摘要：
To study the evolution of the rheological properties of warm mix asphalt (WMA) and its mastic viscoelastic drive prediction model, different dosages of Sasobit (S) and Evotherm 3G (G) WMA were prepared. The temperature sensitivity, high temperature deformation resistance, and low temperature viscoelastic properties of the two WMA were analyzed using a bonding material high and low temperature performance test. By combining the Refatus and Burgers models, the optimal dosage of the two warm mixes (S and G) was determined to be 4.0 % and 0.6 %, respectively. The prediction accuracy of three micromechanical models, including Hashin and Buttlar, for the high temperature performance of mastic was compared through the particle reinforcement theory system. The results showed that the warm mix additives significantly improved the high temperature stability performance of the asphalt, reduced the temperature sensitivity of the asphalt, and the Refatus model parameter a was decreased by 9.15 % and 1.0 % for B-S-4 and B-G-0.6 (denoting 4 % S and 0.6 % G added to the SBS modified asphalt, respectively). The creep deformation of WMA was smaller than that of the asphalt in the original samples, and the warm mix additives effectively improved the low temperature cracking resistance of asphalt. The Buttlar model optimally predicted the high temperature performance in the mastic (71 °C/29 °C correlation coefficient R 2 = 0.991/0.917), and the adsorption of mineral powders with SBS modified asphalt was superior to that of the base asphalt. At the same time, the warming agent had a detrimental effect on the asphalt adsorption. The results provide a theoretical basis for predicting WMA mastic’s multiscale performance and for designing sustainable pavement materials.
To study the evolution of the rheological properties of warm mix asphalt (WMA) and its mastic viscoelastic drive prediction model, different dosages of Sasobit (S) and Evotherm 3G (G) WMA were prepared. The temperature sensitivity, high temperature deformation resistance, and low temperature viscoelastic properties of the two WMA were analyzed using a bonding material high and low temperature performance test. By combining the Refatus and Burgers models, the optimal dosage of the two warm mixes (S and G) was determined to be 4.0 % and 0.6 %, respectively. The prediction accuracy of three micromechanical models, including Hashin and Buttlar, for the high temperature performance of mastic was compared through the particle reinforcement theory system. The results showed that the warm mix additives significantly improved the high temperature stability performance of the asphalt, reduced the temperature sensitivity of the asphalt, and the Refatus model parameter a was decreased by 9.15 % and 1.0 % for B-S-4 and B-G-0.6 (denoting 4 % S and 0.6 % G added to the SBS modified asphalt, respectively). The creep deformation of WMA was smaller than that of the asphalt in the original samples, and the warm mix additives effectively improved the low temperature cracking resistance of asphalt. The Buttlar model optimally predicted the high temperature performance in the mastic (71 °C/29 °C correlation coefficient R 2 = 0.991/0.917), and the adsorption of mineral powders with SBS modified asphalt was superior to that of the base asphalt. At the same time, the warming agent had a detrimental effect on the asphalt adsorption. The results provide a theoretical basis for predicting WMA mastic’s multiscale performance and for designing sustainable pavement materials.
期刊：
Renewable & Sustainable Energy Reviews,2026年226:116230 ISSN：1364-0321
通讯作者：
Chuanchang Li
作者机构：
[Xinrui Yan; Baoshan Xie; Chuanchang Li] Key Laboratory of Renewable Energy Electric-Technology of Hunan Province, School of Energy and Power Engineering, Changsha University of Science and Technology, Changsha, 410114, China
通讯机构：
[Chuanchang Li] K;Key Laboratory of Renewable Energy Electric-Technology of Hunan Province, School of Energy and Power Engineering, Changsha University of Science and Technology, Changsha, 410114, China
摘要：
With the ongoing advancement of aerospace technology, the demand for high-performance materials is rising. Phase change materials (PCMs), known for their unique thermophysical properties and versatility, offer new opportunities for breakthroughs in aerospace applications. PCMs, characterized by their low density, high energy storage density, and robust cycle stability, are ideal for aircraft lightweighting and thermal management of electronic devices. This review provides an overview of PCMs, including their mechanism, classification, preparation methods, and performance optimization. It then outlines the selection criteria for aerospace applications, emphasizing attributes such as lightweight design, long-term cycle stability, high thermal conductivity, resistance to extreme temperatures and radiation, and compatibility with existing equipment. Finally, the review explores recent advancements in PCM applications in aerospace, addressing the associated challenges and future prospects.
With the ongoing advancement of aerospace technology, the demand for high-performance materials is rising. Phase change materials (PCMs), known for their unique thermophysical properties and versatility, offer new opportunities for breakthroughs in aerospace applications. PCMs, characterized by their low density, high energy storage density, and robust cycle stability, are ideal for aircraft lightweighting and thermal management of electronic devices. This review provides an overview of PCMs, including their mechanism, classification, preparation methods, and performance optimization. It then outlines the selection criteria for aerospace applications, emphasizing attributes such as lightweight design, long-term cycle stability, high thermal conductivity, resistance to extreme temperatures and radiation, and compatibility with existing equipment. Finally, the review explores recent advancements in PCM applications in aerospace, addressing the associated challenges and future prospects.
作者：
Tao Li*;Yongfei Lin;Ling Zeng;Xiaowei Tang;Gang Yang;...
期刊：
Soil Dynamics and Earthquake Engineering,2026年200:109721 ISSN：0267-7261
通讯作者：
Tao Li
作者机构：
Key Laboratory of Safety Control of Bridge Engineering, Ministry of Education, Changsha University of Science & Technology, Changsha, China;[Ling Zeng] School of Civil and Environmental Engineering, Changsha University of Science & Technology, Changsha, China;[Xiaowei Tang; Gang Yang] State Key Laboratory of Coastal and Offshore Engineering, Dalian University of Technology, Dalian, China;[Shun Liu] China Institute of Water Resources and Hydropower Research, Beijing, China;[Tao Li; Yongfei Lin] Key Laboratory of Safety Control of Bridge Engineering, Ministry of Education, Changsha University of Science & Technology, Changsha, China<&wdkj&>School of Civil and Environmental Engineering, Changsha University of Science & Technology, Changsha, China
通讯机构：
[Tao Li] K;Key Laboratory of Safety Control of Bridge Engineering, Ministry of Education, Changsha University of Science & Technology, Changsha, China<&wdkj&>School of Civil and Environmental Engineering, Changsha University of Science & Technology, Changsha, China
摘要：
Static liquefaction will occur in saturated sand-clay mixtures under static loading. Understanding the mechanical behavior of sand-clay mixtures is crucial for evaluating its safety and stability. Through a series of isotropically consolidated undrained triaxial compression tests (CU), the effect of clay content (CC) on static shear strength of sand-clay mixtures were studied under constant sand skeleton void ratio (e s ). Test results demonstrate that static liquefaction happened for specimens with CC = 0 % (pure sand), 3 %, 5 %, 7 %, and 10 % under different confining pressures. By contrast, no static liquefaction occurred for specimens with CC = 12 % and 15 %. When CC = 0 %∼15 %, the peak deviator stress (q peak ) and mean effective stress at the steady state (p ss ') of corresponding specimens raise 305.6 %, 94.4 % and 97.6 % as well as 114.1 %, 987.9 % and 266.2 % under three different confining pressures, respectively. In addition, the microscopic characteristics of sand-clay mixtures with various clay contents were observed. It can be found that when CC ≤ 10 %, the clay particles primarily filled in the inter-sand voids, distributed on the surfaces of sand particles and located at the sand-sand contact points. And these clay particles can lubricate and bond sand particles, which can promote the liquefaction of mixed soil. The bonding effect of clay on sand is further enhanced when CC = 12 % and 15 %, and clay play an inhibitory role in the liquefaction of sand. Finally, a calculation equation of clay participation coefficient was proposed in current study, which can consider the effect of content, particle size and plasticity index of clay on the mechanical properties of sand-clay mixtures. The equation demonstrates excellent fitting results for both steady state data and cyclic stress ratio data in the present study and relevant literature.
