Search Results (268)

Object detection technology enables the automatic identification of photovoltaic (PV) panel locations and conditions, significantly enhancing operational efficiency for maintenance teams while reducing the time and cost associated with manual inspections. Challenges arise due to the low resolution of remote sensing images combined with small-sized targets—PV panels intertwined with complex urban or natural backgrounds. To address this, a parallel architecture model based on YOLOv5 was designed, substituting traditional residual connections with parallel convolution structures to enhance feature extraction capabilities and information transmission efficiency. Drawing inspiration from the bottleneck design concept, a primary feature extraction module framework was constructed to optimize the model’s deep learning capacity. The improved model achieved a 4.3% increase in mAP, a 0.07 rise in F1 score, a 6.55% enhancement in recall rate, and a 6.2% improvement in precision. Additionally, the study validated the model’s performance and examined the impact of different loss functions on it, explored learning rate adjustment strategies under various scenarios, and analyzed how individual factors affect learning rate decay during its initial stages. This research notably optimizes detection accuracy and efficiency, holding promise for application in large-scale intelligent PV power station maintenance systems and providing reliable technical support for clean energy infrastructure management. Full article

The rapid growth of the Internet of Things (IoT) has revolutionized various industries by enabling interconnected devices to exchange data seamlessly. However, IoT systems face significant security challenges due to decentralized architectures, resource-constrained devices, and dynamic network environments. These challenges include denial-of-service (DoS) attacks, anomalous network behaviors, and data manipulation, which threaten the security and reliability of IoT ecosystems. New methods based on machine learning have been reported in the literature, addressing topics such as intrusion detection and prevention. This paper proposes an advanced anomaly detection framework for IoT networks expressed in several phases. In the first phase, data preprocessing is conducted using techniques like the Median-KS Test to remove noise, handle missing values, and balance datasets, ensuring a clean and structured input for subsequent phases. The second phase focuses on optimal feature selection using a Genetic Algorithm enhanced with eagle-inspired search strategies. This approach identifies the most significant features, reduces dimensionality, and enhances computational efficiency without sacrificing accuracy. In the final phase, an ensemble classifier combines the strengths of the Decision Tree, Random Forest, and XGBoost algorithms to achieve the accurate and robust detection of anomalous behaviors. This multi-step methodology ensures adaptability and scalability in handling diverse IoT scenarios. The evaluation results demonstrate the superiority of the proposed framework over existing methods. It achieves a 12.5% improvement in accuracy (98%), a 14% increase in detection rate (95%), a 9.3% reduction in false positive rate (10%), and a 10.8% decrease in false negative rate (5%). These results underscore the framework’s effectiveness, reliability, and scalability for securing real-world IoT networks against evolving cyber threats. Full article

Power quality is a key issue in cold ironing (CI) systems, where a stable, clean power supply is essential to meet the needs of moored vessels. According to IEC/ISO/IEEE 80005-1, these systems must deliver high power at standardized voltages (6.6 kV or 11 kV) with minimal harmonic distortion in the presence of vessel load variability. This study proposes a model-free control strategy based on an intelligent proportional–integral (iPI) corrector with adaptive gain, applied to a three-phase modular multilevel converter (MMC) equipped with an LC filter. This architecture, adapted to distributed infrastructures, reduces the number of transformers required while guaranteeing high output voltages. The iPI strategy improves system robustness, dynamically compensates for disturbances, and ensures better power quality. A comparative analysis of three control strategies, proportional–integral (PI), intelligent proportional–integral (iPI), and intelligent proportional–integral adaptive (iPIa), performed in MATLAB/Simulink and complemented by experimental tests on the OPAL-RT platform, revealed a significant THD reduction of 1.18%, in accordance with the IEC/ISO/IEEE 80005-1 standard. These results confirm the effectiveness of the proposed method in meeting the requirements of CI systems. Full article

