Diabetic Retinopathy(DR)is a critical disorder that affects the retina due to the constant rise in diabetics and remains the major cause of blindness across the world.Early detection and timely treatment are essential...Diabetic Retinopathy(DR)is a critical disorder that affects the retina due to the constant rise in diabetics and remains the major cause of blindness across the world.Early detection and timely treatment are essential to mitigate the effects of DR,such as retinal damage and vision impairment.Several conventional approaches have been proposed to detect DR early and accurately,but they are limited by data imbalance,interpretability,overfitting,convergence time,and other issues.To address these drawbacks and improve DR detection accurately,a distributed Explainable Convolutional Neural network-enabled Light Gradient Boosting Machine(DE-ExLNN)is proposed in this research.The model combines an explainable Convolutional Neural Network(CNN)and Light Gradient Boosting Machine(LightGBM),achieving highly accurate outcomes in DR detection.LightGBM serves as the detection model,and the inclusion of an explainable CNN addresses issues that conventional CNN classifiers could not resolve.A custom dataset was created for this research,containing both fundus and OCTA images collected from a realtime environment,providing more accurate results compared to standard conventional DR datasets.The custom dataset demonstrates notable accuracy,sensitivity,specificity,and Matthews Correlation Coefficient(MCC)scores,underscoring the effectiveness of this approach.Evaluations against other standard datasets achieved an accuracy of 93.94%,sensitivity of 93.90%,specificity of 93.99%,and MCC of 93.88%for fundus images.For OCTA images,the results obtained an accuracy of 95.30%,sensitivity of 95.50%,specificity of 95.09%,andMCC of 95%.Results prove that the combination of explainable CNN and LightGBMoutperforms othermethods.The inclusion of distributed learning enhances the model’s efficiency by reducing time consumption and complexity while facilitating feature extraction.展开更多
Underwater acoustic target recognition(UATR)has become increasingly prevalent for ocean detection,localisation,and identification.However,due to the complexity and variability of underwater environments,especially in ...Underwater acoustic target recognition(UATR)has become increasingly prevalent for ocean detection,localisation,and identification.However,due to the complexity and variability of underwater environments,especially in multi ship event environments,where multiple acoustic signals coexist,practical applications face significant challenges.These challenges hinder single-category acoustic recognition algorithms,particularly in extracting time series features and achieving fine-grained or multi-scale feature fusion.This paper innovatively introduce the SKANN framework,which achieve precise submarine sound recognition in underwater mixed ship events environments through timing data enhancement and sampling training module and selective kernel feature extraction module.The timing data enhancement and sampling training module improves time sequence feature extraction through progressive acoustic sampling.The selective kernel feature extraction module effectively fuses multi-scale features by integrating selective kernel(SK)technology.To simulate concurrent ship events,we constructed the mixed ship noise dataset(MDeepShip),providing an experimental basis and test platform for underwater mixed ship event detection.This dataset ensures that the model encounters diverse audio samples during training and validation,improving its ability to extract temporal features.Experimental results show that SKANN achieves a 93.6%recognition rate on the M-DeepShip dataset,demonstrating its effectiveness in recognising underwater mixed ship events.Given the complexity of real underwater environments,this work lays a crucial foundation for the sound recognition of submarine vessels.Future research will focus on real marine environments to validate and refine the models and methods for practical applications.展开更多
Vertical Federated Learning(VFL),which draws attention because of its ability to evaluate individuals based on features spread across multiple institutions,encounters numerous privacy and security threats.Existing sol...Vertical Federated Learning(VFL),which draws attention because of its ability to evaluate individuals based on features spread across multiple institutions,encounters numerous privacy and security threats.Existing solutions often suffer from centralized architectures,and exorbitant costs.To mitigate these issues,in this paper,we propose SecureVFL,a decentralized multi-party VFL scheme designed to enhance efficiency and trustworthiness while guaranteeing privacy.SecureVFL uses a permissioned blockchain and introduces a novel consensus algorithm,Proof of Feature Sharing(PoFS),to facilitate decentralized,trustworthy,and high-throughput federated training.SecureVFL introduces a verifiable and lightweight three-party Replicated Secret Sharing(RSS)protocol for feature intersection summation among overlapping users.Furthermore,we propose a(_(2)^(4))-sharing protocol to achieve federated training in a four-party VFL setting.This protocol involves only addition operations and exhibits robustness.SecureVFL not only enables anonymous interactions among participants but also safeguards their real identities,and provides mechanisms to unmask these identities when malicious activities are performed.We illustrate the proposed mechanism through a case study on VFL across four banks.Finally,our theoretical analysis proves the security of SecureVFL.Experiments demonstrated that SecureVFL outperformed existing multi-party VFL privacy-preserving schemes,such as MP-FedXGB,in terms of both overhead and model performance.展开更多
