Search Results (132)

Recent advances in precision manufacturing and high-end equipment technologies have imposed ever more stringent requirements on the accuracy, real-time performance, and lightweight design of online steel strip surface defect detection systems. To reconcile the persistent trade-off between detection precision and inference efficiency in complex industrial environments, this study proposes StripSurface–YOLO, a novel real-time defect detection framework built upon YOLOv8n. The core architecture integrates an Efficient Cross-Stage Local Perception module (ResGSCSP), which synergistically combines GSConv lightweight convolutions with a one-shot aggregation strategy, thereby markedly reducing both model parameters and computational complexity. To further enhance multi-scale feature representation, this study introduces an Efficient Multi-Scale Attention (EMA) mechanism at the feature-fusion stage, enabling the network to more effectively attend to critical defect regions. Moreover, conventional nearest-neighbor upsampling is replaced by DySample, which produces deeper, high-resolution feature maps enriched with semantic content, improving both inference speed and fusion quality. To heighten sensitivity to small-scale and low-contrast defects, the model adopts Focal Loss, dynamically adjusting to sample difficulty. Extensive evaluations on the NEU-DET dataset demonstrate that StripSurface–YOLO reduces FLOPs by 11.6% and parameter count by 7.4% relative to the baseline YOLOv8n, while achieving respective improvements of 1.4%, 3.1%, 4.1%, and 3.0% in precision, recall, mAP₅₀, and mAP_50:95. Under adverse conditions—including contrast variations, brightness fluctuations, and Gaussian noise—SteelSurface-YOLO outperforms the baseline model, delivering improvements of 5.0% in mAP₅₀ and 4.7% in mAP_50:95, attesting to the model’s robust interference resistance. These findings underscore the potential of StripSurface–YOLO to meet the rigorous performance demands of real-time surface defect detection in the metal forging industry. Full article

The use of CH₄ as an energy source is increasing every day. To increase the efficiency of CH₄ combustion and ensure that the equipment meets ecological requirements, it is necessary to measure the CH₄ concentration in the exhaust gases of combustion systems. To this end, sensors are required that can withstand extreme operating conditions, including temperatures of at least 600 °C, as well as high pressure and gas flow rate. ZnGa₂O₄, being an ultra-wide bandgap semiconductor with high chemical and thermal stability, is a promising material for such sensors. The synthesis and investigation of the structural and CH₄ sensing properties of ceramic pellets made from pure and Er-doped ZnGa₂O₄ were conducted. Doping with Er leads to the formation of a secondary Er₃Ga₅O₁₂ phase and an increase in the active surface area. This structural change significantly enhanced the CH₄ response, demonstrating an 11.1-fold improvement at a concentration of 10⁴ ppm. At the optimal response temperature of 650 °C, the Er-doped ZnGa₂O₄ exhibited responses of 2.91 a.u. and 20.74 a.u. to 100 ppm and 10⁴ ppm of CH₄, respectively. The Er-doped material is notable for its broad dynamic range for CH₄ concentrations (from 100 to 20,000 ppm), low sensitivity to humidity variations within the 30–70% relative humidity range, and robust stability under cyclic gas exposure. In addition to CH₄, the sensitivity of Er-doped ZnGa₂O₄ to other gases at a temperature of 650 °C was investigated. The samples showed strong responses to C₂H₄, C₃H₈, C₄H₁₀, NO_2, and H₂, which, at gas concentrations of 100 ppm, were higher than the response to CH₄ by a factor of 2.41, 2.75, 3.09, 1.16, and 1.64, respectively. The study proposes a plausible mechanism explaining the sensing effect of Er-doped ZnGa₂O₄ and discusses its potential for developing high-temperature CH₄ sensors for applications such as combustion monitoring systems and determining the ideal fuel/air mixture. Full article

