Search

Abstract Underwater object detection is an important application of underwater vehicles. However, traditional underwater object detection algorithms have several shortcomings in underwater settings. These include imprecise feature extraction, slow detection speeds, and a lack of robustness. To address these shortcomings, a novel approach based on YOLOv5‐MH (Multi‐Head) is proposed in this paper. Firstly, an image enhancement technique is utilised. This technique uses adaptive linear mapping to adjust contrast and improve the quality of underwater images. Secondly, the C2f module for feature extraction is employed to enable more effective capture of object characteristics. Subsequently, the multi‐head self‐attention and coordinate attention are integrated into the network's backbone. This integration increases the attention given to input data, thereby enhancing the network's performance in handling complex tasks. Furthermore, a bidirectional feature pyramid is implemented to adeptly handle objects of varying scales and sizes, and elevate model performance. Finally, through comprehensive testing on the 2018 URPC dataset and deep plastic dataset, this method demonstrates superior performance. This performance is compared to the original YOLOv5 and other similar networks. It holds immense promise for practical applications across a wide spectrum of underwater tasks.

Abstract In this paper, we improve networks’ spectral efficiency (SE), extend networks’ lifetime, and maximize networks’ energy efficiency (EE) of two-way full-duplex (FD) relay networks. Firstly, to improve networks’ SE and to extend networks’ lifetime simultaneously, we design a two-way FD relay transmission strategy with simultaneous wireless information and power transfer and direct links (DLs). The designed transmission strategy can complete a bidirectional communication in only one time slot with the exists of DLs and the energy-constrained relay node. With the designed transmission strategy, we further give the characteristics of relay amplification factor, the analysis of the designed transmission strategy, and the EE analysis of traditional half-duplex two-way amplify-and-forward relaying. Secondly, to maximize networks’ EE, we present both the EE maximization problems and analyses of the designed transmission strategy with equal power allocation and optimal power allocation. To solve the EE maximization problems, we further propose the alternating optimal algorithm and give complexity analysis of the algorithm. Simulations show that our designed transmission strategy can improve the SE and EE of the networks.

The purpose of this work is to investigate the function of SNHG1, a long non-coding RNA implicated in disease progression, apoptosis, and proliferation, in order to solve the problem of hypoxic-ischemic encephalopathy (HIE) in newborn care. We investigated the impact of overexpressing SNHG1 on hypoxia-induced apoptosis and studied its expression in BV2 microglial cells under hypoxic circumstances. As a result of modifying YY1 expression, SNHG1′s overexpression prevents apoptosis, as our data demonstrate that it is considerably downregulated under hypoxia. We demonstrate that SNHG1 might potentially reduce microglial ischemia-reperfusion damage by using sophisticated nanoengineering drug delivery technologies to target it. This provides encouraging information for the therapy of ischemic epilepsy.

In order to improve the internal heat dissipation efficiency of the proton exchange membrane fuel cell, this paper designs and builds a set of fuel cell heat dissipation system experimental bench with a heat dissipation capacity of 15 kW based on the separated heat pipe technology, and investigates the heat dissipation characteristics of the HFE-7100 and ethylene glycol aqueous cooling medium under different condenser processes and different ambient temperatures. The results show that at an ambient temperature of 35 °C, under the same condenser flow arrangement, the two-phase heat dissipation efficiency of HFE-7100 is higher than that of glycol aqueous solution liquid cooling, and its heat dissipation and EER (Energy Efficiency Ratio) are increased by 81.2 %∼98.8 % and 68.2 %∼86.6 %, respectively. The best effect is achieved when the condenser is 3-process, with two-phase heat dissipation of 14.1 kW and EER of 20.5 kW/kW, which are 82.8 % and 68.9 % higher than liquid cooling, respectively. As the ambient temperature rises, the two-phase heat dissipation effect is further improved, and at 50 °C ambient temperature, the heat dissipation is 10.44 kW, which is 145 % higher than that of liquid cooling, and the EER is 15 kW/kW, which is 132 % higher.

The rudder steering control of high-speed unmanned surface vessels(USVs) must simultaneously satisfy the requirements of a short adjustment time and small overshoot. To satisfy the parameter tuning requirements for rudder steering proportional integral derivative(PID) control of USVs, a parameter tuning method based on the hybrid mean center opposition-based learning particle swarm optimization(HCOPSO) algorithm was devised in this study. The HCOPSO algorithm was used to optimize the parameters of the PID controller, and this prevented the optimization process from becoming trapped in local optimal solutions. The PID controller parameter tuning effects of the particle swarm optimization(PSO), linear decreasing inertia weight particle swarm optimization(LDIWPSO), and HCOPSO algorithms were compared and studied. The results indicate that the USV rudder PID controller with the HCOPSO algorithm has the best control effect. Compared with those of PSO and LDIWPSO, the adjustment time is reduced by 22% and 15%, the overshoot is reduced by 89% and 74%, and the number of iterations is reduced by 40% and 30%, respectively. Using the developed Jiuhang 750 USV, a marine environment test was performed. The test results indicate that the proposed method is effective for the rudder steering control of small high-speed USVs.