Natural surfaces, such as lotus leaves, springtail cuticles, and pitcher plant peristomes, exhibit extraordinary wetting behaviors due to their unique surface topographies and chemical compositions. These natural architectures have inspired the development of wettability models and the production of artificial surfaces with tailored wettability for advanced applications. Electrodeposited metallic coatings can imitate the wettability behaviors of natural surfaces, showing superhydrophobic, superoleophobic, or slippery characteristics. Such coatings can significantly enhance corrosion resistance by minimizing water–metal contact and promoting self-cleaning effects. This review presents various strategies for fabricating corrosion-resistant metallic coatings, including different electrodeposition techniques in aqueous or non-aqueous baths, followed by post-treatment procedures and surface functionalization methods. However, despite the promising protective properties demonstrated under controlled laboratory conditions, long-term studies under natural exposure conditions are still lacking, which limits the full assessment of the durability and effectiveness of non-wettable electroplated deposits in practical applications. Full article

Poly(acrylonitrile-butadiene-styrene) (ABS) is a common polymer used in toys, automobile parts, and membranes. Membranes fabricated with this copolymer commonly employ toxic solvents and have a dense architecture, which may not work in all applications. This work investigates the synthesis of ABS membranes, using green solvents and the influence of additives on the phase inversion process during the non-solvent-induced phase separation. The addition of water-soluble additives, ethanol, and acetone is hypothesized to provide additional control over viscosity and volatility, and, consequently, impact the phase inversion process. Membranes were fabricated with PolarClean and with various additive concentrations and evaporation times. The resulting membranes were characterized using scanning electron microscopy (SEM) and a pycnometer to visualize the pore structure and obtain porosity information. Membrane performance, including water flux and bovine serum albumin rejection, was evaluated using dead-end cell filtration. Membranes fabricated using only PolarClean had fingerlike pore morphology and relatively low protein rejection. The addition of additives resulted in a change in pore architecture and rejection, which is hypothesized to be a result of additives’ volatility, humidity, and destabilization of liquid–liquid separation. This study provides a more detailed understanding of the impact of additives on the resulting ABS membrane structure and performance, with a focus on safer solvents. Full article

Research on bionic superhydrophobic surfaces draws inspiration from the microstructures and wetting mechanisms of natural organisms such as lotus leaves, water striders, and butterfly wings, offering innovative approaches for developing artificial functional surfaces. By synergistically combining micro/nano hierarchical structures with low surface energy chemical modifications, researchers have devised various fabrication strategies—including laser etching, sol-gel processes, electrochemical deposition, and molecular self-assembly—to achieve superhydrophobic surfaces characterized by contact angles exceeding 150° and sliding angles below 5°. These technologies have found widespread applications in self-cleaning architectural coatings, efficient oil–water separation membranes, anti-icing materials for aviation, and anti-biofouling medical devices. This article begins by examining natural organisms exhibiting superhydrophobic properties, elucidating the principles underlying their surface structures and the wetting states of droplets on solid surfaces. Subsequently, it categorizes and highlights key fabrication methods and application domains of superhydrophobic surfaces, providing an in-depth and comprehensive discussion. Full article

Cleaning operations in narrow pipelines are often hindered by limited maneuverability and low efficiency, necessitating the development of a high-performance and highly adaptable robotic solution. To address this challenge, this study proposes a pipeline-cleaning robot specifically designed for the heat-exchange tubes of industrial heat exchangers. The robot features a dual-wheel cross-drive configuration to enhance motion stability and integrates a gear–rack-based alignment mechanism with a cam-based propulsion system to enable autonomous deployment and cleaning via a flexible arm. The robot adopts a modular architecture with a separated body and cleaning arm, allowing for rapid assembly and maintenance through bolted connections. A vision-guided control system is implemented to support accurate positioning and task scheduling within the primary pipeline. Experimental results demonstrate that the robot can stably execute automatic navigation and sub-pipe cleaning, achieving pipe-switching times of less than 30 s. The system operates reliably and significantly improves cleaning efficiency. The proposed robotic system exhibits strong adaptability and generalizability, offering an effective solution for automated cleaning in confined pipeline environments. Full article

