期刊：
Journal of Tribology,2026年148(1):011701 ISSN：0742-4787
通讯作者：
Hua-Xi Zhou
作者机构：
[Jing-Lun Xie; Chang-Guang Zhou] Department of Mechanical Engineering, Nanjing University of Science and Technology, Nanjing 210094, China;[Xiao-Yi Wang; Hua-Xi Zhou; Yi Zhang] Department of Mechanical Engineering, Wuhan Polytechnic University, Wuhan 430048, China
通讯机构：
[Hua-Xi Zhou] D;Department of Mechanical Engineering, Wuhan Polytechnic University, Wuhan 430048, China
通讯机构：
[Chen, Y ] W;Wuhan Polytech Univ, Coll Mech Engn, Wuhan 430048, Hubei, Peoples R China.
关键词：
3D point cloud;Genetic algorithm-based wavelet neural network;Mean absolute percentage error;Poultry viscera;Root mean square error
摘要：
In order to avoid damaging viscera during poultry evisceration and enhance the economic value of poultry products, this paper proposes a predictive method for poultry carcass visceral dimensions based on 3D point cloud and a Genetic Algorithm-based Wavelet Neural Network (GA-WNN). In this study, a data set of poultry carcasses was obtained through the use of 3D point cloud scanning equipment combined with reverse engineering software. The inputs and predicted targets of the model were determined through correlation analysis of various carcass dimensions. Then, a prediction model of poultry visceral size (GA-WNN) was built by K-fold cross validation method, Genetic Algorithm and Wavelet Neural Network (WNN). By comparing the prediction results and analyzing Mean Absolute Percentage Error (MAPE) and Root Mean Square Error (RMSE) of the six models, it was determined that the GA-WNN model had the best prediction results. Finally, in order to verify the generalizability of the method, generalizability experiments were conducted on different breeds of poultry, which proved that the method of this study had superior generalizability ability. In the comparative analysis of the six models, the MAPE and RMSE of the GA-WNN model for the prediction of the three visceral dimensions were the lowest except for the RMSE for the prediction of visceral length. Compared with the largest of the two kinds of errors, the MAPE and RMSE for the prediction of the position of the upper end of the left liver by the method of this study were lower by 5.56% and 0.915 cm, respectively, and the prediction effect had a significant advantage. The experimental results showed that the model built in this paper based on 3D point cloud and GA-WNN network can accurately predict the size of the viscera of poultry carcasses, thus providing theoretical references for the automated evisceration technology without damaging the viscera.
In order to avoid damaging viscera during poultry evisceration and enhance the economic value of poultry products, this paper proposes a predictive method for poultry carcass visceral dimensions based on 3D point cloud and a Genetic Algorithm-based Wavelet Neural Network (GA-WNN). In this study, a data set of poultry carcasses was obtained through the use of 3D point cloud scanning equipment combined with reverse engineering software. The inputs and predicted targets of the model were determined through correlation analysis of various carcass dimensions. Then, a prediction model of poultry visceral size (GA-WNN) was built by K-fold cross validation method, Genetic Algorithm and Wavelet Neural Network (WNN). By comparing the prediction results and analyzing Mean Absolute Percentage Error (MAPE) and Root Mean Square Error (RMSE) of the six models, it was determined that the GA-WNN model had the best prediction results. Finally, in order to verify the generalizability of the method, generalizability experiments were conducted on different breeds of poultry, which proved that the method of this study had superior generalizability ability. In the comparative analysis of the six models, the MAPE and RMSE of the GA-WNN model for the prediction of the three visceral dimensions were the lowest except for the RMSE for the prediction of visceral length. Compared with the largest of the two kinds of errors, the MAPE and RMSE for the prediction of the position of the upper end of the left liver by the method of this study were lower by 5.56% and 0.915 cm, respectively, and the prediction effect had a significant advantage. The experimental results showed that the model built in this paper based on 3D point cloud and GA-WNN network can accurately predict the size of the viscera of poultry carcasses, thus providing theoretical references for the automated evisceration technology without damaging the viscera.
作者机构：
[Tahir, Muhammad; Tahir, M; Ahmad, Waheed; Dai, Jun; Dai, J] Beijing Inst Technol, Sch Mechatron Engn, Beijing 100081, Peoples R China.;[Bibi, Batoul] Ghazi Univ, Dept Chem, DG Khan 32200, Pakistan.;[Peng, Zhen; He, L; Xiong, Yibo; Khan, Arif Ullah; He, Liang; Ul Nisa, Fazal; Ma, Zeyu; Naseem, Mizna] Sichuan Univ, Sch Mech Engn, State Key Lab Intelligent Construct & Hlth Operat, Chengdu 610065, Peoples R China.;[Gong, Fengming] Sichuan Univ, West China Hosp 2, Dept Gynecol & Obstet, Dev & Related Dis Women & Children Key Lab Sichuan, Chengdu 610041, Peoples R China.;[Tang, Hui] Univ Elect Sci & Technol China, Sch Mat & Energy, Chengdu 611731, Peoples R China.