Static liquefaction will occur in saturated sand-clay mixtures under static loading. Understanding the mechanical behavior of sand-clay mixtures is crucial for evaluating its safety and stability. Through a series of isotropically consolidated undrained triaxial compression tests (CU), the effect of clay content (CC) on static shear strength of sand-clay mixtures were studied under constant sand skeleton void ratio (e s ). Test results demonstrate that static liquefaction happened for specimens with CC = 0 % (pure sand), 3 %, 5 %, 7 %, and 10 % under different confining pressures. By contrast, no static liquefaction occurred for specimens with CC = 12 % and 15 %. When CC = 0 %∼15 %, the peak deviator stress (q peak ) and mean effective stress at the steady state (p ss ') of corresponding specimens raise 305.6 %, 94.4 % and 97.6 % as well as 114.1 %, 987.9 % and 266.2 % under three different confining pressures, respectively. In addition, the microscopic characteristics of sand-clay mixtures with various clay contents were observed. It can be found that when CC ≤ 10 %, the clay particles primarily filled in the inter-sand voids, distributed on the surfaces of sand particles and located at the sand-sand contact points. And these clay particles can lubricate and bond sand particles, which can promote the liquefaction of mixed soil. The bonding effect of clay on sand is further enhanced when CC = 12 % and 15 %, and clay play an inhibitory role in the liquefaction of sand. Finally, a calculation equation of clay participation coefficient was proposed in current study, which can consider the effect of content, particle size and plasticity index of clay on the mechanical properties of sand-clay mixtures. The equation demonstrates excellent fitting results for both steady state data and cyclic stress ratio data in the present study and relevant literature.
关键词：
Patient gesture recognition;Rehabilitation medicine;Feature fusion mechanism;Cognitive impairment;Gathering and distribution
摘要：
As for patients with cognitive impairments, it is vital to accurately identify their hand gestures for both the assessment and rehabilitation of their cognitive function. In recent years, computer vision has gradually been applied to hand gesture recognition. However, it is still challenging to accurately recognize gestures in complex environments, due to insufficient fusion of global and local semantic information, and dynamic hand gestures. To address these problems, this paper proposes a novel gesture recognition model based on the You Only Look Once (YOLO) architecture for patients with cognitive impairment, named YOLO-LG. This model integrates gesture features at different levels by applying a gathering and distribution mechanism, to enhance its multi-scale feature fusion ability. Meanwhile, we reinforce the spatial pyramid pooling module in the YOLO architecture by introducing large separable kernel attention mechanism, to promote the aggregation ability of important features of patient gestures. Furthermore, we adopt a dynamically non-monotonic focusing mechanism and an optimized Wise-IoU loss function, for further elevating the processing ability of complex gesture details. Eventually, we evaluate the performance of our proposed YOLO-LG model by experiments on both the public dataset HaGRID and self-made dataset CGDS collected from patients with cognitive impairment. Extensive experimental results show that the YOLO-LG model outperforms the state-of-the-art gesture recognition model in terms of [email protected] and precision. Specifically, the model increases average [email protected] values with 0.58% and 1.4% respectively on the HaGRID and CGDS datasets, while still maintaining efficient model inference of gesture recognition.
As for patients with cognitive impairments, it is vital to accurately identify their hand gestures for both the assessment and rehabilitation of their cognitive function. In recent years, computer vision has gradually been applied to hand gesture recognition. However, it is still challenging to accurately recognize gestures in complex environments, due to insufficient fusion of global and local semantic information, and dynamic hand gestures. To address these problems, this paper proposes a novel gesture recognition model based on the You Only Look Once (YOLO) architecture for patients with cognitive impairment, named YOLO-LG. This model integrates gesture features at different levels by applying a gathering and distribution mechanism, to enhance its multi-scale feature fusion ability. Meanwhile, we reinforce the spatial pyramid pooling module in the YOLO architecture by introducing large separable kernel attention mechanism, to promote the aggregation ability of important features of patient gestures. Furthermore, we adopt a dynamically non-monotonic focusing mechanism and an optimized Wise-IoU loss function, for further elevating the processing ability of complex gesture details. Eventually, we evaluate the performance of our proposed YOLO-LG model by experiments on both the public dataset HaGRID and self-made dataset CGDS collected from patients with cognitive impairment. Extensive experimental results show that the YOLO-LG model outperforms the state-of-the-art gesture recognition model in terms of [email protected] and precision. Specifically, the model increases average [email protected] values with 0.58% and 1.4% respectively on the HaGRID and CGDS datasets, while still maintaining efficient model inference of gesture recognition.
作者机构：
[Shan Cheng; Kehui Yao; Linxi Guo; Zihui Xu; Hong Tian] School of Energy and Power Engineering, Key Laboratory of Renewable Energy Electric-Technology of Hunan Province, Changsha University of Science and Technology, Changsha 410114, China;State Grid Hunan Electric Power Company Limited Research Institute, Changsha 410208, China;Hunan Key Laboratory of Clean & Efficient Power Generation Technologies, Changsha 410208, China;[Wen Chen] State Grid Hunan Electric Power Company Limited Research Institute, Changsha 410208, China<&wdkj&>Hunan Key Laboratory of Clean & Efficient Power Generation Technologies, Changsha 410208, China
通讯机构：
[Hong Tian] S;School of Energy and Power Engineering, Key Laboratory of Renewable Energy Electric-Technology of Hunan Province, Changsha University of Science and Technology, Changsha 410114, China
摘要：
Pyrolysis of sludge is a promising method for energy and resource recovery from solid waste. However, the emission of odorous gases, especially those containing sulfur and nitrogen, poses significant environmental challenges. Therefore, this study investigated the release characteristics of organic sulfur, when it coexist with organic nitrogen. Focusing on the mechanism by which organic nitrogen affects the transformation pathways of organic sulfur. Co-pyrolysis experiments with organic sulfur model compounds ‘benzyl sulfide (BS) and 4,4′-dihydroxydiphenyl sulfide (DHS)’ and nitrogen-containing model compounds ‘proline (Pro) and aspartic acid (Asp)’. The presence of Pro and Asp lower the pyrolysis temperature and enhance the reaction extent of BS and DHS. The functional groups of organic nitrogen compounds, such as −H, –OH, and −C=O, promoted the production of sulfur-containing gases from organic sulfur compounds. Pro and Asp increase the yield of gas-S by 1.5 ∼ 2 times and 3 times, respectively. Pro also reduced the energy barriers for key steps in H 2 S formation from BS, including the removal of −SH radical from benzyl mercaptan and thiophenol, and −SH hydrogenation, by 83.92 kJ/mol, 39.97 kJ/mol, and 135 kJ/mol, respectively. Asp promoted the cleavage of the C aliphatic -S bond in BS and the C aromatic -S bond in DHS, lowering the energy barriers by 74.05 kJ/mol and 160.27 kJ/mol, respectively. These findings elucidate the role of organic nitrogen compounds in organic sulfur release during sewage sludge pyrolysis, thereby providing a potential way for the synergistic removal of sulfur- and nitrogen-containing odorous gases.