Driven by the global energy transition and the pursuit of “dual carbon” goals, sustainability risks within the coal supply chain have emerged as a central obstacle impeding the low-carbon transformation of high-carbon industries. To address the critical gap in systematic and multidimensional risk assessments for coal supply chains, this study proposes a hybrid framework that integrates the analytic hierarchy process (AHP) with the fuzzy comprehensive evaluation (FCE) method. Utilizing the Delphi method and the coefficient of variation technique, this study develops a risk assessment system encompassing eight primary criteria and forty sub-criteria. These indicators cover economic, operational safety, ecological and environmental, management policy, demand, sustainable supply, information technology, and social risks. An empirical analysis is conducted, using a prominent Chinese coal enterprise as a case study. The findings demonstrate that the overall risk level of the enterprise is “moderate”, with demand risk, information technology risk, and social risk ranking as the top three concerns. This underscores the substantial impact of accelerated energy substitution, digital system vulnerabilities, and stakeholder conflicts on supply chain resilience. Further analysis elucidates the transmission mechanisms of critical risk nodes, including financing constraints, equipment modernization delays, and deficiencies in end-of-pipe governance. Targeted strategies are proposed, such as constructing a diversified financing matrix, developing a blockchain-based data-sharing platform, and establishing a community co-governance mechanism. These measures offer scientific decision-making support for the coal industry’s efforts to balance “ensuring supply” with “reducing carbon emissions”, and provide a replicable risk assessment paradigm for the sustainable transformation of global high-carbon supply chains. Full article

The agricultural sector faces growing pressure to enhance productivity and sustainability, prompting innovation in machinery design. Traditional materials such as steel still dominate but are a cause of increased weight, soil compaction, increased fuel consumption, and corrosion. Composite materials—and, more specifically, fiber-reinforced polymers (FRPs)—offer appealing alternatives due to their high specific strength and stiffness, corrosion resistance, and design flexibility. Meanwhile, increasing environmental awareness has triggered interest in biocomposites, which contain natural fibers (e.g., flax, hemp, straw) and/or bio-based resins (e.g., PLA, biopolyesters), aligned with circular economy principles. This review offers a comprehensive overview of synthetic composites and biocomposites for agricultural machinery and equipment (AME). It briefly presents their fundamental constituents—fibers, matrices, and fillers—and recapitulates relevant mechanical and environmental properties. Key manufacturing processes such as hand lay-up, compression molding, resin transfer molding (RTM), pultrusion, and injection molding are discussed in terms of their applicability, benefits, and limits for the manufacture of AME. Current applications in tractors, sprayers, harvesters, and planters are covered in the article, with advantages such as lightweighting, corrosion resistance, flexibility and sustainability. Challenges are also reviewed, including the cost, repairability of damage, and end-of-life (EoL) issues for composites and the moisture sensitivity, performance variation, and standardization for biocomposites. Finally, principal research needs are outlined, including material development, long-term performance testing, sustainable and scalable production, recycling, and the development of industry-specific standards. This synthesis is a practical guide for researchers, engineers, and manufacturers who want to introduce innovative material solutions for more efficient, longer lasting, and more sustainable agricultural machinery. Full article

Chili pepper (Capsicum annuum L.), a globally important economic crop, faces production challenges characterized by high labor intensity, cost, and inefficiency. Intelligent technologies offer key opportunities for sector transformation. This review begins by outlining the diversity of major chili pepper cultivars, differences in key quality indicators, and the resulting specific harvesting needs. It then reviews recent progress in intelligent perception, recognition, and automation within the chili pepper industry. For perception and recognition, the review covers the evolution from traditional image processing to deep learning-based methods (e.g., YOLO and Mask R-CNN achieving a mAP > 90% in specific studies) for pepper detection, segmentation, and fine-grained cultivar identification, analyzing the performance and optimization in complex environments. In terms of automation, we systematically discuss the principles and feasibility of different mechanized harvesting machines, consider the potential of vision-based keypoint detection for the point localization of picking, and explore motion planning and control for harvesting robots (e.g., robotic systems incorporating diverse end-effectors like soft grippers or cutting mechanisms and motion planning algorithms such as RRT) as well as seed cleaning/separation techniques and simulations (e.g., CFD and DEM) for equipment optimization. The main current research challenges are listed including the environmental adaptability/robustness, efficiency/real-time performance, multi-cultivar adaptability/flexibility, system integration, and cost-effectiveness. Finally, future directions are given (e.g., multimodal sensor fusion, lightweight models, and edge computing applications) in the hope of guiding the intelligent growth of the chili pepper industry. Full article