通讯机构：
[He, L ] S;[Tahir, M; Dai, J ] B;Beijing Inst Technol, Sch Mechatron Engn, Beijing 100081, Peoples R China.;Sichuan Univ, Sch Mech Engn, State Key Lab Intelligent Construct & Hlth Operat, Chengdu 610065, Peoples R China.
关键词：
Zeolitic imidazolate frameworks;Metal-organic frameworks;Electrochemical gas sensors
摘要：
Zeolitic imidazolate framework-8 (ZIF-8), a subgroup of metal-organic frameworks, has garnered significant focus due to its remarkable structural tunability, chemical stability, and high surface area. These attributes make ZIF-8 and its derivatives desirable materials for drug delivery, catalysis, absorption of different gases, and gas sensing applications. This review comprehensively explains the current advances in the synthesis of ZIF-8 and its derivatives, emphasizing the various approaches, including solvothermal, mechanochemical, sonochemical, and seed-assisted methods, etc. The influence of critical parameters like linkers, solvents, pH, temperature, reaction flow rate, viscosity, interfacial tension, and reaction conditions on the morphology and performance of ZIF-8 is systematically analyzed. Furthermore, it explores the roles of ZIF-8-based materials in electrochemical sensing, highlighting their performance in sensing gases such as CO 2 , NH 3 , H 2 S, volatile organic compounds, NO 2 , and H 2 . The mechanisms behind the sensing capabilities of pure, composite, and hybrid ZIF-8 are explained, emphasizing the structure-property relationship and the development of composites. Challenges such as reproducibility, scalability, and environmental stability are also addressed, alongside prospects for integrating ZIF-8 into next-generation gas sensors. This review provides insight into the synthesis-application nexus, aiming to guide future research towards the rational design of ZIF-8-based gas sensors with improved performance and practical utility.
Zeolitic imidazolate framework-8 (ZIF-8), a subgroup of metal-organic frameworks, has garnered significant focus due to its remarkable structural tunability, chemical stability, and high surface area. These attributes make ZIF-8 and its derivatives desirable materials for drug delivery, catalysis, absorption of different gases, and gas sensing applications. This review comprehensively explains the current advances in the synthesis of ZIF-8 and its derivatives, emphasizing the various approaches, including solvothermal, mechanochemical, sonochemical, and seed-assisted methods, etc. The influence of critical parameters like linkers, solvents, pH, temperature, reaction flow rate, viscosity, interfacial tension, and reaction conditions on the morphology and performance of ZIF-8 is systematically analyzed. Furthermore, it explores the roles of ZIF-8-based materials in electrochemical sensing, highlighting their performance in sensing gases such as CO 2 , NH 3 , H 2 S, volatile organic compounds, NO 2 , and H 2 . The mechanisms behind the sensing capabilities of pure, composite, and hybrid ZIF-8 are explained, emphasizing the structure-property relationship and the development of composites. Challenges such as reproducibility, scalability, and environmental stability are also addressed, alongside prospects for integrating ZIF-8 into next-generation gas sensors. This review provides insight into the synthesis-application nexus, aiming to guide future research towards the rational design of ZIF-8-based gas sensors with improved performance and practical utility.
摘要：
The vat photopolymerization (VPP) technology offers exceptional flexibility in design and manufaction of high-performance piezoceramic structures. However, due to the high refractive index and strong UV absorption characteristics of Pb-based ceramics, ceramic slurries are required to have a low solid content in for efficient deep-curing. This limitation often results in sintering distortion, cracking, low density, and deteriorated piezoelectric properties. To address these challenges, a heat treatment process was initially employed to spheroidize the PZT powder, which effectively decreased light scattering and absorption. Subsequently, high-refractive-index and highly reactive photosensitive resins were introduced to enhance the penetration of UV light and reduce the critical exposure energy of the slurry. Additionally, a hindered amine light stabilizer is added to capture the peroxy radicals to overcome the oxygen inhibition effect and further increase the curing depth. As a result, a PZT slurry with a high solid content of 45 vol.%, was successfully formulated, facilitating the manufacturing of PZT ceramics with complex structures. The sintered samples achieved a high relative density of 96.5% and a d 33 value of 573 pC/N. These optimized strategies provide a solid foundation for the development of high-performance PZT devices with intricate geometries.
通讯机构：
[Yang, L ] W;Wuhan Polytech Univ, Coll Mech Engn, Wuhan 430048, Hubei, Peoples R China.
关键词：
Vertical rice mill;Interaction mechanism;Rice;Discrete element method
摘要：
The external milling vertical rice mill (EMVRM) is one of the new rice milling process equipment. The grains moving and interaction mechanism are basic points for milling, rice blade-sand bar spacing H, sieve structure are important structural parameters need considering. The EMVRM 3D model and DEM contact working model are established. The rice blade spacing H and sieve structure effect on grains motion and interaction in is analyzed. Grains motion velocity and density in milling chamber gradually reduced along axial flow direction. The milling chamber grains average motion velocity difference decreases with spacing H increasing. The grain normal and tangential force show down parabolic relation with axial distance from inlet. The sieve direction has little effect on grain motion velocity and force. Normal and tangential force in milling chamber reduced downward along the axial. The research provides practical guidance for EMVRM mill design.