Pyrolysis of sludge is a promising method for energy and resource recovery from solid waste. However, the emission of odorous gases, especially those containing sulfur and nitrogen, poses significant environmental challenges. Therefore, this study investigated the release characteristics of organic sulfur, when it coexist with organic nitrogen. Focusing on the mechanism by which organic nitrogen affects the transformation pathways of organic sulfur. Co-pyrolysis experiments with organic sulfur model compounds ‘benzyl sulfide (BS) and 4,4′-dihydroxydiphenyl sulfide (DHS)’ and nitrogen-containing model compounds ‘proline (Pro) and aspartic acid (Asp)’. The presence of Pro and Asp lower the pyrolysis temperature and enhance the reaction extent of BS and DHS. The functional groups of organic nitrogen compounds, such as −H, –OH, and −C=O, promoted the production of sulfur-containing gases from organic sulfur compounds. Pro and Asp increase the yield of gas-S by 1.5 ∼ 2 times and 3 times, respectively. Pro also reduced the energy barriers for key steps in H 2 S formation from BS, including the removal of −SH radical from benzyl mercaptan and thiophenol, and −SH hydrogenation, by 83.92 kJ/mol, 39.97 kJ/mol, and 135 kJ/mol, respectively. Asp promoted the cleavage of the C aliphatic -S bond in BS and the C aromatic -S bond in DHS, lowering the energy barriers by 74.05 kJ/mol and 160.27 kJ/mol, respectively. These findings elucidate the role of organic nitrogen compounds in organic sulfur release during sewage sludge pyrolysis, thereby providing a potential way for the synergistic removal of sulfur- and nitrogen-containing odorous gases.
期刊：
International Journal of Heat and Fluid Flow,2026年117:110035 ISSN：0142-727X
通讯作者：
Yanfeng Yang
作者机构：
[Chaolin Liu; Chaofan Xiao] School of Energy and Power Engineering, Changsha University of Science and Technology, Changsha 410114, China;[Sanqi Liu] Chengnan College, Changsha University of Science and Technology, Changsha 410015, China;Hebei Key Laboratory of Physics and Energy Technology, Baoding 071003, China;[Yanfeng Yang] School of Energy and Power Engineering, Changsha University of Science and Technology, Changsha 410114, China<&wdkj&>Hebei Key Laboratory of Physics and Energy Technology, Baoding 071003, China
通讯机构：
[Yanfeng Yang] S;School of Energy and Power Engineering, Changsha University of Science and Technology, Changsha 410114, China<&wdkj&>Hebei Key Laboratory of Physics and Energy Technology, Baoding 071003, China
摘要：
In this study, a two-dimensional cylindrical tube array model was established using the finite element method. The effects of sound wave frequency (50–200 Hz) and sound pressure level (118–140 dB) on the flow characteristics in the inter-tube gap under laminar ( Re = 150) and turbulent ( Re = 500) conditions were systematically investigated. The results show that under laminar conditions, the average gap flow velocity increases from 0.025 m/s to 0.048 m/s when the sound pressure level rises from 118 dB to 130 dB at a 50 Hz sound wave, representing a 177 % increase. However, at a constant sound pressure level of 130 dB, the flow velocity of the front row tubes (C1-C7) significantly decreases as the frequency increases from 50 Hz to 200 Hz. Under turbulent conditions, the flow velocity increases linearly by 33 % within the range of 130–140 dB at 50 Hz sound waves. Even though the flow velocity decreases when the frequency increases to 140 dB sound pressure, it is still higher than that without sound waves. The study found that low-frequency sound waves have a more significant effect on enhancing the flow of the front row tubes, while high-frequency sound waves need to consider energy dissipation. Overall, low-frequency and high sound pressure level sound waves can effectively increase the inter-tube flow velocity, and the enhancement effect is more obvious under laminar conditions. This provides a theoretical basis for the optimization of sound wave parameters in engineering applications.
In this study, a two-dimensional cylindrical tube array model was established using the finite element method. The effects of sound wave frequency (50–200 Hz) and sound pressure level (118–140 dB) on the flow characteristics in the inter-tube gap under laminar ( Re = 150) and turbulent ( Re = 500) conditions were systematically investigated. The results show that under laminar conditions, the average gap flow velocity increases from 0.025 m/s to 0.048 m/s when the sound pressure level rises from 118 dB to 130 dB at a 50 Hz sound wave, representing a 177 % increase. However, at a constant sound pressure level of 130 dB, the flow velocity of the front row tubes (C1-C7) significantly decreases as the frequency increases from 50 Hz to 200 Hz. Under turbulent conditions, the flow velocity increases linearly by 33 % within the range of 130–140 dB at 50 Hz sound waves. Even though the flow velocity decreases when the frequency increases to 140 dB sound pressure, it is still higher than that without sound waves. The study found that low-frequency sound waves have a more significant effect on enhancing the flow of the front row tubes, while high-frequency sound waves need to consider energy dissipation. Overall, low-frequency and high sound pressure level sound waves can effectively increase the inter-tube flow velocity, and the enhancement effect is more obvious under laminar conditions. This provides a theoretical basis for the optimization of sound wave parameters in engineering applications.
作者：
Saize Zhang;Jiwei Liu;Fujun Niu*;Tianchun Dong;Xin Pan
期刊：
Cold Regions Science and Technology,2026年241:104683 ISSN：0165-232X
通讯作者：
Fujun Niu
作者机构：
[Saize Zhang; Jiwei Liu] Industry Key Laboratory of Traffic Infrastructure Security Risk Management, School of Civil and Environmental Engineering, Changsha University of Science and Technology, Changsha 410114, China;[Fujun Niu] School of Environmental and Geographical Sciences, Shanghai Normal University, Shanghai 200234, China;[Tianchun Dong] China Railway Qinghai-Tibet Group Co., Ltd., Xining 810000, China;China Communications Fourth Harbor Engineering Bureau Co., Ltd., Guangzhou 510290, China;[Xin Pan] Industry Key Laboratory of Traffic Infrastructure Security Risk Management, School of Civil and Environmental Engineering, Changsha University of Science and Technology, Changsha 410114, China<&wdkj&>China Communications Fourth Harbor Engineering Bureau Co., Ltd., Guangzhou 510290, China
通讯机构：
[Fujun Niu] S;School of Environmental and Geographical Sciences, Shanghai Normal University, Shanghai 200234, China
摘要：
In permafrost regions, due to climate warming and other factors, subgrade deformation may persist for extended periods. Accurately predicting the settlement of frozen soil subgrades is crucial for the stable operation of transportation infrastructure. However, the deformation rules of frozen soil subgrades are diverse, influenced by complex factors, and challenging to monitor. Settlement monitoring data can comprehensively reflect the effects of multiple influencing factors. Against this background, this study focuses solely on historical deformation monitoring data and compares three commonly used types of embankment settlement prediction methods, including curve fitting, grey models, and machine learning approaches. Based on this, a Stacking ensemble algorithm was employed to integrate different categories of prediction models, validated using settlement data from eight monitoring sites, and a Stacking-based model for embankment settlement in permafrost regions was developed and compared with traditional models. The results demonstrate that individual prediction models tend to exhibit inconsistent performance across different monitoring sites and working conditions, often lacking sufficient generalization capability. In contrast, the Stacking Hybrid Ensemble Model effectively leverages the strengths of multiple models, significantly improving overall prediction accuracy while maintaining stable and reliable performance across diverse conditions and locations. This highlights its superior adaptability and generalization ability, underscoring its potential for practical engineering applications in cold-region infrastructure monitoring and maintenance.
