. Despite all the inductors being equal in value, wide passband can still be achieved. The measured results show that the center frequency of UBW BPF is 11.43 GHz, the 3-dB fractional bandwidth (FBW) is 133% (3.87 GHz-18.99 GHz). And the 3-dB FBWs of DBBPF centered at 6.5 GHz and 16.2 GHz are 92% (3.51-9.49 GHz) and 35% (13.33-18.98 GHz), respectively. The stopband of UBW BPF and DBBPF can be extended to 47.7 GHz and 44.8 GHz in sizes of 321 um × 856 um and 235 um × 846 um, respectively (excluding measuring pads).

Non-conventional renewable energy sources integration into distribution systems, data science, and enabling technological infrastructures present formidable challenges in transforming distribution systems, particularly in accommodating active demand. With escalating energy demand emphasizing the need for dependable tracking and predictive methodologies. These methodologies are essential for managing Distributed Energy Resources (DERs) and digital infrastructure. Effectively monitoring active demand requires a comprehensive understanding of the transactional system concept, encompassing digital infrastructure, and decentralized demand. Despite the increasing prominence of metaheuristic techniques in demand response integration, existing literature predominantly focuses on specific techniques rather than providing a characterization of dynamic transaction integration for active demand. Technological advancements, such as smart meters and communication systems, propel the evolution of the demand concept from rudimentary consumption measurement to active consumer involvement. This paper reviews the evolutionary trajectories of essential concepts in the energy sector, namely active demand, DERs, and transactive systems. Through in-depth analysis, Simultaneously, the paper systematically examines prevalent metaheuristic techniques in the literature, specifically focusing on their role in integrating and predicting the behaviors of active demand and DERs. The paper presents a methodology that, if utilized as a roadmap, would facilitate the evidence of stages required for the integration of DERs .

This letter introduces an innovative approach for minimizing energy consumption in multi-UAV (Unmanned Aerial Vehicles) networks using Deep Reinforcement Learning (DRL), with a focus on optimizing the Age of Information (AoI) in disaster environments. We propose a hierarchical UAV deployment strategy that facilitates cooperative trajectory planning, ensuring timely data collection and transmission while minimizing energy consumption. By formulating the inter-UAV network path planning problem as a Markov Decision Process (MDP), we apply a Deep Q-Network (DQN) strategy to enable real-time decision-making that accounts for dynamic environmental changes, obstacles, and UAV battery constraints. Our extensive simulation results, conducted in both rural and urban scenarios, demonstrate the effectiveness of employing a memory access approach within the DQN framework, significantly reducing energy consumption up to 33.25\% in rural settings and 74.20\% in urban environments compared to non-memory approaches. By integrating AoI considerations with energy-efficient UAV control, this work offers a robust solution for maintaining fresh data in critical applications, such as disaster response, where ground-based communication infrastructures are compromised. The use of replay memory approach, particularly the online history approach, proves crucial in adapting to changing conditions and optimizing UAV operations for both data freshness and energy consumption.

The type parameter helps in the scattering mechanism analysis and scattering center identification. However, current approximate solution methods based on spectral estimation are noise-sensitive and have poor accuracy. In this letter, a high-precision approach for the type parameter based on the alternating direction method of multipliers (ADMM) is proposed. We use logarithmic transformation to separate the type parameter and the amplitude from the coupling term, and initially obtain the closed-form solution of the type parameter. Then, the regularization term of l2-norm is used for denoising. Finally, a joint optimization model based on ADMM is constructed to effectively estimate the type parameter. Simulation results confirm the high accuracy of the proposed approach.