The external milling vertical rice mill (EMVRM) is one of the new rice milling process equipment. The grains moving and interaction mechanism are basic points for milling, rice blade-sand bar spacing H, sieve structure are important structural parameters need considering. The EMVRM 3D model and DEM contact working model are established. The rice blade spacing H and sieve structure effect on grains motion and interaction in is analyzed. Grains motion velocity and density in milling chamber gradually reduced along axial flow direction. The milling chamber grains average motion velocity difference decreases with spacing H increasing. The grain normal and tangential force show down parabolic relation with axial distance from inlet. The sieve direction has little effect on grain motion velocity and force. Normal and tangential force in milling chamber reduced downward along the axial. The research provides practical guidance for EMVRM mill design.
摘要：
Fractionation allows the separation of components in beef tallow. This study compared the physicochemical characteristics and cholesterol content of beef tallow and its liquid fraction, evaluating their frying performance as potential deep-fat frying oils against plant oils. Results showed effective separation of unsaturated components from beef tallow through fractionation. Beef tallow exhibited superior physicochemical properties during frying, with lower deterioration levels than plant oils. Benzo[ a ]pyrene content increased in plant oils but remained low in beef tallow and its liquid fraction. The liquid fraction had a significantly shorter oxidative induction time of 0.38 h compared to 5.85 h and 5.24 h for plant oils. This study revealed that alterations were observed in beef tallow and its liquid fraction when used as frying oils, with beef tallow demonstrating stronger antioxidative properties compared to the liquid fraction, which exhibited lower levels of cholesterol and saturated fatty acids.
Fractionation allows the separation of components in beef tallow. This study compared the physicochemical characteristics and cholesterol content of beef tallow and its liquid fraction, evaluating their frying performance as potential deep-fat frying oils against plant oils. Results showed effective separation of unsaturated components from beef tallow through fractionation. Beef tallow exhibited superior physicochemical properties during frying, with lower deterioration levels than plant oils. Benzo[ a ]pyrene content increased in plant oils but remained low in beef tallow and its liquid fraction. The liquid fraction had a significantly shorter oxidative induction time of 0.38 h compared to 5.85 h and 5.24 h for plant oils. This study revealed that alterations were observed in beef tallow and its liquid fraction when used as frying oils, with beef tallow demonstrating stronger antioxidative properties compared to the liquid fraction, which exhibited lower levels of cholesterol and saturated fatty acids.
摘要：
This study comprehensively investigates the corrosion mechanism of laser-cladded FeCoNiCrCu high-entropy alloy (HEA) coatings fabricated with varying initial powder particle sizes in a 3.5 % NaCl solution. Subsequent to orthogonal experimental optimization, optimal laser cladding parameters were obtained to generate fine coatings with exceptional quality. It's noteworthy that finer initial powder particles possess higher specific surface energy, promoting the formation of coatings with fewer defects. Aside from that, steady increases in corrosion current density and decreases in corrosion potential were observed with coarser initial powder particles, accompanied by lower charge transfer resistance. Corrosion preferentially initiates at surface defects, with more severe defects directly degrading corrosion resistance. The Cu accumulation at these defects generates a less extensively protective passivation film, which hinders the formation of Cr oxides. As evidently demonstrated by XPS analysis, smaller powder particles form protective films with more Cr oxides and fewer Cu oxides in comparison with those formed by larger particles. To sum up, coatings prepared from finer initial powder particle sizes display superior corrosion resistance.
This study comprehensively investigates the corrosion mechanism of laser-cladded FeCoNiCrCu high-entropy alloy (HEA) coatings fabricated with varying initial powder particle sizes in a 3.5 % NaCl solution. Subsequent to orthogonal experimental optimization, optimal laser cladding parameters were obtained to generate fine coatings with exceptional quality. It's noteworthy that finer initial powder particles possess higher specific surface energy, promoting the formation of coatings with fewer defects. Aside from that, steady increases in corrosion current density and decreases in corrosion potential were observed with coarser initial powder particles, accompanied by lower charge transfer resistance. Corrosion preferentially initiates at surface defects, with more severe defects directly degrading corrosion resistance. The Cu accumulation at these defects generates a less extensively protective passivation film, which hinders the formation of Cr oxides. As evidently demonstrated by XPS analysis, smaller powder particles form protective films with more Cr oxides and fewer Cu oxides in comparison with those formed by larger particles. To sum up, coatings prepared from finer initial powder particle sizes display superior corrosion resistance.