In permafrost regions, due to climate warming and other factors, subgrade deformation may persist for extended periods. Accurately predicting the settlement of frozen soil subgrades is crucial for the stable operation of transportation infrastructure. However, the deformation rules of frozen soil subgrades are diverse, influenced by complex factors, and challenging to monitor. Settlement monitoring data can comprehensively reflect the effects of multiple influencing factors. Against this background, this study focuses solely on historical deformation monitoring data and compares three commonly used types of embankment settlement prediction methods, including curve fitting, grey models, and machine learning approaches. Based on this, a Stacking ensemble algorithm was employed to integrate different categories of prediction models, validated using settlement data from eight monitoring sites, and a Stacking-based model for embankment settlement in permafrost regions was developed and compared with traditional models. The results demonstrate that individual prediction models tend to exhibit inconsistent performance across different monitoring sites and working conditions, often lacking sufficient generalization capability. In contrast, the Stacking Hybrid Ensemble Model effectively leverages the strengths of multiple models, significantly improving overall prediction accuracy while maintaining stable and reliable performance across diverse conditions and locations. This highlights its superior adaptability and generalization ability, underscoring its potential for practical engineering applications in cold-region infrastructure monitoring and maintenance.
期刊：
Biomedical Signal Processing and Control,2026年112:108616 ISSN：1746-8094
通讯作者：
Hong Yan
作者机构：
[Gengbiao Chen; Haolong Li] College of Mechanical and Vehicle Engineering, Changsha University of Science and Technology, Changsha 410114, China;[Hong Yan] Department of Stomatology, General Hospital of Northern Theater Command, Shenyang 110016, China
通讯机构：
[Hong Yan] D;Department of Stomatology, General Hospital of Northern Theater Command, Shenyang 110016, China
摘要：
Background Accurate classification of MI from EEG signals is crucial for non-invasive BCIs, especially for individuals with motor impairments. However, existing methods often overlook the synergies across multiple frequency bands, limiting their discriminative power. To address this limitation, we propose DAS-LSTM, a hybrid framework that integrates FBCSP for multi-band feature extraction, a simplified LSTM variant with reduced gating complexity, and a dual attention mechanism that prioritizes task-relevant temporal and spectral features.
Accurate classification of MI from EEG signals is crucial for non-invasive BCIs, especially for individuals with motor impairments. However, existing methods often overlook the synergies across multiple frequency bands, limiting their discriminative power. To address this limitation, we propose DAS-LSTM, a hybrid framework that integrates FBCSP for multi-band feature extraction, a simplified LSTM variant with reduced gating complexity, and a dual attention mechanism that prioritizes task-relevant temporal and spectral features.
Method The proposed model combines the foundational architecture of LSTM networks with an advanced attention mechanism. DAS-LSTM consists of two attention layers and one newly designed LSTM variant. The model’s effectiveness was evaluated using the BCI-IV-2a and BCI-IV-2b datasets. Additionally, ablation experiments were conducted to assess the impact of the newly proposed LSTM variant on the model’s overall performance.
The proposed model combines the foundational architecture of LSTM networks with an advanced attention mechanism. DAS-LSTM consists of two attention layers and one newly designed LSTM variant. The model’s effectiveness was evaluated using the BCI-IV-2a and BCI-IV-2b datasets. Additionally, ablation experiments were conducted to assess the impact of the newly proposed LSTM variant on the model’s overall performance.
Conclusion This study demonstrates the efficacy of the DAS-LSTM framework in accurately classifying motor imagery signals across different EEG categories. The findings highlight the potential of the proposed model to contribute to the development of more intuitive and natural prosthetic control systems.
This study demonstrates the efficacy of the DAS-LSTM framework in accurately classifying motor imagery signals across different EEG categories. The findings highlight the potential of the proposed model to contribute to the development of more intuitive and natural prosthetic control systems.
作者机构：
[Binbin Chen; Hong Tian; Zhen Zhou; Yanni Xuan; Siying Liu] School of Energy and Power Engineering, Changsha University of Science and Technology, Changsha 410114, China;[Zhijie Wang] Hunan Province Key Laboratory of Efficient and Clean Power Generation Technologies, Changsha 410007, China
通讯机构：
[Hong Tian; Yanni Xuan] S;School of Energy and Power Engineering, Changsha University of Science and Technology, Changsha 410114, China
摘要：
A combined pretreatment of straw was carried out using acid washing and torrefaction methods. Metal-modified HZSM-5 core–shell molecular sieves were prepared by loading Zn (2, 4, and 6 wt%) and Ni (6, 8, and 10 wt%) on HZSM-5 molecular sieves and introducing MCM-41 core–shell structure. PY-GC/MS and a tubular furnace were employed to study the effects of pretreatment conditions and catalysts on the product composition distribution during wheat straw catalytic pyrolysis. XRD, SEM, BET, TPD and ICP were used to characterize the catalyst performance. It was found that the combined acid washing and torrefaction pretreatment reduced the oxygenated compounds in the bio-oil from straw catalytic pyrolysis and increased the bio-oil yield to 29.37 %. The incorporation of modified catalysts promoted the deoxygenation, zwitterionization and aromatization reactions during the straw-catalyzed pyrolysis. The monometallic loading of 4%Zn/HZ and 8%Ni/HZ catalyzed acid washing and torrefaction straw pyrolysis resulted in 54.2 % and 57.06 % yields of MAHs and 42.86 % and 38.58 % yields of BTX in bio-oil, respectively.Compared with the monometallic loading, 4%Zn8%Ni/HZ further optimized the bio-oil compositional distribution, with a MAHs yield of 64.76 %, a BTX yield of 54.69 %, and a deoxygenation performance of 81.34%.MCM-41-coated HZSM-5 produces a large mesoporous structure with channels with sufficient transport capacity, accelerating the cleavage of various types of oxygen-containing compounds in the bio-oil into smaller molecules for better conversion into aromatics. The 4%Zn8%Ni/H@M catalyst achieved a MAHs yield of 72.68 % and BTX yield of 63.43 % in the bio-oil during pyrolysis of acid-washed and torrefied wheat straw, with oxygen removal efficiency reaching 85.37 %. Therefore, the combination of feedstock pretreatment and metal-modified core–shell HZSM-5 molecular sieve could synergistically optimize both compositional distribution and production yield of bio-oil derived from biomass pyrolysis.