Although both noise suppression and speech restoration are fundamental to speech enhancement, many Deep neural network (DNN)-based approaches tend to focus disproportionately on one, often overlooking the importance of their joint handling. In this study, we propose a dual-path architecture designed to balance noise suppression and speech restoration. The main path consists of an encoder and two specialized decoders: one dedicated to estimating the clean speech spectrum and the other to predicting a noise suppression mask. To reinforce the joint modeling of noise suppression and speech restoration, we introduce an auxiliary refinement path. This path consists of a separate encoder–decoder structure and is designed to further refine the enhanced speech by incorporating complementary information, learned independently from the main path. By using this dual-path architecture, the model better preserves fine speech details while reducing residual noise. Experimental results on the VoiceBank + DEMAND dataset show that our model surpasses conventional methods across multiple evaluation metrics in the causal setup. Specifically, it achieves a PESQ score of 3.33, reflecting improved speech quality, and a CSIG score of 4.48, indicating enhanced intelligibility. Furthermore, it demonstrates superior noise suppression, achieving an SNRseg of 10.44 and a CBAK score of 3.75. Full article

The colour degradation of murals presents a significant challenge in the conservation of architectural heritage. Previous research has often concentrated on localized pigment changes while paying insufficient attention to the interaction between colour variation and indoor environmental conditions. Although non-destructive analytical techniques are widely used in heritage studies, their integrated application in combination with colourimetry has been limited, particularly in the context of Tibetan Buddhist murals in highland continental climates. This study investigates the murals of Liuli Hall in Meidai Lamasery, Inner Mongolia, as a representative case. We employed a comprehensive methodology that combines non-destructive analytical tools, gas chromatography–mass spectrometry, and quantitative colour analysis to examine pigment composition, binding material, and surface deterioration. Through joint analysis using the CIE Lab and CIE LCh colour space systems, we quantified mural colour changes and explored their correlation with material degradation and environmental exposure. The pigments identified include cinnabar, atacamite, azurite, and chalk, with animal glue and drying oils as binding materials. Colourimetric results revealed pronounced yellowing on the east and west walls, primarily caused by the ageing of organic binders. In contrast, a notable reduction in brightness on the south wall was attributed to dust accumulation. These findings support tailored conservation measures such as regular surface cleaning for the south wall and antioxidant stabilization treatments for the east and west walls. Initial cleaning efforts proved effective. The integrated approach adopted in this study provides a replicable model for mural diagnostics and conservation under complex environmental conditions. Full article

In the context of the rapid popularization of clean energy, the precise identification of surface defects on photovoltaic modules has become a core technical bottleneck limiting the operational efficiency of power stations. In response to the shortcomings of existing detection methods in identifying tiny defects and model efficiency, this study innovatively constructed the EER-DETR detection framework: firstly, a feature reconstruction module WDBB with a differentiable branch structure was introduced to significantly enhance the feature retention ability for fine cracks and other small targets; secondly, an adaptive feature pyramid network EHFPN was innovatively designed, which achieved efficient integration of multi-level features through a dynamic weight allocation mechanism, reducing the model complexity by 9.7% while maintaining detection accuracy, solving the industry problem of “precision—efficiency imbalance” in traditional feature pyramid networks; finally, an enhanced upsampling component was introduced to effectively address the problem of detail loss that occurs in traditional methods during image resolution enhancement. Experimental verification shows that the improved algorithm increased the average precision (mAP@0.5) on the panel dataset by 1.9%, and its comprehensive performance also exceeded RT-DETR. Based on the industry standard PVEL-AD, the detection rate of typical defects significantly improved compared to the baseline model. The core innovation of this research lies in the combination of differentiable architecture design and dynamic feature management, providing a detection tool for the intelligent operation and maintenance of photovoltaic power stations that possesses both high precision and lightweight characteristics. It has significant engineering application value and academic reference significance. Full article