摘要：
Localized reverse polarity (LRP) in proton exchange membrane fuel cells (PEMFCs) poses a critical degradation challenge by inducing localized redox reactions that compromise catalyst layer integrity and membrane conductivity, yet conventional voltage monitoring fails to detect its concealed electrochemical anomalies in real-time. Given the urgency of PEMFCs in sustainable energy transitions for decarbonizing transportation and stationary power systems, addressing LRP-driven degradation is vital to enhancing durability, efficiency, and safety, thereby reducing lifecycle costs and environmental impacts. While prior durability tests have quantified structural damage (e.g., 55.4 % catalyst layer thinning and 59.4 % platinum agglomeration in LRP-affected regions), no framework currently correlates these degradations with in situ impedance dynamics. This study fills this gap by introducing a non-destructive electrochemical impedance spectroscopy (EIS)-based diagnostic framework that establishes a three-dimensional quantitative model linking ohmic impedance (R ohm ) to LRP characteristics (area, duration, intensity), demonstrating a 92.0 % linear correlation with affected area. The findings enable real-time health monitoring, adaptive operational strategies to mitigate LRP, and actionable engineering insights for catalyst layer and membrane electrode assembly (MEA) optimization, thereby advancing impedance-based diagnostics as a cornerstone for sustainable, long-lived PEMFC systems and bridging the critical link between electrochemical impedance and spatially resolved degradation.
Localized reverse polarity (LRP) in proton exchange membrane fuel cells (PEMFCs) poses a critical degradation challenge by inducing localized redox reactions that compromise catalyst layer integrity and membrane conductivity, yet conventional voltage monitoring fails to detect its concealed electrochemical anomalies in real-time. Given the urgency of PEMFCs in sustainable energy transitions for decarbonizing transportation and stationary power systems, addressing LRP-driven degradation is vital to enhancing durability, efficiency, and safety, thereby reducing lifecycle costs and environmental impacts. While prior durability tests have quantified structural damage (e.g., 55.4 % catalyst layer thinning and 59.4 % platinum agglomeration in LRP-affected regions), no framework currently correlates these degradations with in situ impedance dynamics. This study fills this gap by introducing a non-destructive electrochemical impedance spectroscopy (EIS)-based diagnostic framework that establishes a three-dimensional quantitative model linking ohmic impedance (R ohm ) to LRP characteristics (area, duration, intensity), demonstrating a 92.0 % linear correlation with affected area. The findings enable real-time health monitoring, adaptive operational strategies to mitigate LRP, and actionable engineering insights for catalyst layer and membrane electrode assembly (MEA) optimization, thereby advancing impedance-based diagnostics as a cornerstone for sustainable, long-lived PEMFC systems and bridging the critical link between electrochemical impedance and spatially resolved degradation.
摘要：
Investigations using hot compression tests on a new high-strength weathering steel revealed specific deformation behaviors across different conditions. These tests were performed at temperatures ranging from 850 to 1050°C and at strain rates from 0.01 to 5s(-1). Results indicated that a decrease in the deformation temperature combined with an increase in strain rate notably enhanced both the maximum stress and strain achieved. Notably, above 900°C and with strain rates below 0.1s(-1), the flow stress of the material reached a steady state at certain strain levels. At a strain rate of 1s(-1), irrespective of the temperature, the steel shows a continuous strain hardening behavior, achieving no stable flow stress state. Notably, when the true strain exceeds 0.8, an unusual increase in flow stress occurs, predominantly due to secondary work hardening effects. The microstructural changes in the deformed samples were examined using electron backscatter diffraction (EBSD), which helped elucidate the softening mechanisms inherent in this high-strength steel. Further, processing maps developed from true strains of 0.1-0.9, derived from the experimental flow stress data, suggest controlling the strain within 0.2-0.4 to minimize instability during hot working.
作者机构：
[Xie, Jun; Xie, J; Cai, Song] Hunan First Normal Univ, Sch Intelligent Mfg, Changsha 410205, Peoples R China.;[Zhang, Yi] TravelSky Technol Ltd, Beijing 100000, Peoples R China.;[Tang, Yun] Hunan Univ Sci & Technol, Sch Phys & Elect Sci, Xiangtan 411201, Peoples R China.;[Ji, Yi; Cai, Song] Wuhan Polytech Univ, Sch Mech Engn, Wuhan 430074, Peoples R China.
通讯机构：
[Xie, J ] H;Hunan First Normal Univ, Sch Intelligent Mfg, Changsha 410205, Peoples R China.