A combined pretreatment of straw was carried out using acid washing and torrefaction methods. Metal-modified HZSM-5 core–shell molecular sieves were prepared by loading Zn (2, 4, and 6 wt%) and Ni (6, 8, and 10 wt%) on HZSM-5 molecular sieves and introducing MCM-41 core–shell structure. PY-GC/MS and a tubular furnace were employed to study the effects of pretreatment conditions and catalysts on the product composition distribution during wheat straw catalytic pyrolysis. XRD, SEM, BET, TPD and ICP were used to characterize the catalyst performance. It was found that the combined acid washing and torrefaction pretreatment reduced the oxygenated compounds in the bio-oil from straw catalytic pyrolysis and increased the bio-oil yield to 29.37 %. The incorporation of modified catalysts promoted the deoxygenation, zwitterionization and aromatization reactions during the straw-catalyzed pyrolysis. The monometallic loading of 4%Zn/HZ and 8%Ni/HZ catalyzed acid washing and torrefaction straw pyrolysis resulted in 54.2 % and 57.06 % yields of MAHs and 42.86 % and 38.58 % yields of BTX in bio-oil, respectively.Compared with the monometallic loading, 4%Zn8%Ni/HZ further optimized the bio-oil compositional distribution, with a MAHs yield of 64.76 %, a BTX yield of 54.69 %, and a deoxygenation performance of 81.34%.MCM-41-coated HZSM-5 produces a large mesoporous structure with channels with sufficient transport capacity, accelerating the cleavage of various types of oxygen-containing compounds in the bio-oil into smaller molecules for better conversion into aromatics. The 4%Zn8%Ni/H@M catalyst achieved a MAHs yield of 72.68 % and BTX yield of 63.43 % in the bio-oil during pyrolysis of acid-washed and torrefied wheat straw, with oxygen removal efficiency reaching 85.37 %. Therefore, the combination of feedstock pretreatment and metal-modified core–shell HZSM-5 molecular sieve could synergistically optimize both compositional distribution and production yield of bio-oil derived from biomass pyrolysis.
作者机构：
[Chen Li; Jianxin Li; Furong Wan] Department of Business Administration, School of Economics and Management, Changsha University of Science and Technology, Changsha, 410076, China;[Shibin Sheng] Collat School of Business, University of Alabama at Birmingham, Birmingham, AL, 35294, USA;[Tengqi Yi] School of Information Engineering and Business, Changsha Vocational and Technical College, Changsha, 410217, China
通讯机构：
[Shibin Sheng] C;Collat School of Business, University of Alabama at Birmingham, Birmingham, AL, 35294, USA
关键词：
Domestic tourism;Destination marketing;Celebrity endorsement;Celebrity identity;Meaning transfer model;Celebrity-destination relationship
摘要：
In celebrity endorsement of tourism destinations, celebrity identity constitutes and conveys multilayered symbolic meanings to tourists. However, prior studies often overlook how these meanings emerge through the contextual and interactive construction of identity. Drawing on the Meaning Transfer Model, we elucidate how these symbolic meanings are constituted and transferred along the ascription and interaction dimensions of celebrity-destination relationships. The findings from four experiments with 2758 participants reveal that the interaction dimension plays a pivotal role in shaping identity salience and tourist responses via endorsement authenticity, contingent on destination types. This study contributes to the literature of celebrity endorsement by incorporating a dual-dimensional identity matrix and offering a nuanced understanding of the mechanisms through which celebrity identity shapes tourists' responses to the endorsement. We also introduce the FOCUS framework as managerial guidelines. Future research is warranted to assess its cross-cultural validity in domestic tourism and its applicability to international contexts.
In celebrity endorsement of tourism destinations, celebrity identity constitutes and conveys multilayered symbolic meanings to tourists. However, prior studies often overlook how these meanings emerge through the contextual and interactive construction of identity. Drawing on the Meaning Transfer Model, we elucidate how these symbolic meanings are constituted and transferred along the ascription and interaction dimensions of celebrity-destination relationships. The findings from four experiments with 2758 participants reveal that the interaction dimension plays a pivotal role in shaping identity salience and tourist responses via endorsement authenticity, contingent on destination types. This study contributes to the literature of celebrity endorsement by incorporating a dual-dimensional identity matrix and offering a nuanced understanding of the mechanisms through which celebrity identity shapes tourists' responses to the endorsement. We also introduce the FOCUS framework as managerial guidelines. Future research is warranted to assess its cross-cultural validity in domestic tourism and its applicability to international contexts.
摘要：
In comparison to the classical liquid-contact ion-selective electrode (LC-ISE), the potential stability remains one of the major challenges currently faced by the solid-contact ion-selective electrode (SC-ISE). This challenge can be addressed by introducing a conducting layer between the conductive substrate and the polymer membrane to stabilize the potential output. In this paper, hollow carbon nanospheres loaded with zinc cobaltate nanoparticles (ZnCo 2 O 4 @HCNs) were synthesized by using a one-pot method, which served as an ion-electron conducting layer between the potassium ion-selective polymer membrane and the conductive glassy carbon substrate. Electrochemical impedance spectroscopy (EIS) and chronopotentiometry tests demonstrated a high capacitance and excellent potential stability for the solid-contact potassium ion-selective electrode (K + /SC-ISE). Water layer tests further confirmed that ZnCo 2 O 4 @HCNs exhibited strong hydrophobicity, effectively preventing the formation of water contact layers at the electrode/membrane interface, thereby ensuring a highly stable signal-response. Under the optimized conditions, the prepared K + /SC-ISE showed a good linear response in the concentration range of 1.0 × 10 −6 –1.0 × 10 −2 mol/L with a Nernst slope of 56.03 ± 0.55 mV/decade. The detection limit was calculated to be 6.16 × 10 −7 mol/L. Additionally, the electrode demonstrated excellent selectivity, stability, and a long lifetime of over 24 months. Finally, the electrode was successfully applied to the on-site detection of K + in actual cigarette paper samples at the factory workshop, yielding results consistent with those using ion chromatography, highlighting its broad application prospect.
In comparison to the classical liquid-contact ion-selective electrode (LC-ISE), the potential stability remains one of the major challenges currently faced by the solid-contact ion-selective electrode (SC-ISE). This challenge can be addressed by introducing a conducting layer between the conductive substrate and the polymer membrane to stabilize the potential output. In this paper, hollow carbon nanospheres loaded with zinc cobaltate nanoparticles (ZnCo 2 O 4 @HCNs) were synthesized by using a one-pot method, which served as an ion-electron conducting layer between the potassium ion-selective polymer membrane and the conductive glassy carbon substrate. Electrochemical impedance spectroscopy (EIS) and chronopotentiometry tests demonstrated a high capacitance and excellent potential stability for the solid-contact potassium ion-selective electrode (K + /SC-ISE). Water layer tests further confirmed that ZnCo 2 O 4 @HCNs exhibited strong hydrophobicity, effectively preventing the formation of water contact layers at the electrode/membrane interface, thereby ensuring a highly stable signal-response. Under the optimized conditions, the prepared K + /SC-ISE showed a good linear response in the concentration range of 1.0 × 10 −6 –1.0 × 10 −2 mol/L with a Nernst slope of 56.03 ± 0.55 mV/decade. The detection limit was calculated to be 6.16 × 10 −7 mol/L. Additionally, the electrode demonstrated excellent selectivity, stability, and a long lifetime of over 24 months. Finally, the electrode was successfully applied to the on-site detection of K + in actual cigarette paper samples at the factory workshop, yielding results consistent with those using ion chromatography, highlighting its broad application prospect.