Path planning for autonomous underwater vehicles (AUVs) is a key research focus in the marine domain, requiring consideration of the underwater environment’s complexity and the efficiency of the planning algorithms. Firstly, a variety of strategies such as the memory function are integrated into the artificial jellyfish search algorithm (JS) to improve its convergence accuracy, and the improved artificial jellyfish search algorithm (IJS) is obtained. Secondly, this paper establishes a good objective function including the ocean current disturbance model, which helps the IJS algorithm better plan the paths to avoid obstacles and strong side currents. Furthermore, the optimal smoothed paths are obtained by using a cubic spline midpoint interpolation method. Finally, multiple simulation experiments are performed on the multi-obstacle ocean current model with realistic terrain data. The comparison results show that the IJS algorithm with a short running time has the optimal time cost and ocean current penalty cost for the planned path. In addition, the IJS algorithm is also shown to be adaptable in the field of multi-AUV movements.

Task assignment is of paramount importance in multi-AUV systems, particularly in applications such as bridge inspection where task execution is direction-specific. In such scenarios, the underactuation of AUVs is a critical factor that cannot be ignored. Therefore, it is essential to consider the AUV’s kinematic model comprehensively to ensure minimal energy consumption during task execution. In this paper, we introduce an improved Reeds–Shepp algorithm in conjunction with a distributed auction approach. We treat AUVs as car-like models in our approach, paying meticulous attention to their operational characteristics during path planning. Importantly, we effectively utilize their backward driving capabilities. Our analysis reveals that this model successfully fulfills the directional requirements of detection tasks. Furthermore, the distributed auction approach optimizes the overall task distribution in the multi-AUV system. We support our method with simulation results that underscore its effectiveness.

Abstract Color-changeable fibers can provide diverse functions for intelligent wearable devices such as novel information displays and human–machine interfaces when woven into fabric. This work develops a low-cost, effective, and scalable strategy to produce thermochromic fibers by wet spinning. Through a combination of different thermochromic microcapsules, flexible fibers with abundant and reversible color changes are obtained. These color changes can be clearly observed by the naked eye. It is also found that the fibers exhibit excellent color-changing stability even after 8000 thermal cycles. Moreover, the thermochromic fibers can be fabricated on a large scale and easily woven or implanted into various fabrics with good mechanical performance. Driven by their good mechanical and physical characteristics, applications of thermochromic fibers in dynamic colored display are demonstrated. Dynamic quick response (QR) code display and recognition are successfully realized with thermochromic fabrics. This work well confirms the potential applications of thermochromic fibers in smart textiles, wearable devices, flexible displays, and human–machine interfaces. Graphical Abstract

Powder-based laser direct metal deposition (DMD), one of the directed energy deposition, was applied in air and underwater to repair pre-machined NV E690 steel plates. Systematic investigations on the effects of underwater environment and ambient pressures (0.01–0.35 MPa) on the microstructure evolution, phase transformation, and mechanical properties were conducted. The water quenching effect refined the grain size and increased the dislocation density and lath martensite content. The theoretical models of the underwater pressurized nitriding process and the precipitation kinetics of (Ti, V)N particles were established. Moreover, the microstructure evolution and the mechanical properties of other underwater DMD repaired samples did not show obvious relation with the underwater ambient pressures. This investigation not only provides a candidate for the underwater restoration technique but also bridges marine engineering and emerging DMD technology.

Abstract With the rapid developments of marine resource exploitation, mounts of marine engineering equipment are settled on the ocean. When it is not possible to move the damaged equipment into a dry dock, welding operations must be performed in underwater environments. The underwater laser welding/cladding technique is a promising and advanced technique which could be widely applied to the maintenance of the damaged equipment. The present review paper aims to present a critical analysis and engineering overview of the underwater laser welding/cladding technique. First, we elaborated recent advances and key issues of drainage nozzles all over the world. Next, we presented the underwater laser processing and microstructural-mechanical behavior of repaired marine materials. Then, the newly developed powder-feeding based and wire-feeding based underwater laser direct metal deposition techniques were reviewed. The differences between the convection, conduction, and the metallurgical kinetics in the melt pools during underwater laser direct metal deposition and in-air laser direct metal deposition were illustrated. After that, several challenges that need to be overcame to achieve the full potential of the underwater laser welding/cladding technique are proposed. Finally, suggestions for future directions to aid the development of underwater laser welding/cladding technology and underwater metallurgical theory are provided. The present review will not only enrich the knowledge in the underwater repair technology, but also provide important guidance for the potential applications of the technology on the marine engineering.