摘要：
This paper presents a novel mechanistic explanation for microwave-assisted LIBS to mitigate self-absorption effects. A set of plasma characteristic equations in cylindrical coordinates, including plasma isothermal expansion and modified adiabatic expansion kinetics equations, were formulated. These equations were subsequently coupled with microwave electric field equations to develop a microwave energy-assisted model. These models were employed to numerically analyze the plasma dimensions, velocity, and spatial distribution characteristics of plasma concentration, as well as the energy consumption during plasma expansion and the microwave-assisted energy. This analysis aims to explain the self-absorption mechanism and uncover how microwave-assisted LIBS mitigates self-absorption. LIBS experiments, both with and without microwave assistance, were conducted. An improved Saha-Boltzmann planar method was proposed to quantify the degree of self-absorption in the spectral lines of Al, Si, and Ca. Based on the measured spectral data, the plasma temperature of Al was calculated using this improved method, while the electron densities of Si, Ca, and Al plasmas were determined independently of self-absorption effects, as no spectral line intensity information was involved. The evolution of plasma expansion was captured using an intensified charge-coupled device (ICCD). The experiments confirmed that microwave- assisted LIBS did not alter plasma electron temperature or electron density, but provides sufficient energy for uniform plasma expansion, thereby reducing self-absorption. This finding offers a theoretical reference for mitigating jitter following femtosecond laser wire formation. Furthermore, the experiments demonstrated that microwaves effectively reduce self-absorption and enhance spectral intensity, validating both the accuracy and feasibility of the plasma characteristic equations and the microwave energy-assisted model. These results provide theoretical guidance and experimental optimization for plasma characteristics in LIBS applications.
摘要：
Degumming, a critical process in the edible oil industry, is essential for removing phospholipids. Traditional methods such as water-degumming and acid-degumming have limitations in this regard. This study aimed to assess the efficacy of different degumming techniques, including acid, single and multi-enzyme methods, on rice bran oil (RBO). The investigation also focused on the impact of these techniques on the physicochemical characteristics and preservation of micronutrients in RBO during the degumming process. The primary phospholipids identified were phosphatidylethanolamine (29.51 %), phosphatidylcholine (37.00 %), and phosphatidylinositol (24.49 %). Acid degumming removed 84.23 % of phospholipids, while a significantly higher removal rate of 98.7 % was achieved with the combination of phospholipase A1 & phospholipase C. The degumming process effectively inhibited oxidation in RBO, leading to a substantial increase in the oxidation induction time from 5.7 to 10.0 hours. Furthermore, multi-enzyme degumming showed slightly greater radical-scavenging activity compared to single enzyme degumming in RBO. However, the levels of micronutrients such as phenols, sterols, tocopherols, squalene, and oryzanol were reduced by 6.27–22.17 %. This study provides a comprehensive analysis of the effects of different degumming processes on the physicochemical properties, fatty acid profiles, antioxidant capacities, and preservation of micronutrients in RBO.
Degumming, a critical process in the edible oil industry, is essential for removing phospholipids. Traditional methods such as water-degumming and acid-degumming have limitations in this regard. This study aimed to assess the efficacy of different degumming techniques, including acid, single and multi-enzyme methods, on rice bran oil (RBO). The investigation also focused on the impact of these techniques on the physicochemical characteristics and preservation of micronutrients in RBO during the degumming process. The primary phospholipids identified were phosphatidylethanolamine (29.51 %), phosphatidylcholine (37.00 %), and phosphatidylinositol (24.49 %). Acid degumming removed 84.23 % of phospholipids, while a significantly higher removal rate of 98.7 % was achieved with the combination of phospholipase A1 & phospholipase C. The degumming process effectively inhibited oxidation in RBO, leading to a substantial increase in the oxidation induction time from 5.7 to 10.0 hours. Furthermore, multi-enzyme degumming showed slightly greater radical-scavenging activity compared to single enzyme degumming in RBO. However, the levels of micronutrients such as phenols, sterols, tocopherols, squalene, and oryzanol were reduced by 6.27–22.17 %. This study provides a comprehensive analysis of the effects of different degumming processes on the physicochemical properties, fatty acid profiles, antioxidant capacities, and preservation of micronutrients in RBO.
摘要：
The granularity effect is a key factor in many experiments, significantly influencing the properties and performance of materials. This study prepared oleogels and hydrogels using beeswax and gelatin as gelators, respectively. The study examined the effects of hydrogel particle size on the performance of the bigels and the influence of storage conditions on hydrogel particle size. The results indicate that the Fourier Transform Infrared Spectroscopy (FTIR) spectral characteristics of the bigels are predominantly governed by the oleogel component. Differential scanning calorimetry (DSC) analysis further revealed that the bigels began to melt at approximately 37 °C, reached a peak endothermic response at 55 °C, and completed melting at 60 °C, exhibiting more complex and broader thermal behavior than the individual components. As the hydrogel particle size within the bigels decreased, hardness increased by 2.20 times, elastic modulus by 2.23 times, and both storage and loss moduli also rose, indicating enhanced elasticity and energy dissipation. Smaller hydrogel particle sizes improved oil retention capacity, with BG5 exhibiting a 15.73% higher oil retention rate than BG1 under non-freeze-thaw conditions. Freeze-thaw cycles significantly affected hydrogel particle size, with the particle size of BG1 increasing by 38.33 μm and BG5 increasing by only 12.61 μm.