期刊：
Electric Power Systems Research,2026年250:112102 ISSN：0378-7796
通讯作者：
Yong Li
作者机构：
College of Electrical and Information Engineering, Hunan University, Changsha, 410082, China;Greater Bay Area Institute for Innovation, Hunan University, Guangzhou, 511300, China;[Yijia Cao] School of Electrical & Information Engineering, Changsha University of Science and Technology, Changsha, 410114, China;[Fang Wu; Rui Li; Jiuqing Cai] Wuhan second ship design and research institute, Wuhan, 44227, China;[Yinglong Zhao; Yong Li; Sijia Hu] College of Electrical and Information Engineering, Hunan University, Changsha, 410082, China<&wdkj&>Greater Bay Area Institute for Innovation, Hunan University, Guangzhou, 511300, China
通讯机构：
[Yong Li] C;College of Electrical and Information Engineering, Hunan University, Changsha, 410082, China<&wdkj&>Greater Bay Area Institute for Innovation, Hunan University, Guangzhou, 511300, China
关键词：
Electrified ship;Electric power system;Transformer design;Vibration reduction;Finite element model;Multiple optimization
摘要：
The optimization of transformer vibration and power density in marine electrical power systems poses a challenging task due to the constraints imposed by ship noise and limited space. This paper introduces a multi-optimization design method for transformers based on multi-objective optimization model and finite element model, with the objective of minimizing vibration and improving power density. The proposed approach leverages the merits of both methodologies by initially utilizing a multi-objective optimization technique to attain an optimal transformer preliminary design, featuring optimized volume, loss, and vibration acceleration. Subsequently, based on this preliminary design, a finite element model is constructed to further refine the transformer’s placement configuration and thermal limits, ultimately yielding an optimal design scheme for a transformer that boasts both low vibration and high power density. Experimental results demonstrate that the proposed method effectively reducing transformer vibrations and volume. Compared to previous-generation transformers not utilizing this method, the proposed approach leads to a 55.81% reduction in vibrational acceleration and 44.93% reduction in volume. Additionally, the calculation values of the transformer from the proposed method exhibit high precision compared to actual measurements.
The optimization of transformer vibration and power density in marine electrical power systems poses a challenging task due to the constraints imposed by ship noise and limited space. This paper introduces a multi-optimization design method for transformers based on multi-objective optimization model and finite element model, with the objective of minimizing vibration and improving power density. The proposed approach leverages the merits of both methodologies by initially utilizing a multi-objective optimization technique to attain an optimal transformer preliminary design, featuring optimized volume, loss, and vibration acceleration. Subsequently, based on this preliminary design, a finite element model is constructed to further refine the transformer’s placement configuration and thermal limits, ultimately yielding an optimal design scheme for a transformer that boasts both low vibration and high power density. Experimental results demonstrate that the proposed method effectively reducing transformer vibrations and volume. Compared to previous-generation transformers not utilizing this method, the proposed approach leads to a 55.81% reduction in vibrational acceleration and 44.93% reduction in volume. Additionally, the calculation values of the transformer from the proposed method exhibit high precision compared to actual measurements.
摘要：
Porcelain insulator is an important component of power transmission systems, and its condition detection is essential to ensure safe operation of the power grid. Nevertheless, it is difficult for existing detection models to effectively solve the contradiction between detection accuracy and resource consumption. To address this issue, a high-precision lightweight insulator defect detection model (BCM-YOLO) based on an improved YOLOv8 is proposed. Firstly, bidirectional feature pyramid network (BiFPN), with a simplified bidirectional information flow mechanism, is employed to replace the path aggregation network with feature pyramid network in YOLOv8 to alter the feature fusion mode, thereby reducing the model size. Secondly, a cross-stage partial Bottleneck with 2 convolutions partially replaced by a context-guided block (C2f_CG) structure with parameter sharing is designed using the improved context-guided block to optimise the cross-stage partial Bottleneck with 2 convolutions (C2f) modules, thus further decreasing the number of model parameters. Finally, multiscale dilated attention is introduced into the BiFPN network to enhance the perception ability of different scales of features to improve the detection performance. Experimental results indicate that compared to YOLOv8s, the BCM-YOLO model reduces the number of parameters by 50.5%, lowers floating-point operations by 31.3% and increases mean average precision at intersection over union = 0.5 (mAP0.5) by 2.8%. The proposed model not only improves detection accuracy but also decreases parameter counts, making it more suitable for deployment on edge devices.
摘要：
Compared with traditional electronic devices, spintronic devices offer the benefits of lower power consumption, faster transmission speeds, and higher integration densities. Therefore, seeking a direct and efficient method to flexibly generate spin currents in a single device, especially fully spin-polarized current (FSPC) and pure spin current (PSC), remains crucial. Inspired by this, we design a spin optoelectronic device based on the half-metal YSi 2 N 4 , and investigate its transport behavior influenced by photogalvanic effects. Remarkably, the YSi 2 N 4 spin device can generate FSPC and PSC under linearly polarized light irradiation with parallel electrode magnetic configuration (PC) or anti-parallel configuration (APC). For the YSi 2 N 4 device in PC, FSPC can be generated at any polarization angle when the photon energy is less than 2.36 eV, fulfilling the spin filtering effect. For the device in APC, PSC can be obtained across an extensive range of photon energy, which acts as an important carrier for spin transport. More interestingly, the magnetoresistance ratio of the YSi 2 N 4 device generally exceeds 90%, demonstrating excellent spin-valve effect. Our work suggests that the monolayer YSi 2 N 4 spin optoelectronic device can generate FSPC and PSC flexibly and efficiently, making it an advanced candidate for multifunctional spin devices.
摘要：
In order to establish an efficient microbial transformation platform based on seaweed feedstocks, experiments were performed to isolate a salt-tolerant strain capable of producing alginate lyase and 2,3-butanediol (2,3-BDO). Its physiological and biochemical characteristics, carbon source utilization, and product synthesis capabilities were investigated, and then the process for co-producing alginate lyase and 2,3-BDO from Laminaria japonica was optimized. Results showed that the isolated strain was identified as Vibrio alginolyticus, which was capable of utilizing multiple carbon sources to produce alginate lyase and 2,3-BDO even in the presence of 5 % NaCl. The highest reducing sugar yield was achieved as the Laminaria japonica pretreated with 1 % (v/v) sulfuric acid at 120 °C for 18 min. The enzymatic hydrolysis was boosted by devising a novel tween 80-assisted enzyme complex containing 60 FPU/g of cellulase, 15 U/g of pectinase, 20 U/g of alginate lyase, and 90 mg/g of tween 80. After establishing a semi-simultaneous saccharification and fermentation (S-SSF) strategy, the sugars could be fully utilized, yielding 14.83 g/L 2,3-BDO and 11.02 kU/L alginate lyase, respectively. Mass balance calculations indicating that up to 140 g of 2,3-butanediol and 120 kU of alginate lyase can be obtained from per kg of Laminaria japonica via this integrated approach.
In order to establish an efficient microbial transformation platform based on seaweed feedstocks, experiments were performed to isolate a salt-tolerant strain capable of producing alginate lyase and 2,3-butanediol (2,3-BDO). Its physiological and biochemical characteristics, carbon source utilization, and product synthesis capabilities were investigated, and then the process for co-producing alginate lyase and 2,3-BDO from Laminaria japonica was optimized. Results showed that the isolated strain was identified as Vibrio alginolyticus, which was capable of utilizing multiple carbon sources to produce alginate lyase and 2,3-BDO even in the presence of 5 % NaCl. The highest reducing sugar yield was achieved as the Laminaria japonica pretreated with 1 % (v/v) sulfuric acid at 120 °C for 18 min. The enzymatic hydrolysis was boosted by devising a novel tween 80-assisted enzyme complex containing 60 FPU/g of cellulase, 15 U/g of pectinase, 20 U/g of alginate lyase, and 90 mg/g of tween 80. After establishing a semi-simultaneous saccharification and fermentation (S-SSF) strategy, the sugars could be fully utilized, yielding 14.83 g/L 2,3-BDO and 11.02 kU/L alginate lyase, respectively. Mass balance calculations indicating that up to 140 g of 2,3-butanediol and 120 kU of alginate lyase can be obtained from per kg of Laminaria japonica via this integrated approach.