Timely and accurate detection of agricultural disasters is crucial for ensuring food security and enhancing post-disaster response efficiency. This paper proposes a deployable UAV-based multimodal agricultural disaster detection framework that integrates multispectral and RGB imagery to simultaneously capture the spectral responses and spatial structural features of affected crop regions. To this end, we design an innovative stride–cross-attention mechanism, in which stride attention is utilized for efficient spatial feature extraction, while cross-attention facilitates semantic fusion between heterogeneous modalities. The experimental data were collected from representative wheat and maize fields in Inner Mongolia, using UAVs equipped with synchronized multispectral (red, green, blue, red edge, near-infrared) and high-resolution RGB sensors. Through a combination of image preprocessing, geometric correction, and various augmentation strategies (e.g., MixUp, CutMix, GridMask, RandAugment), the quality and diversity of the training samples were significantly enhanced. The model trained on the constructed dataset achieved an accuracy of 93.2%, an F1 score of 92.7%, a precision of 93.5%, and a recall of 92.4%, substantially outperforming mainstream models such as ResNet50, EfficientNet-B0, and ViT across multiple evaluation metrics. Ablation studies further validated the critical role of the stride attention and cross-attention modules in performance improvement. This study demonstrates that the integration of lightweight attention mechanisms with multimodal UAV remote sensing imagery enables efficient, accurate, and scalable agricultural disaster detection under complex field conditions. Full article

Aiming at the challenge of trajectory planning for a continuum manipulator in the confined spaces of gas-insulated switchgear (GIS) chambers during intelligent operation and maintenance of power equipment, this paper proposes a configuration space (C-space) obstacle-avoidance planning method based on an improved RRT-Connect algorithm. By constructing a virtual joint-space obstacle map, the collision-avoidance problem in Cartesian space is transformed into a joint-space path search problem, significantly reducing the computational burden of frequent inverse kinematics solutions inherent in traditional methods. Compared to the RRT-Connect algorithm, improvements in node expansion strategies and greedy optimization mechanisms effectively minimize redundant nodes and enhance path generation efficiency: the number of iterations is reduced by 68% and convergence speed is improved by 35%. Combined with polynomial-driven trajectory planning, the method successfully resolves and smoothens driving cable length variations, achieving efficient and stable control for both the end-effector and arm configuration of a dual-segment continuum manipulator. Simulation and experimental results demonstrate that the proposed algorithm rapidly generates collision-free arm configuration trajectories with high trajectory coincidence in typical GIS chamber scenarios, verifying its comprehensive advantages in real-time performance, safety, and motion smoothness. This work provides theoretical support for the application of continuum manipulator in precision operation and maintenance of power equipment. Full article

Accurate prediction of the remaining useful life (RUL) of rolling bearings was crucial for ensuring the safe operation of machinery and reducing maintenance losses. However, due to the high nonlinearity and complexity of mechanical systems, traditional methods failed to meet the requirements of medium- and long-term prediction tasks. To address this issue, this paper proposed a recurrent neural network with a dual attention model. By employing path weight selection methods, Discrete Fourier transform, and selection mechanisms, the prediction accuracy and generalization ability in complex time series analysis were significantly improved. Evaluation results based on mean absolute error (MAE) and root mean square error (RMSE) indicated that the dual attention mechanism effectively focused on key features, optimized feature extraction, and improved prediction performance. An end-to-end RUL prediction model was established based on the MS-DAN network, and the effectiveness of the method was validated using the IEEE PHM 2012 Data Challenge dataset, providing more accurate decision support for equipment maintenance engineers. Full article