The aim of sequential recommendation (SR) is to predict a user’s next interaction by analyzing their historical behavioral sequences. The proposed framework leverages the inherent symmetry in user behavior patterns, where temporal and spectral representations exhibit complementary structures that can be harmonized for robust recommendation. Conventional SR methods predominantly utilize implicit feedback (e.g., clicks and views) as model inputs, whereby observed interactions are treated as positive instances, while unobserved ones are considered negative samples. However, the inherent randomness and diversity in user behaviors inevitably introduce noise into such implicit feedback, potentially compromising the accuracy of recommendations. Recent studies have explored noise mitigation through two primary approaches: temporal-domain methods that reweight interactions to distill clean samples for comprehensive user preference modeling, and frequency-domain techniques that purify item embeddings to reduce the propagation of noise. While temporal approaches excel in sample refinement, frequency-based methods demonstrate superior capability in learning noise-resistant representations through spectral analysis. Motivated by the desire to synergize these complementary advantages, we propose SR-DFN, a novel framework that systematically addresses noise interference through coordinated time–frequency processing. Self-guided sample purification is implemented in the temporal domain of our architecture via adaptive interaction weighting, effectively distilling behaviorally significant patterns. The refined sequence is then transformed into the frequency domain, where learnable spectral filters operate to further attenuate residual noise components while preserving essential preference signals. Drawing on the convolution theorem’s revelation regarding frequency-domain operations capturing behavioral periodicity, we critically examine conventional position encoding schemes and propose an efficient parameterization strategy that eliminates redundant positional embeddings without compromising temporal awareness. Comprehensive experiments conducted on four real-world benchmark datasets demonstrate SR-DFN’s superior performance over state-of-the-art baselines, with ablation studies validating the effectiveness of our dual-domain denoising mechanism. Our findings suggest that coordinated time–frequency processing offers a principled solution for noise-resilient sequential recommendation while challenging conventional assumptions about positional encoding requirements in spectral-based approaches. The symmetry principles underlying our dual-domain approach demonstrate how the balanced processing of temporal and frequency domains can achieve superior noise resilience. Full article

We propose RobustQuote, a novel defense framework designed to enhance the adversarial robustness of vision transformers. The core idea is to leverage trusted reference images drawn from a dynamically changing pool unknown to the attacker as contextual anchors to detect and correct adversarial perturbations in input images. By quoting semantic features from uncorrupted references, RobustQuote mitigates the propagation of corrupted features through the model and uses these references to compute the rectification term. RobustQuote consists of two key modules: a quotation mechanism that propagates the global semantic tokens (i.e., the [cls] tokens) from reference images, and a rectification mechanism that adjusts the image tokens of the adversarial input using contextual signals from those references. During training, the model is explicitly guided to detect and rectify adversarial inputs more aggressively than clean ones. This approach is modular and plug-and-play, making it compatible with a wide range of vision transformer architectures, such as DeiT. Experiments show that RobustQuote achieves adversarial robustness on par with TRADES-trained DeiT under strong threat models, even when the attacker is aware of the reference set. In a more realistic setting where the attacker lacks access to the references, RobustQuote outperforms the second-best defense by +12.2% adversarial accuracy against the C&W attack. Our findings highlight the underexplored potential of incorporating external, attacker-unknown context as a robust defense strategy. RobustQuote offers a promising direction for addressing evolving adversarial threats in modern vision models, directly aligning with the critical challenges in adversarial machine learning and cybersecurity. Full article