The Static Frequency Converter (SFC) in a pumped storage power plant often causes harmonic problems in the dragging processes, which may lead to the false operation of automatic devices in the power station, and even damage to the power equipment. These harmonics caused by SFC contain both characteristic and non-characteristic degrees, and the components are more complex. Hence, the existing harmonic suppression methods using APF or PPF compensation all have some problems. The SFC starting control strategy based on Linear Active Disturbance Rejection Control (LADRC) is proposed in this paper to reduce the non-characteristic subharmonics. Firstly, the factors affecting the non-characteristic harmonic content are analyzed, and a mathematical model of the conventional SFC starting transfer function is established. On this basis, the LADRC controller is used as the speed loop of SFC starting to replace the PI controller. The stability is judged by drawing the Bode plot and the zero-pole plot. Finally, A Matlab/Simulink simulation model of LADRC-based SFC starting is established in a pumped storage power plant with actual parameters, and simulation accurate measurements verify the effectiveness of the proposed control strategy for non-characteristic harmonic current suppression.

Carbon emissions limit the output demand of traditional fuel-generating units, significantly affecting the electric power and energy balance scheduling mechanisms of modern power systems. To address this, we first analyzed the short-term electric power and energy balance from a dual perspective, along with the electro-carbon coupling mechanism of dynamic dispatching on the source-load side. Based on this analysis, we constructed a framework for low-carbon bilateral demand response (LCBDR). Subsequently, based on the carbon emission flow theory, the carbon emission index information of the demand response system was obtained. Therefore, the short-term electric power and energy balance optimal scheduling model for LCBDR is established to achieve a low-carbon economy. In this study, the enhanced decision tree classifier (EDTC) algorithm is used to predict the electricity consumption behavior of transfer load (TL) user load, and an improved ‘ ε-greedy’ strategy particle swarm optimization (PSO) algorithm is proposed to realize the coordinated optimization of daily operational cost and daily carbon emission of the regional power grid. Finally, by setting up three different perspectives of the simulation examples, the proposed method can effectively realize the regional power grid’s low-carbon, economical, and efficient operation. The feasibility of the model can be verified via the IEEE-30 bus system.

Nikel Cobalt layered double hydroxides (NiCo LDHs) have emerged as ideal electrode materials for supercapacitor due to their high specific surface area and excellent cycling stability. Morphology control plays a unique role in regulating the performance of NiCo LDHs, but there are rare reports to regulate the morphology during energy storage. Herein, the morphology of NiCo-LDHs electrode is optimized for enhancing energy storage by simple activation process with different t concentrations of the electrolyte. During the activation process, electrochemical morphology reconstructed occurs on the electrode surface. With2 M KOH electrolyte the NiCo-LDH electrode transforms from nanosheets to nanoflower, which aids in reducing the distance of ion transport. The reconstructed NiCo-LDH (NiCo-LDH-2) exhibits an ultra-high specific capacitance of 5428 F g-1 at a current density of 1 A g-1, outperforming most of NiCo LDHs. Even at a high current density of 10 A g-1, the capacitance retention rate remains above 77.6% after 1000 charge-discharge cycles. The strategy proposed in the study, which involves concentration-controlled morphology optimization for energy storage enhancement, holds great practical significance for the field of supercapacitors.

For telecom operators, customer service is integral to their business. Traditional service systems, responsible for managing large amounts of data and complex knowledge bases, need more time retrieval processes and a lack of precision, hindering their ability to respond quickly to customer requests. To address these issues, this paper uses the LangChain programming framework to create a customized Large Language Model (LLM) specifically for the customer service context of telecom operators. It also uses reinforcement learning to improve the performance of the models and reduce the production of incorrect information. Experimental results show that the acceptance of our model’s recommended knowledge has increased from 15% to 70%, confirming its reliable operation in resource-constrained environments.

Edge computing has been proved an efficient approach to provisioning computation offloading service to vehicles on road through Road-Side Units (RSUs). However, the traffic volume on road is highly dynamic, while RSU-based edge servers are static in terms of geographical location and computation capacity. To address this problem, this paper proposes a mobile edge server placement strategy using cruising UAVs along the roads based on the genetic algorithm. We first build a mathematical model to characterize the deployment cost of these UAV-mounted servers and their routes. Next, we design a heuristic UAV-mounted edge server deployment scheme based on K-medoid clustering and genetic algorithms. Experimental results verify that our proposed UAV deployment scheme satisfies the offloading demand of IoV nodes while reducing the total deployment cost by 17.05% to 48.94% compared with existing popular approaches.