作者机构：
[Li, Xinzhuo; Tian, Hong; Sun, Liutao; Dai, Pengfei; Xu, Chenghui; Huang, Zhangjun; Li, XZ] Changsha Univ Sci & Technol, Sch Energy & Power Engn, Changsha 410114, Peoples R China.
通讯机构：
[Li, XZ ] C;Changsha Univ Sci & Technol, Sch Energy & Power Engn, Changsha 410114, Peoples R China.
关键词：
Ammonia/methane mixture combustion;Mechanism reduction;Mechanism optimization;Artificial Neural Network;NO x emissions
摘要：
The application of ammonia/methane (NH 3 /CH 4 ) blended fuels in gas turbines has received considerable attention, and the development of their combustors requires the implementation of more precise and compact reaction mechanisms. In this work, we propose a new optimization mechanism for ammonia/methane and comprehensively verify the performance of the optimization mechanism. A detailed chemical mechanism with 65 species and 466 reactions (Detailed-Mech) was first assembled using models from the literature. A directed relation graph with error propagation (DRGEP) and computational singular perturbation (CSP) method were then used to obtain a 23-species, 73-reaction compact reaction model (Reduced-Mech). Finally, the pre-exponential factor ( A ) and activation energy ( E a ) of five significant elementary reactions were optimized using an Artificial Neural Network (ANN) to obtain the optimized mechanism (ANN-Mech). The ANN-Mech was validated at ignition delay times (IDT), laminar burning velocity (LBV), plug flow reactor (PFR) species distribution, and in the 3-D combustion chamber. The study found that the logarithmic mean errors of IDT decreased by 3.9 %. The mean error of laminar burning velocity is reduced from 18.5 % to 9.5 %, and the prediction error of NO X in ANN-Mech is reduced by 47.5 %. The results of the premixed flames simulation indicate that the temperature and velocity fields of ANN-Mech at different ammonia fractions better agree with the Detailed-Mech. Additionally, the NO error of the outlet was reduced by 30 %. The calculation speed was also increased by ten times compared to the Detailed-Mech.
The application of ammonia/methane (NH 3 /CH 4 ) blended fuels in gas turbines has received considerable attention, and the development of their combustors requires the implementation of more precise and compact reaction mechanisms. In this work, we propose a new optimization mechanism for ammonia/methane and comprehensively verify the performance of the optimization mechanism. A detailed chemical mechanism with 65 species and 466 reactions (Detailed-Mech) was first assembled using models from the literature. A directed relation graph with error propagation (DRGEP) and computational singular perturbation (CSP) method were then used to obtain a 23-species, 73-reaction compact reaction model (Reduced-Mech). Finally, the pre-exponential factor ( A ) and activation energy ( E a ) of five significant elementary reactions were optimized using an Artificial Neural Network (ANN) to obtain the optimized mechanism (ANN-Mech). The ANN-Mech was validated at ignition delay times (IDT), laminar burning velocity (LBV), plug flow reactor (PFR) species distribution, and in the 3-D combustion chamber. The study found that the logarithmic mean errors of IDT decreased by 3.9 %. The mean error of laminar burning velocity is reduced from 18.5 % to 9.5 %, and the prediction error of NO X in ANN-Mech is reduced by 47.5 %. The results of the premixed flames simulation indicate that the temperature and velocity fields of ANN-Mech at different ammonia fractions better agree with the Detailed-Mech. Additionally, the NO error of the outlet was reduced by 30 %. The calculation speed was also increased by ten times compared to the Detailed-Mech.
作者机构：
National Engineering Research Center of Highway Maintenance Technology, Changsha University of Science & Technology, Changsha, HN 410114, China;[Liyan Liu] School of Transportation, Changsha University of Science & Technology, Changsha, HN 410114, China;[Shirong Zhou; Zhong Zhou] School of Civil Engineering, Central South University, Changsha, HN 410075, China;[Hao Yang] National Engineering Research Center of Highway Maintenance Technology, Changsha University of Science & Technology, Changsha, HN 410114, China<&wdkj&>School of Transportation, Changsha University of Science & Technology, Changsha, HN 410114, China
通讯机构：
[Shirong Zhou; Zhong Zhou] S;School of Civil Engineering, Central South University, Changsha, HN 410075, China
摘要：
Defects in tunnel linings accelerate structural deterioration, reduce service life, and pose serious safety risks. Existing algorithms for detecting defect signals in ground-penetrating radar (GPR) images often struggle to balance accuracy and efficiency, with limited capacity to extract meaningful features. To address these limitations, this paper proposes a lightweight algorithm, MGD-DETR, for accurate recognition of internal tunnel lining defects, using RT-DETR as the base model. First, a Multi-HGNet backbone feature extraction network is introduced to reduce model size (MS) and enhance dynamic fusion and interaction between feature layers, thereby improving feature extraction. Second, the lightweight convolution module GSConv replaces standard convolution operations to reduce the parameter count. Third, a dual attention module (DAM) is integrated to dynamically adjust spatial and channel feature weights, improving the model’s generalization performance. Five models—RT-DETR, YOLO-LD, YOLOv10, YOLOv11, and SSD—were used for comparative evaluation. Experimental results show that MGD-DETR outperforms the other models across all metrics, achieving a mean average precision (mAP) of 0.834, mean F1 score (mF1) of 0.818, MS of 26.9 M, and frames per second (FPS) of 91.2f/s, enabling fast and accurate recognition of defect signals and facilitate subsequent deployment into tunnel detection mobile devices.
Defects in tunnel linings accelerate structural deterioration, reduce service life, and pose serious safety risks. Existing algorithms for detecting defect signals in ground-penetrating radar (GPR) images often struggle to balance accuracy and efficiency, with limited capacity to extract meaningful features. To address these limitations, this paper proposes a lightweight algorithm, MGD-DETR, for accurate recognition of internal tunnel lining defects, using RT-DETR as the base model. First, a Multi-HGNet backbone feature extraction network is introduced to reduce model size (MS) and enhance dynamic fusion and interaction between feature layers, thereby improving feature extraction. Second, the lightweight convolution module GSConv replaces standard convolution operations to reduce the parameter count. Third, a dual attention module (DAM) is integrated to dynamically adjust spatial and channel feature weights, improving the model’s generalization performance. Five models—RT-DETR, YOLO-LD, YOLOv10, YOLOv11, and SSD—were used for comparative evaluation. Experimental results show that MGD-DETR outperforms the other models across all metrics, achieving a mean average precision (mAP) of 0.834, mean F1 score (mF1) of 0.818, MS of 26.9 M, and frames per second (FPS) of 91.2f/s, enabling fast and accurate recognition of defect signals and facilitate subsequent deployment into tunnel detection mobile devices.