Underwater images contain abundant information, but many challenges remain for underwater object detection tasks. Various salient object detection methods may encounter low detection precision, and the segmented map has an incomplete region of the target object. To deal with blurry underwater scenes and vague detection problems, a novel fusion underwater salient object detection algorithm (FUSOD) is proposed based on multi-scale saliency and spatial optimization. Firstly, an improved underwater color restoration was utilized to restore the color information for afterward color contrast saliency calculation. Secondly, a more accurate multi-scale fusion saliency map was obtained by fully considering both the global and local feature contrast information. Finally, the fusion saliency was optimized by the proposed spatial optimization method to enhance the spatial coherence. The proposed FUSOD algorithm may process turbid and complex underwater scenes and preserve a complete structure of the target object. Experimental results on the USOD dataset show that the proposed FUSOD algorithm can segment the salient object with a comparatively higher detection precision than the other traditional state-of-the-art algorithms. An ablation experiment showed that the proposed spatial optimization method increases the detection precision by 0.0325 scores in the F-Measure.

Energy band alignment theory is used to understand interface charge transfer in semiconductor/semiconductor junctions but many abnormal results cannot be well explained. Here, the authors demonstrate a Faradaic junction theory with isoenergetic charge transfer at semiconductor/semiconductor interfaces to explain these abnormal results.

Abstract In recent years, researchers have been making great progress in the steganalysis technology based on convolution neural networks (CNN). However, experts ignore the contribution of nonlinear residual and joint domain detection to steganalysis, and how to detect the adaptive steganographic algorithms with low embedding rates is still challenging. In this paper, we propose a CNN steganalysis model that uses a joint domain detection mechanism and a nonlinear detection mechanism. For the nonlinear detection mechanism, based on the spatial rich model (SRM), we introduce the maximum and minimum nonlinear residual feature acquisition method into the model to adapt to the nonlinear distribution of steganography information. For the joint domain detection mechanism, we not only apply the high-pass filters from the SRM for spatial residuals, but also apply the patterns from the discrete cosine transform residual (DCTR) for transformation steganographic impacts, so as to fully capture the interference trace of spatial steganography to transform domain. We also apply a new transfer learning method to improve the model’s performance. That is, we apply the low embedding rate steganography samples to initialize the model, because we think that the method makes the network more sensitive than applying high embedding rate steganography samples to initialize the model. The simulation results also confirm this assumption. Combined with the above improved methods, the detection accuracy of the model for WOW and S-UNIWARD is higher than that of SRM+EC, Ye-Net, Xu-Net, Yedroudj-Net and Zhu-Net, which is about 4 ∼6% higher than that of the optimal Zhu-Net. The results can provide a certain reference for steganalysis and image forensics tasks.

Knowledge tracing (KT) is based on modeling students’ behavior sequences to obtain students’ knowledge state and predict students’ future performance. The KT task aims to model students’ knowledge state in real-time according to their historical learning behavior, so as to predict their future learning performance. Online education has become more critical in recent years due to the impact of COVID-19, and KT has also attracted much attention due to its importance in the education field. However, previous KT models generally have the following three problems. Firstly, students’ learning and forgetting behaviors affect their knowledge state, and past KT models have yet to exploit this fully. Secondly, the input of traditional KT models is mainly limited to students’ exercise sequence and answers. In the learning process, students’ answering performance can reflect their knowledge level. Finally, the context of students’ learning sequence also affects their judgment of the knowledge state. In this paper, we combined educational psychology theories to propose enhanced learning and forgetting behavior for contextual knowledge tracing (LFEKT). LFEKT enriches the features of exercises by introducing difficulty information and considers the influence of students’ answering behavior on the knowledge state. In order to model students’ learning and forgetting behavior, LFEKT integrates multiple influencing factors to build a knowledge acquisition module and a knowledge retention module. Furthermore, LFEKT introduces a long short-term memory (LSTM) network to capture the contextual relations of learned sequences. From the experimental results, it can be seen that LFEKT had better prediction performance than existing models on four public datasets, which indicates that LFEKT can better trace students’ knowledge state and has better prediction performance.