High-throughput image analysis is a key tool for the efficient assessment of quantitative plant phenotypes. A typical approach to the computation of quantitative plant traits from image data consists of two major steps including (i) image segmentation followed by (ii) calculation of quantitative traits of segmented plant structures. Despite substantial advancements in deep learning-based segmentation techniques, minor artifacts of image segmentation cannot be completely avoided. For several commonly used traits including plant width, height, convex hull, etc., even small inaccuracies in image segmentation can lead to large errors. Ad hoc approaches to cleaning ’small noisy structures’ are, in general, data-dependent and may lead to substantial loss of relevant small plant structures and, consequently, falsified phenotypic traits. Here, we present a straightforward end-to-end approach to direct computation of phenotypic traits from image data using a deep learning regression model. Our experimental results show that image-to-trait regression models outperform a conventional segmentation-based approach for a number of commonly sought plant traits of plant morphology and health including shoot area, linear dimensions and color fingerprints. Since segmentation is missing in predictions of regression models, visualization of activation layer maps can still be used as a blueprint to model explainability. Although end-to-end models have a number of limitations compared to more complex network architectures, they can still be of interest for multiple phenotyping scenarios with fixed optical setups (such as high-throughput greenhouse screenings), where the accuracy of routine trait predictions and not necessarily the generalizability is the primary goal. Full article

Against the backdrop of the steady advancement of the national rural revitalization strategy and the dual-carbon goals, the low-carbon transformation of rural energy systems is of critical importance. This study first proposes a comprehensive architecture for rural energy supply systems, incorporating four key dimensions: investment, system configuration, user demand, and policy support. Leveraging the abundant wind, solar, and biomass resources available in rural areas, a low-carbon optimization model for rural energy system operation is developed. The model accounts for diverse load characteristics and the integration of elevated water towers, which serve both energy storage and agricultural functions. The optimization framework targets the multi-energy demands of rural production and daily life—including electricity, heating, cooling, and gas—and incorporates the stochastic nature of wind and solar generation. To address renewable energy uncertainty, the Fisher optimal segmentation method is employed to extract representative scenarios. A representative rural region in China is used as the case study, and the system’s performance is evaluated across multiple scenarios using the Gurobi solver. The objective functions include maximizing clean energy benefits and minimizing carbon emissions. Within the system, flexible resources participate in demand response based on their specific response characteristics, thereby enhancing the overall decarbonization level. The energy storage aggregator improves renewable energy utilization and gains economic returns by charging and discharging surplus wind and solar power. The elevated water tower contributes to renewable energy absorption by storing and releasing water, while also supporting irrigation via a drip system. The simulation results demonstrate that the proposed clean energy system and its associated operational strategy significantly enhance the low-carbon performance of rural energy consumption while improving the economic efficiency of the energy system. Full article

Within the framework of the SOCIETAL CHALLENGES—Secure, Clean, and Efficient Energy objective under the European Horizon 2020 research and innovation funding program, the SO-FREE project has developed a future-ready solid oxide fuel cell (SOFC) system with high-efficiency heat recovery. The system concept prioritizes low emissions, fuel flexibility, modular power production, and efficient thermal management. A key design feature is the interchangeability of two different SOFC stack types, allowing for operation under different temperature conditions. The system was developed with a strong emphasis on simplicity, minimizing the number of components to reduce overall plant costs while maintaining high performance. This paper presents the simulation results of the proposed flexible SOFC system, conducted using Aspen Plus^® software version 11 to establish a baseline architecture for real plant development. The simulated layout consists of an autothermal reformer (ATR), a high-temperature blower, an SOFC stack, a burner, and a heat recovery system incorporating four heat exchangers. Simulations were performed for two different anodic inlet temperatures (600 °C and 700 °C) and three fuel compositions (100% CH₄, 100% H₂, and 50% H₂ + 50% CH₄), resulting in six distinct operating scenarios. The results demonstrate a system utilization factor (UF_F) exceeding 90%, electrical efficiency ranging from 60% to 77%, and an effective heat recovery rate above 60%. These findings were instrumental in the development of the Piping and Instrumentation Diagram (P&ID) required for the design and implementation of the real system. The proposed SOFC system represents a cost-effective and adaptable energy conversion solution, contributing to the advancement of high-efficiency and low-emission power generation technologies. Full article