The safety helmet is the equipment that construction workers must wear, and it plays an important role in protecting their lives. However, there are still many construction workers who do not pay attention to the wearing of helmets. Therefore, the real-time high-precision intelligent detection of construction workers’ helmet wearing is crucial. To this end, this paper proposes a lightweight helmet-wearing detection algorithm based on StarNet-YOLOv10. Firstly, the StarNet network structure is used to replace the backbone network part of the original YOLOv10 model while retaining the original Spatial Pyramid Pooling Fast (SPPF) and Partial Self-attention (PSA) parts. Secondly, the C2f module in the neck network is optimised by combining the PSA attention module and the GhostBottleneckv2 module, which improves the extraction of feature information and the expression ability of the model. Finally, optimisation is performed in the head network by introducing the Large Separable Kernel Attention (LSKA) attention mechanism to improve the detection accuracy and detection efficiency of the detection head. The experimental results show that compared with the existing Faster R-CNN, YOLOv5s, YOLOv6, and the original YOLOv10 models, the StarNet-YOLOv10 model proposed in this paper has a greater degree of improvement in the accuracy, recall, average precision mean, computational volume, and frame rate, in which the accuracy is as high as 83.36%, the recall rate can be up to 81.17%, and the average precision mean can reach 78.66%. Meanwhile, compared with the original YOLOv10 model, this model improves 1.7% in accuracy, 1.62% in recall, and 4.43% in mAP. Therefore, the present model can well meet the detection requirements of helmet wearing and can effectively reduce the safety hazards caused by not wearing helmets on construction sites. Full article

With the improvement in industrial equipment intelligence and reliability requirements, bearing fault diagnosis has become a key technology to ensure the stable operation of mechanical equipment. Traditional bearing fault diagnosis methods are ineffective in diagnosing complex faults and mostly rely on the manual adjustment of hyperparameters. To this end, this paper proposes a domain adversarial migratory learning bearing fault diagnosis model incorporating structural adjustment modules. First, the pre-trained model of the source domain is applied to the target domain dataset through an adversarial domain adaptation technique. Then, the network depth and width are dynamically adjusted in the Optuna optimization framework to accommodate more complex fault types in the target domain. Finally, the performance of the model is further improved by automatically optimizing the hyperparameters. The experimental results show that the model exhibits high accuracy in the diagnosis of different fault types, especially in the face of complex and variable industrial environments, demonstrating strong adaptability and robustness. The method provides an effective solution for fault diagnosis of intelligent devices. Full article

In order to address the impact of equipment fault diagnosis and repair delays on production schedule execution in the dynamic scheduling of flexible job shops, this paper proposes a multi-resource, multi-objective dynamic scheduling optimization model, which aims to minimize delay time and completion time. It integrates the scheduling of the workpieces, machines, and maintenance personnel to improve the response efficiency of emergency equipment maintenance. To this end, a self-learning Ant Colony Algorithm based on deep reinforcement learning (ACODDQN) is designed in this paper. The algorithm searches the solution space by using the ACO, prioritizes the solutions by combining the non-dominated sorting strategies, and achieves the adaptive optimization of scheduling decisions by utilizing the organic integration of the pheromone update mechanism and the DDQN framework. Further, the generated solutions are locally adjusted via the feasible solution optimization strategy to ensure that the solutions satisfy all the constraints and ultimately generate a Pareto optimal solution set with high quality. Simulation results based on standard examples and real cases show that the ACODDQN algorithm exhibits significant optimization effects in several tests, which verifies its superiority and practical application potential in dynamic scheduling problems. Full article

In emergency scenarios, we focus on studying how to manipulate legged robot dogs equipped with robotic arms to move and operate in a small space, known as legged emergency manipulation. Although the legs of the robotic dog are mainly used for movement, we found that implementing a whole-body control strategy can enhance its operational capabilities. This means that the robotic dog’s legs and mechanical arms can be synchronously controlled, thus expanding its working range and mobility, allowing it to flexibly enter and exit small spaces. To this end, we propose a framework that can utilize visual information to provide feedback for whole-body control. Our method combines low-level and high-level strategies: the low-level strategy utilizes all degrees of freedom to accurately track the body movement speed of the robotic dog and the position of the end effector of the robotic arm; the advanced strategy is based on visual input, intelligently planning the optimal moving speed and end effector position. At the same time, considering the uncertainty of visual guidance, we integrate fully supervised learning into the advanced strategy to construct a teacher network and use it as a benchmark network for training the student network. We have rigorously trained these two levels of strategies in a simulated environment, and through a series of extensive simulation validations, we have demonstrated that our method has significant improvements over baseline methods in moving various objects in a small space, facing different configurations and different target objects. Full article