Background: Acute pancreatitis (AP) is a potentially life-threatening inflammatory disease with multiple etiologies. The prevalence of hypertriglyceridemia-induced acute pancreatitis (HTG-AP) has been increasing in recent years. It is reported that early triglyceride (TG) levels were associated with the severity of the disease, and TG- lowering therapies, including medical treatment and blood purification, may impact the clinical outcomes. However, there is no consensus regarding the optimal TG-lowering therapy, and clinical practice varies greatly among different centers. Our objective is to evaluate the TG-lowering effects of different therapies and their impact on clinical outcomes in HTG-AP patients with worrisome features.Methods: This is a multicenter, observational, prospective cohort study. A total of approximately 300 patients with HTG-AP with worrisome features are planned to be enrolled. The primary objective of the study is to evaluate the relationship between TG decline and the evolution of organ failure, and patients will be dichotomized depending on the rate of TG decline. The primary outcome is organ failure (OF) free days to 14 days after enrollment. Secondary outcomes include new-onset organ failure, new-onset multiple-organ failure (MOF), new-onset persistent organ failure (POF), new receipt of organ support, requirement of ICU admission, ICU free days to day 14, hospital free days to day 14, 60-day mortality, AP severity grade (Based on the Revised Atlanta Classification), and incidence of systemic and local complications. Generalized linear model (GLM), Fine and Gray competing risk regression, and propensity score matching will be used for statistical analysis.Discussion: Results of this study will reveal the current practice of TG-lowering therapy in HTG-AP and provide necessary data for future trials.

Introduction/aim: The supraphysiologic chloride concentration of normal saline may contribute to acute kidney injury (AKI). Balanced crystalloids can decrease chloride concentration and AKI in critically ill patients. We aim to test the hypothesis that, in patients with predicted severe acute pancreatitis (pSAP), compared with saline, fluid therapy with balanced crystalloids will decrease plasma chloride concentration.Methods/Design: This is a multicenter, stepped-wedge, cluster-randomized, controlled trial. All eligible patients presenting to the 11 participating sites across China during the study period will be recruited. All sites will use saline for the first month and sequentially change to balanced crystalloids at the pre-determined and randomly allocated time point. The primary endpoint is the plasma chloride concentration on day 3 of enrollment. Secondary endpoints will include major adverse kidney events on hospital discharge or day 30 (MAKE 30) and free and alive days to day 30 for intensive care admission, invasive ventilation, vasopressors, and renal replacement therapy. Additional endpoints include daily serum chloride and sequential organ failure assessment (SOFA) score over the first seven days of enrollment.Discussion: This study will provide data to define the impact of normal saline vs. balanced crystalloids on plasma chloride concentration and clinical outcomes in pSAP patients. It will also provide the necessary data to power future large-scale randomized trials relating to fluid therapy.Ethics and Dissemination: This study was approved by the ethics committee of Jinling Hospital, Nanjing University (2020NZKY-015-01) and all the participating sites. The results of this trial will be disseminated in peer-reviewed journals and at scientific conferences.Trial registration: The trial has been registered at the Chinese Clinical Trials Registry (ChiCTR2100044432).

The call for green shipping is increasing, and the reduction of greenhouse gas emissions from ships becomes more and more important. Traditional ship energy efficiency monitoring is based on the noon reports, which are susceptible to human error and have a time delay. Many ship energy efficiency monitoring systems have been designed and developed, but they usually cannot send data to the shore in time. In order to identify abnormal fuel consumption in time, this paper realizes a big data collection system for ship energy efficiency monitoring based on the BeiDou System. The system installed on two sister container ships has already collected a lot of data. Big data analysis methods, such as principal component analysis (PCA) and correlation analysis, are applied in the system to realize data visualization and analysis. Using PCA, it turns out that the shaft power of the main engine is related to a certain ship speed, which is also affected by load and weather conditions, and is the biggest factor in determining fuel consumption. To realize the assessment of hull fouling and the optimization of ship trim, a useful physics-based analysis is proposed. The analysis shows that the fouling of ship body greatly increases its resistance. Our analysis method can also find the best trim under specific loading condition. All these points are important for reducing fuel consumption and improving ship efficiency.

Cooperative hunting of multi-autonomous underwater vehicle (AUV) is an important research topic. Current studies concentrate on AUVs with the same speed abilities and mostly do not consider their speed differences. In fact, AUVs in a hunting group are often of different types and possess different maximum sailing speeds. For inhomogeneous multi-AUV, a novel time competition mechanism is proposed to construct an efficient dynamic hunting alliance. Hunting team with AUVs possessing higher speed abilities is more suitable for the vast underwater environment. In the pursuing stage, AUV needs to act fast enough to avoid the escape of evader. To achieve a quick and accurate pursuit, a combined path planning approach is presented, which combines a Glasius bio-inspired neural network model and a belief function. Simulation experiments demonstrate the feasibility and efficiency of the proposed algorithm in the cooperative hunting of inhomogeneous multi-AUV under dynamic underwater environment with intelligent evaders and multi-obstacle.