In order to maximize the promotion effect of renewable energy policies, this paper proposes a capacity allocation optimization method of wind power generation, solar power and energy storage in power grid planning under different policy objectives. Firstly, based on the policy quantification, grey relation analysis is used to calculate the correlation degree of the policy indicators on the planning capacity of renewable energy. Further, a multi-objective capacity estimation model is comprehensively presented. Some highly correlated policy indicators are transformed into the special constraints. And the economy and the stability of the power grid are integrated as the objective function. Meanwhile, the carbon trading and punishment for wind power and solar power abandonment is considered. Finally, the proposed model is solved by NSGA-Ⅱ-PSO algorithm. The novelty of the algorithm is that the crossover operation of NSGA-Ⅱ is replaced by the position updating of particle swarm. The calculation result of the case study can effectively evaluate the optimal planning capacity of renewable energy under different policies, while ensure the economic and the stability of the power system. The paper can provide the reasonable basis and the valid analytical method for the policy formulation and the renewable energy development.

Automatic Violence Detection and Classification (AVDC) with deep learning has garnered significant attention in computer vision research. This paper presents a novel approach to combining a custom Deep Convolutional Neural Network (DCNN) with a Gated Recurrent Unit (GRU) in developing a new AVDC model called BrutNet. Specifically, we develop a time-distributed DCNN (TD-DCNN) to generate a compact 2D representation with 512 spatial features per frame from a set of equally-spaced frames of dimension 160×90 in short video segments. Further to leverage the temporal information, a GRU layer is utilised, generating a condensed 1D vector that enables binary classification of violent or non-violent content through multiple dense layers. Overfitting is addressed by incorporating dropout layers with a rate of 0.5, while the hidden and output layers employ rectified linear unit (ReLU) and sigmoid activations, respectively. The model is trained on the NVIDIA Tesla K80 GPU through Google Colab, demonstrating superior performance compared to existing models across various video datasets, including hockey fights, movie fights, AVD, and RWF-2000. Notably, our model stands out by requiring only 3.416 million parameters and achieving impressive test accuracies of 97.62%, 100%, 97.22%, and 86.43% on the respective datasets. Thus, BrutNet exhibits the potential to emerge as a highly efficient and robust AVDC model in support of greater public safety, content moderation and censorship, computer-aided investigations, and law enforcement.

Targets with rotating components generate micro-motion (MM) modulation effect in addition to the main body. Extracting MM parameters is challenging due to interference from the target’s main body, necessitating the separation of modulation signals. This letter proposes a robust complex local mean decomposition (RCLMD) method with self-adaptive sifting stopping, aiming at the problem of component redundancy due to multiple iterations during break and the loss of modulation components during the separation process. The proposed method sets the objective function and self-adaptive stopping criterion, combined with the modulation signal characteristics, enhancing the accuracy and efficiency of MM component extraction. Simulation experiments indicate that at a low signal-to-noise ratio (SNR) of 3 dB, the separation effect of RCLMD is still 14.72\% higher than that of the conventional complex local mean decomposition (CLMD) method, and the separation efficiency is improved by 54.92\%. Furthermore, the measured radar signals verify the effectiveness of the proposed method in real scenarios.

Power system unreliability tracing model allocates the system’s reliability index to individual components, identifying potential weaknesses. This study expands its scope by considering the impact of storage resources. Unreliable factors leading to load shedding are categorized into two groups: objective factors inherent to the component and insufficient storage resources. The latter requires a retrospective analysis of other components that caused unreliability previously. When allocating responsibility for load shedding at a certain time, it begins by allocating it among components based on differences between fixed expected output and actual supply. Expected output insufficiency is considered the unreliable factor. This insufficiency due to insufficient storage resources is then decomposed into segments, each caused by excessive output in earlier instances of the same component. The expected output excess is attributed to the expected output insufficiency of other components in previous times, for which responsibility has been allocated to each component. Consequently, the expected output insufficiency at a particular time can be traced back based on a temporal recursive model, with the load shedding further allocated to components before that time. Case studies based on several systems demonstrate that the proposed model’s allocation results are reasonable and more accurate than the traditional model.