The granularity effect is a key factor in many experiments, significantly influencing the properties and performance of materials. This study prepared oleogels and hydrogels using beeswax and gelatin as gelators, respectively. The study examined the effects of hydrogel particle size on the performance of the bigels and the influence of storage conditions on hydrogel particle size. The results indicate that the Fourier Transform Infrared Spectroscopy (FTIR) spectral characteristics of the bigels are predominantly governed by the oleogel component. Differential scanning calorimetry (DSC) analysis further revealed that the bigels began to melt at approximately 37 °C, reached a peak endothermic response at 55 °C, and completed melting at 60 °C, exhibiting more complex and broader thermal behavior than the individual components. As the hydrogel particle size within the bigels decreased, hardness increased by 2.20 times, elastic modulus by 2.23 times, and both storage and loss moduli also rose, indicating enhanced elasticity and energy dissipation. Smaller hydrogel particle sizes improved oil retention capacity, with BG5 exhibiting a 15.73% higher oil retention rate than BG1 under non-freeze-thaw conditions. Freeze-thaw cycles significantly affected hydrogel particle size, with the particle size of BG1 increasing by 38.33 μm and BG5 increasing by only 12.61 μm.
作者：
Mou, Yi;Zhou, Long;Chen, Weizhen;Liu, Jianguo;Li, Teng
期刊：
Algorithms,2025年18(7):424- ISSN：1999-4893
通讯作者：
Yi Mou
作者机构：
[Chen, Weizhen; Liu, Jianguo; Li, Teng] School of Electrical and Electronic Engineering, Wuhan Polytechnic University, Wuhan 430024, China;[Zhou, Long] School of Mechanical Engineering, Wuhan Polytechnic University, Wuhan 430024, China;Author to whom correspondence should be addressed.;[Mou, Yi] School of Electrical and Electronic Engineering, Wuhan Polytechnic University, Wuhan 430024, China<&wdkj&>Author to whom correspondence should be addressed.
通讯机构：
[Yi Mou] S;School of Electrical and Electronic Engineering, Wuhan Polytechnic University, Wuhan 430024, China<&wdkj&>Author to whom correspondence should be addressed.
关键词：
partial least squares;regression analysis;filter learning;content prediction
摘要：
Partial Least Squares (PLS) regression has been widely used to model the relationship between predictors and responses. However, PLS may be limited in its capacity to handle complex spectral data contaminated with significant noise and interferences. In this paper, we propose a novel filter learning-based PLS (FPLS) model that integrates an adaptive filter into the PLS framework. The FPLS model is designed to maximize the covariance between the filtered spectral data and the response. This modification enables FPLS to dynamically adapt to the characteristics of the data, thereby enhancing its feature extraction and noise suppression capabilities. We have developed an efficient algorithm to solve the FPLS optimization problem and provided theoretical analyses regarding the convergence of the model, the prediction variance, and the relationships among the objective functions of FPLS, PLS, and the filter length. Furthermore, we have derived bounds for the Root Mean Squared Error of Prediction (RMSEP) and the Cosine Similarity (CS) to evaluate model performance. Experimental results using spectral datasets from Corn, Octane, Mango, and Soil Nitrogen show that the FPLS model outperforms PLS, OSCPLS, VCPLS, PoPLS, LoPLS, DOSC, OPLS, MSC, SNV, SGFilter, and Lasso in terms of prediction accuracy. The theoretical analyses align with the experimental results, emphasizing the effectiveness and robustness of the FPLS model in managing complex spectral data.
作者机构：
[Yu Zhai; Jianjun Yang; Xiao Rang; Zean Wang; Houchang Pei; Shaoyun Song] College of Mechanical Engineering, Wuhan Polytechnic University, Wuhan, China
会议名称：
2025 5th International Conference on Artificial Intelligence and Industrial Technology Applications (AIITA)
会议时间：
28 March 2025
会议地点：
Xi'an, China
会议论文集名称：
2025 5th International Conference on Artificial Intelligence and Industrial Technology Applications (AIITA)
关键词：
Steel material defects detection;YOLOv7;Attention mechanism;Data enhancement
摘要：
The detection of steel material defects is vital for enhancing product quality, safety, and reliability. However, conventional deep learning approaches, such as YOLOv7 and SSD, suffer from slow detection speed and suboptimal accuracy. To address these issues, we present an enhanced YOLOv7 algorithm for steel defect detection. Our approach optimizes the YOLOv7 model by integrating ResNet Channel Attention Connection modules into the backbone network to improve feature extraction capabilities. Furthermore, a self-Coordinate Attention mechanism is introduced to enhance detection accuracy. Additionally, we modify downsampling using the unfold model to improve the detection of small objects. Experimental results demonstrate the effectiveness of our proposed enhanced YOLOv7 model in accurately identifying steel defects. Compared to the original model, we achieved an 11.4% increase in mean average precision (mAP) and reduced training time by 2.795 hours.