摘要：
Single-atom nanozymes (SAzymes) with well-defined metal–nitrogen–carbon coordination structures are of great interest for the development of colorimetric biosensing. However, their catalytic efficiency and specificity is restricted due to the limited number of single metal atoms that can serve as catalytic centre. Therefore, the construction of SAzymes with high activity and specificity is vital but remains challenging. To address these issues, we prepared a bimetallic SAzymes with an independent iron and cobalt structure (FeCo/NC), and the oxidase-like activity was enhanced by >112.8% relative to Fe/NC. This preparation strategy increased the amount of single metal atoms loaded, resulting in a strong synergistic effect and proximity-orientation effects due to the unique structure of single Co and Fe atoms coexisting on graphene. The oxidase-mimicking activity of FeCo/NC was specifically enhanced by co doping, although the activities of peroxidase-, superoxide dismutase-, or catalase-like were not significantly affected. In light of these discoveries, as a symbol of the proof-of-concept, FeCo/NC-based colorimetric immunoassays were developed for sensitive detection of aflatoxin B1 (AFB 1 ), achieving a linear range of 0.01−10 ng/mL and a detection limit of 0.005 ng/mL. This study provides a convenient strategy for promoting the catalytic activity and specificity of SAzymes, thereby enhancing their potential in biosensing.
Single-atom nanozymes (SAzymes) with well-defined metal–nitrogen–carbon coordination structures are of great interest for the development of colorimetric biosensing. However, their catalytic efficiency and specificity is restricted due to the limited number of single metal atoms that can serve as catalytic centre. Therefore, the construction of SAzymes with high activity and specificity is vital but remains challenging. To address these issues, we prepared a bimetallic SAzymes with an independent iron and cobalt structure (FeCo/NC), and the oxidase-like activity was enhanced by >112.8% relative to Fe/NC. This preparation strategy increased the amount of single metal atoms loaded, resulting in a strong synergistic effect and proximity-orientation effects due to the unique structure of single Co and Fe atoms coexisting on graphene. The oxidase-mimicking activity of FeCo/NC was specifically enhanced by co doping, although the activities of peroxidase-, superoxide dismutase-, or catalase-like were not significantly affected. In light of these discoveries, as a symbol of the proof-of-concept, FeCo/NC-based colorimetric immunoassays were developed for sensitive detection of aflatoxin B1 (AFB 1 ), achieving a linear range of 0.01−10 ng/mL and a detection limit of 0.005 ng/mL. This study provides a convenient strategy for promoting the catalytic activity and specificity of SAzymes, thereby enhancing their potential in biosensing.
期刊：
Construction and Building Materials,2025年489:142267 ISSN：0950-0618
通讯作者：
Chaochao Liu
作者机构：
National Engineering Research Center of Highway Maintenance Technology, Changsha University of Science & Technology, Changsha, Hunan 410114, China;Xiangjiang Laboratory, Changsha University of Science & Technology, Changsha, Hunan 410114, China;Guangxi Key Laboratory of Road Structure and Materials, Guangxi Transportation Science and Technology Group Co., Ltd,, Nanning, Guangxi 530007, China;[Mengjie Chen; Chaochao Liu; Zhiyu Yang; Yanhua Xue; Songtao Lv] National Engineering Research Center of Highway Maintenance Technology, Changsha University of Science & Technology, Changsha, Hunan 410114, China<&wdkj&>Xiangjiang Laboratory, Changsha University of Science & Technology, Changsha, Hunan 410114, China;[Jie Chen] National Engineering Research Center of Highway Maintenance Technology, Changsha University of Science & Technology, Changsha, Hunan 410114, China<&wdkj&>Guangxi Key Laboratory of Road Structure and Materials, Guangxi Transportation Science and Technology Group Co., Ltd,, Nanning, Guangxi 530007, China
通讯机构：
[Chaochao Liu] N;National Engineering Research Center of Highway Maintenance Technology, Changsha University of Science & Technology, Changsha, Hunan 410114, China<&wdkj&>Xiangjiang Laboratory, Changsha University of Science & Technology, Changsha, Hunan 410114, China
摘要：
Polymer modified asphalt requires prolonged shear mixing and swelling to ensure adequate polymer dispersion and interaction. In order to alleviate the performance degradation caused by thermal oxidative aging during the preparation process, a closed vacuum shear (CVS) device was developed. The process was compared with the conventional open shear (TOS) process to evaluate the performance differences. The rheological and fatigue properties of asphalt binders prepared using CVS were evaluated by temperature sweep (TS), frequency sweep, linear amplitude sweep (LAS), and bending beam rheometer (BBR) tests. Fourier transform infrared spectroscopy (FTIR), atomic force microscopy (AFM), and scanning electron microscopy (SEM) were used to examine the microstructural changes resulting from different preparation processes. The results showed that the J nr values of SBR, SBS and rubber modified asphalt under CVS decreased by 9.18 %, 16.98 % and 12.93 %, respectively, compared with TOS. Correspondingly, the S value decreased by 19.79 %, 30.46 % and 17.93 %, and the m value decreased by 2.2 %, 5.1 % and 18.0 %. At a strain level of 5 %, the fatigue life increased by 21.95 %, 31.97 % and 6.67 %, respectively. The CVS process significantly improves the high-temperature plasticity resistance, low-temperature crack resistance and fatigue durability of the modified asphalt binder. Microstructural analysis showed that the CVS process eliminated air bubbles entrapped during mixing, resulting in a more uniform and smooth adhesive surface. In addition, the Ic=o index and Is=o index of the CVS modified asphalt decreased, and the Ic=c index increased, further indicating that the degree of aging was reduced. The CVS process can inhibit oxidative aging during the preparation of the binder. It is worth noting that SBS modified asphalt is most sensitive to the preparation process. Therefore, it is recommended to use the CVS process to process SBS modified binder to better maintain the integrity of the SBS polymer network.
Polymer modified asphalt requires prolonged shear mixing and swelling to ensure adequate polymer dispersion and interaction. In order to alleviate the performance degradation caused by thermal oxidative aging during the preparation process, a closed vacuum shear (CVS) device was developed. The process was compared with the conventional open shear (TOS) process to evaluate the performance differences. The rheological and fatigue properties of asphalt binders prepared using CVS were evaluated by temperature sweep (TS), frequency sweep, linear amplitude sweep (LAS), and bending beam rheometer (BBR) tests. Fourier transform infrared spectroscopy (FTIR), atomic force microscopy (AFM), and scanning electron microscopy (SEM) were used to examine the microstructural changes resulting from different preparation processes. The results showed that the J nr values of SBR, SBS and rubber modified asphalt under CVS decreased by 9.18 %, 16.98 % and 12.93 %, respectively, compared with TOS. Correspondingly, the S value decreased by 19.79 %, 30.46 % and 17.93 %, and the m value decreased by 2.2 %, 5.1 % and 18.0 %. At a strain level of 5 %, the fatigue life increased by 21.95 %, 31.97 % and 6.67 %, respectively. The CVS process significantly improves the high-temperature plasticity resistance, low-temperature crack resistance and fatigue durability of the modified asphalt binder. Microstructural analysis showed that the CVS process eliminated air bubbles entrapped during mixing, resulting in a more uniform and smooth adhesive surface. In addition, the Ic=o index and Is=o index of the CVS modified asphalt decreased, and the Ic=c index increased, further indicating that the degree of aging was reduced. The CVS process can inhibit oxidative aging during the preparation of the binder. It is worth noting that SBS modified asphalt is most sensitive to the preparation process. Therefore, it is recommended to use the CVS process to process SBS modified binder to better maintain the integrity of the SBS polymer network.