Welcome to this Special Issue of Wiley’s Healthcare Technology Letters (HTL) journal dedicated to the 2023 edition of the Augmented Environments for Computer-Assisted Interventions (AE-CAI), Computer Assisted and Robotic Endoscopy (CARE), and Context-aware Operating Theatres (OR 2.0) joint workshop. We are pleased to present the proceedings of this exciting scientific gathering held in conjunction with the Medical Image Computing and Computer-Assisted Interventions (MICCAI) conference on October 8th, 2023 in Vancouver, British Columbia, Canada. We hope that you will enjoy reading this Special Issue and we look forward to your continuing support and participation in future editions of the AE-CAI, CARE and OR 2.0 workshops. Their continued success demands our ongoing commitment and support, and we hope to welcome you all to the next edition of the workshop at MICCAI 2024 in Marrakesh, Marocco.

In this letter, we focus on the problem of direct position determination (DPD) for moving targets. Compared with the traditional two-step localization methods, the DPD methods is more robust at low signal-to-noise ratio (SNR). However, to guarantee the optimal location results the computational complexity of DPD with grid search is too high, especially for moving targets. Therefore, we propose a new DPD method for moving targets with low computational complexity. First, using a proposed cost function, we obtain the position information from the received array signals directly. Second, we use the method of successional difference which averages the difference result of position over time to extract the velocity. It avoids the multi-dimensional parameters grid search and reduces the the computational complexity greatly. Simulation results demonstrate that the proposed algorithm outperforms other methods in a wide range of scenarios, especially at low SNR.

To solve the problem of negative-sequence and reactive power in electrified railway, this paper proposes a comprehensive compensation method based on V/v transformers and electromagnetic single-phase var compensator (ESVC) for co-phase power supply. First, based on the principles of static var generator (SVG) and single-phase rotary phase shifting transformer (SRPST), the topology and model of ESVC have been established, and the compensation mechanism has been analyzed. Second, the topological structure and mathematical model of co-phase power comprehensive compensation device (CPCD) are put forward based on the principles of V/v transformers and ESVC, and the operation mode of CPCD in multiple scenarios is designed. Then the strategy of CPCD with double closed loop control is analyzed: corresponding compensation mode is selected according to negative-sequence unbalance degree and expected power factor value. In this strategy, the compensated current output by ESVC is taken as the quantity of external loop control, and the rotation angle of ESVC is taken as the quantity of internal loop control to realize double-closed loop control of CPCD. Finally, the comprehensive CPCD compensation model is built on the simulation platform and validated the accuracy of the mathematical model and the effectiveness of the control strategy.

Frequency response analysis is widely used for the offline diagnosis of winding deformations in power transformers. To apply it to a working transformer, the magnitude of signals of the response current, which is of the order of microamperes, needs to be measured by using Rogowski coil sensors against the load current of the order of thousands of amperes. The saturation of the power frequency magnetic field in current sensors must be inhibited to ensure the accurate measurement of response currents with such a small magnitude. The authors of this paper propose a method to suppress the power frequency magnetic field by using a sensing system involving a special connection of three-phase current sensors based on the rule that the sum of the three-phase power frequency load currents of the transformer is close to zero. Each sensor consists of two secondary-side coils: a measuring coil and an anti-saturation coil. The anti-saturation coils are connected in parallel with one anothes through small inductors to eliminate the power frequency magnetic field in the cores of the sensors. We use theoretical analysis to derive the solution to this system. The results of experiments to verify the proposed method showed that it enables the sensors to function with a transformer carrying a load current of 2333 A, while incurring a relative error in the response current that is smaller than 2%.