East Dongting Lake is an important habitat for migratory birds. Accurately counting the number of migratory birds is crucial to assessing the health of the wetland ecological environment. Traditional manual observation and low-precision methods make it difficult to meet this demand. To this end, this paper proposes a weighted feature fusion network based on multi-level supervision (MS-WFFNet) to count migratory birds. MS-WFFNet consists of three parts: an EEMA-VGG16 sub-network, a multi-source feature aggregation (MSFA) module, and a density map regression (DMR) module. Among them, the EEMA-VGG16 sub-network cross-injects enhanced efficient multi-scale attention (EEMA) into the truncated VGG16 structure. It uses multi-head attention to nonlinearly learn the relative importance of different positions in the same direction. With only a few parameters added, EEMA effectively suppresses the noise interference caused by a cluttered background. The MSFA module integrates a weighted mechanism to fully preserve low-level detail information and high-level semantic information. It achieves this by aggregating multi-source features and enhancing the expression of key features. The DMR module applies density map regression to the output of each path in the MSFA module. It ensures local consistency and spatial correlation among multiple regression results by using distributed supervision. In addition, this paper presents the migratory bird counting dataset DTH, collected using local monitoring equipment in East Dongting Lake. It is combined with other object counting datasets for extensive experiments, showcasing the proposed method’s excellent performance and generalization capability. Full article

With the wide application of a new generation of information technology, remote technical services are receiving an increasing amount of attention in the manufacturing field. In view of the fact that most mechanical and electrical equipment manufacturing enterprises still need to send a substantial number of employees to the site to provide operation and maintenance services for customers, and the operation and maintenance costs of enterprises remain high, a remote operation and maintenance method of mechanical and electrical equipment based on digital twin technology is proposed. A digital twin remote operation and maintenance services model is constructed, and digital twin remote operation and maintenance technology is divided into four basic levels: virtual simulation, software/hardware-in-the-loop virtual commissioning, virtual and real synchronization, and cloud–end interconnection. With this, we conduct in-depth research on the key technologies involved in these four levels. The digital twin remote operation and maintenance service platform has been built; this platform can provide maintenance, repair, overhaul, and operation services for enterprise equipment. The feasibility of both the digital twin remote operation and maintenance services model and the digital twin remote operation and maintenance service platform was verified through cases, which provided an efficient and feasible solution for enterprises to improve service efficiency and reduce labor costs. Full article

In this paper, the Efficient Channel Attention (ECA) mechanism is incorporated at the terminal layer of the YOLOv10 backbone network to enhance the feature expression capability. In addition, Transformer is introduced into the C3 module in the feature extraction process to construct the C3TR module to replace the original C2F module as the deepening network extraction module. In this study, both the ECA mechanism and the self-attention mechanism of Transformer are thoroughly analyzed and integrated into YOLOv10. The C3TR module is used as an important part to deepen the effect of network extraction in backbone network feature extraction. The self-attention mechanism is used to model the long-distance dependency relationship, capture the global contextual information, make up for the limitation of the local sensory field, and enhance the feature expression capability. The ECA module is added to the end of the backbone to globally model the channels of the feature map, distribute channel weights more equitably, and enhance feature expression capability. Extensive experiments on the electrical equipment dataset have demonstrated the high accuracy of the method, with a mAP of 89.4% compared to the original model, representing an improvement of 3.2%. Additionally, the mAP@[0.5, 0.95] reaches 61.8%, which is 5.2% higher than that of the original model. Full article