摘要：
The broken rice rate (BR) is a critical metric which influences the appearance, processing, and economic value of rice. However, current machine vision and machine learning approaches engender significant errors when calculating BR. This study introduces a novel restoring method for identifying BR by leveraging grading and morphological features. A three-class classification model using Convolutional Neural Network (CNN) was devised to distinguish broken rice types of crescent head, elliptical tail, and quadrilateral midst based on their morphological characteristics. After training, the accuracy of classfication model is over 98.7%. Taking the longest 10% of rice grains in the image to be identified as head rice references, the broken grains are filtered by calculating the length proportion to the head rice via machine vision. The filtered broken grains are classified to one of three morphological categories with the trained CNN. The broken grains are virtually ‘restored' to head rice equivalents based on the classified shape and the grading size. Finally, the BR is determined by comparing the counts of original and restored grains. The results of two testing conditions which including all and lacking some broken grains demonstrate that the proposed method can accurately and effectively identify the BR in real-time (2.5s).
The broken rice rate (BR) is a critical metric which influences the appearance, processing, and economic value of rice. However, current machine vision and machine learning approaches engender significant errors when calculating BR. This study introduces a novel restoring method for identifying BR by leveraging grading and morphological features. A three-class classification model using Convolutional Neural Network (CNN) was devised to distinguish broken rice types of crescent head, elliptical tail, and quadrilateral midst based on their morphological characteristics. After training, the accuracy of classfication model is over 98.7%. Taking the longest 10% of rice grains in the image to be identified as head rice references, the broken grains are filtered by calculating the length proportion to the head rice via machine vision. The filtered broken grains are classified to one of three morphological categories with the trained CNN. The broken grains are virtually ‘restored' to head rice equivalents based on the classified shape and the grading size. Finally, the BR is determined by comparing the counts of original and restored grains. The results of two testing conditions which including all and lacking some broken grains demonstrate that the proposed method can accurately and effectively identify the BR in real-time (2.5s).
摘要：
To address the challenges associated with conducting large-scale and complex acoustic vibration experiments on dual-layer cylindrical shells, which are both difficult and costly, this paper proposes a method that predicts the vibrational response of a full-scale dual-layer cylindrical shell using the vibrational response of a single-layer scaled-down model. Finite element models of both the full-scale dual-layer cylindrical shell and its scaled-down counterpart were developed using Virtual.Lab Acoustic software. The vibrational responses of these models were analyzed within the 100-500 Hz frequency range, and a comparative analysis was conducted to validate the accuracy of the scaling method and the similarity of vibrational responses. The results indicate that the displacement ratio between the scaled-down model and the original model aligns closely with the geometric scaling factor. Furthermore, the similarity coefficients of the vibrational responses for both the inner and outer shells of the scaled-down model exceed 0.9, confirming the effectiveness of the proposed approach.
通讯机构：
[Shen, SN ; Li, H] W;Wuhan Univ, Inst Technol Sci, Wuhan 430072, Peoples R China.
关键词：
In-situ monitoring;Video frame interpolation;Super resolution reconstruction;Laser powder bed fusion
摘要：
In the laser powder bed fusion (L-PBF) process, the incorporation of in-situ monitoring systems plays a vital role in guaranteeing the quality of the additive manufacturing (AM) process. Nevertheless, the monitoring system based on high-speed cameras is hindered by the high cost of the required high-speed cameras, making it difficult to achieve accurate in-situ monitoring. This paper studies in-situ video frame interpolation and super resolution reconstruction for accurate monitoring of L-PBF process. It introduces a novel in-situ video frame interpolation algorithm, termed CS-EMA-VFI, aiming to improve the temporal resolution of monitoring video. The visual transformer-based video super resolution (ViTSR) algorithm was employed to enhance the spatial resolution of the interpolated video. A U-Net algorithm was utilized for extracting the geometric characteristics of the molten pool during the L-PBF process subsequent to video frame interpolation and super resolution reconstruction. Comparing the CS-EMA-VFI with seven state-of-the-art video frame interpolation methods, the CS-EMA-VFI achieves the highest peak signal-to-noise ratio (PSNR) of 28.16 dB and the highest structural similarity index measure (SSIM) of 0.917 while being lightweight. The ViTSR achieved PSNR of 28.18 dB and 25.31 dB on the original video sequence and interpolated video sequence, respectively. The inference time for the CS-EMA-VFI with fixed timestep, ViTSR, and U-Net were recorded as 18.5 ms, 48.0 ms, and 20.5 ms, respectively. The total inference time of the three-stage strategy varies from 87.0 ms to 142.5 ms, depending on the temporal resolution enhancement multiples. Additionally, the proposed three-stage method achieves a segmentation accuracy of 90.15 % with fixed timestep interpolation, simultaneously enhancing temporal and spatial resolution, thus enabling accurate and real-time monitoring. This paper promotes the wide adoption of in-situ monitoring system in the AM field.
In the laser powder bed fusion (L-PBF) process, the incorporation of in-situ monitoring systems plays a vital role in guaranteeing the quality of the additive manufacturing (AM) process. Nevertheless, the monitoring system based on high-speed cameras is hindered by the high cost of the required high-speed cameras, making it difficult to achieve accurate in-situ monitoring. This paper studies in-situ video frame interpolation and super resolution reconstruction for accurate monitoring of L-PBF process. It introduces a novel in-situ video frame interpolation algorithm, termed CS-EMA-VFI, aiming to improve the temporal resolution of monitoring video. The visual transformer-based video super resolution (ViTSR) algorithm was employed to enhance the spatial resolution of the interpolated video. A U-Net algorithm was utilized for extracting the geometric characteristics of the molten pool during the L-PBF process subsequent to video frame interpolation and super resolution reconstruction. Comparing the CS-EMA-VFI with seven state-of-the-art video frame interpolation methods, the CS-EMA-VFI achieves the highest peak signal-to-noise ratio (PSNR) of 28.16 dB and the highest structural similarity index measure (SSIM) of 0.917 while being lightweight. The ViTSR achieved PSNR of 28.18 dB and 25.31 dB on the original video sequence and interpolated video sequence, respectively. The inference time for the CS-EMA-VFI with fixed timestep, ViTSR, and U-Net were recorded as 18.5 ms, 48.0 ms, and 20.5 ms, respectively. The total inference time of the three-stage strategy varies from 87.0 ms to 142.5 ms, depending on the temporal resolution enhancement multiples. Additionally, the proposed three-stage method achieves a segmentation accuracy of 90.15 % with fixed timestep interpolation, simultaneously enhancing temporal and spatial resolution, thus enabling accurate and real-time monitoring. This paper promotes the wide adoption of in-situ monitoring system in the AM field.
摘要：
Silver powders play a pivotal role in electronic applications due to their superior conductivity and chemical stability, where morphology and particle size critically determine performance. This work presents a controllable liquid-phase reduction strategy to synthesize spherical silver powders using ascorbic acid as an eco-friendly reductant and polyvinylpyrrolidone (PVP) as a morphology-directing agent. Key synthesis parameters, including reactant mixing mode, temperature (30 degrees C), silver nitrate concentration (0.2 mol/L), PVP dosage (5 wt%), and reducing solution pH (3.3), were systematically optimized to achieve monodisperse spherical particles with an average size of 2 mu m. When formulated into sintered conductive pastes with 75 wt% silver loading, the optimized powder enabled exceptional electrical and mechanical performance: a low sheet resistance of 5.3 m Omega/sq and robust adhesion exceeding 20 N. These results demonstrate a scalable route to tailor silver powders for high-performance electronic interconnects in photovoltaic and 5G modules, balancing cost-efficiency with functional reliability.
摘要：
Research on poultry part partitioning techniques is crucial for the advancement of automated poultry partitioning equipment. In this study, a semantic segmentation method for chicken parts, based on a lightweight DeepLabv3+, was introduced to cater to real-time and precise requirements of segmenting varying poultry sizes. Initially, the backbone network was replaced with an improved lightweight MobileNetV2, enhancing the predictive speed and decreasing computational parameters. Subsequently, the SENet was incorporated, enhancing the capacity to discern high-level features and negate irrelevant information. Furthermore, two shallow feature layers of different scales were integrated into the decoder, augmenting the richness of shallow features and mitigating inaccuracies at segmentation edges. Finally, the Dice Loss and Cross Entropy Loss (CE Loss) functions were combined to minimize the imbalance between positive and negative samples. Experimental findings demonstrated that the lightweight DeepLabv3+ improved the MIoU (Mean Intersection over Union) and MPA (Mean Pixel Accuracy) scores of the original model by 5.42% and 3%, respectively, and amplified the detection speed by 1.89 times. Remarkably, the model size was a mere 10.95% of the original, indicating substantial enhancements in segmentation accuracy and detection speed. Therefore, the proposed algorithm could potentially provide certain technical insights for automatic segmentation of different poultry.
摘要：
With rising living standards, the demand for health and nutrition has increased, sparking interest in food antioxidants. Known for neutralizing free radicals, antioxidants protect cells from oxidative damage, potentially aiding in disease prevention and anti-aging. In the food industry, they also enhance preservation and quality. Thus, studying food antioxidant mechanisms, detection methods, and applications holds theoretical and practical value. This review mainly discusses the mechanisms, detection methods, and applications of food antioxidants in nutrition. Firstly, the main research status and development trends of food antioxidants are described. Then, the action mechanisms of food antioxidants are introduced. Food antioxidants can effectively remove free radicals and prevent free radicals from causing damage to human cells, thus delaying aging and preventing disease. Secondly, the methods of detecting food antioxidants are discussed, including liquid chromatography, liquid chromatography-tandem mass spectrometry, gas chromatography, and gas chromatography-mass spectrometry. These methods can be used to analyze antioxidant components in various samples of foods, drugs, plants, etc. Finally, the research progress of plant antioxidants is discussed, including the applications of a variety of highly effective antioxidant components extracted from different plants. This review provides the theoretical basis and application reference for the research of food antioxidants.