Electronic nose and thermal images are effective ways to diagnose the presence of gases in real-time realtime.Multimodal fusion of these modalities can result in the development of highly accurate diagnostic systems.T...Electronic nose and thermal images are effective ways to diagnose the presence of gases in real-time realtime.Multimodal fusion of these modalities can result in the development of highly accurate diagnostic systems.The low-cost thermal imaging software produces low-resolution thermal images in grayscale format,hence necessitating methods for improving the resolution and colorizing the images.The objective of this paper is to develop and train a super-resolution generative adversarial network for improving the resolution of the thermal images,followed by a sparse autoencoder for colorization of thermal images and amultimodal convolutional neural network for gas detection using electronic nose and thermal images.The dataset used comprises 6400 thermal images and electronic nose measurements for four classes.A multimodal Convolutional Neural Network(CNN)comprising an EfficientNetB2 pre-trainedmodel was developed using both early and late feature fusion.The Super Resolution Generative Adversarial Network(SRGAN)model was developed and trained on low and high-resolution thermal images.Asparse autoencoder was trained on the grayscale and colorized thermal images.The SRGAN was trained on lowand high-resolution thermal images,achieving a Structural Similarity Index(SSIM)of 90.28,a Peak Signal-to-Noise Ratio(PSNR)of 68.74,and a Mean Absolute Error(MAE)of 0.066.The autoencoder model produced an MAE of 0.035,a Mean Squared Error(MSE)of 0.006,and a Root Mean Squared Error(RMSE)of 0.0705.The multimodal CNN,trained on these images and electronic nose measurements using both early and late fusion techniques,achieved accuracies of 97.89% and 98.55%,respectively.Hence,the proposed framework can be of great aid for the integration with low-cost software to generate high quality thermal camera images and highly accurate detection of gases in real-time.展开更多
The development of generative architectures has resulted in numerous novel deep-learning models that generate images using text inputs.However,humans naturally use speech for visualization prompts.Therefore,this paper...The development of generative architectures has resulted in numerous novel deep-learning models that generate images using text inputs.However,humans naturally use speech for visualization prompts.Therefore,this paper proposes an architecture that integrates speech prompts as input to image-generation Generative Adversarial Networks(GANs)model,leveraging Speech-to-Text translation along with the CLIP+VQGAN model.The proposed method involves translating speech prompts into text,which is then used by the Contrastive Language-Image Pretraining(CLIP)+Vector Quantized Generative Adversarial Network(VQGAN)model to generate images.This paper outlines the steps required to implement such a model and describes in detail the methods used for evaluating the model.The GAN model successfully generates artwork from descriptions using speech and text prompts.Experimental outcomes of synthesized images demonstrate that the proposed methodology can produce beautiful abstract visuals containing elements from the input prompts.The model achieved a Frechet Inception Distance(FID)score of 28.75,showcasing its capability to produce high-quality and diverse images.The proposed model can find numerous applications in educational,artistic,and design spaces due to its ability to generate images using speech and the distinct abstract artistry of the output images.This capability is demonstrated by giving the model out-of-the-box prompts to generate never-before-seen images with plausible realistic qualities.展开更多
Landslide hazard detection is a prevalent problem in remote sensing studies,particularly with the technological advancement of computer vision.With the continuous and exceptional growth of the computational environmen...Landslide hazard detection is a prevalent problem in remote sensing studies,particularly with the technological advancement of computer vision.With the continuous and exceptional growth of the computational environment,the manual and partially automated procedure of landslide detection from remotely sensed images has shifted toward automatic methods with deep learning.Furthermore,attention models,driven by human visual procedures,have become vital in natural hazard-related studies.Hence,this paper proposes an enhanced YOLOv5(You Only Look Once version 5)network for improved satellite-based landslide detection,embedded with two popular attention modules:CBAM(Convolutional Block Attention Module)and ECA(Efficient Channel Attention).These attention mechanisms are incorporated into the backbone and neck of the YOLOv5 architecture,distinctly,and evaluated across three YOLOv5 variants:nano(n),small(s),and medium(m).The experiments use opensource satellite images from three distinct regions with complex terrain.The standard metrics,including F-score,precision,recall,and mean average precision(mAP),are computed for quantitative assessment.The YOLOv5n+CBAM demonstrates the most optimal results with an F-score of 77.2%,confirming its effectiveness.The suggested attention-driven architecture augments detection accuracy,supporting post-landslide event assessment and recovery.展开更多
Hydrocarbons,carbon monoxide and other pollutants from the transportation sector harm human health in many ways.Fuel cell(FC)has been evolving rapidly over the past two decades due to its efficient mechanism to transf...Hydrocarbons,carbon monoxide and other pollutants from the transportation sector harm human health in many ways.Fuel cell(FC)has been evolving rapidly over the past two decades due to its efficient mechanism to transform the chemical energy in hydrogen-rich compounds into electrical energy.The main drawback of the standalone FC is its slow dynamic response and its inability to supply rapid variations in the load demand.Therefore,adding energy storage systems is necessary.However,to manage and distribute the power-sharing among the hybrid proton exchange membrane(PEM)fuel cell(FC),battery storage(BS),and supercapacitor(SC),an energy management strategy(EMS)is essential.In this research work,an optimal EMS based on a spotted hyena optimizer(SHO)for hybrid PEM fuel cell/BS/SC is proposed.The main goal of an EMS is to improve the performance of hybrid FC/BS/SC and to reduce the amount of hydrogen consumption.To prove the superiority of the SHO method,the obtained results are compared with the chimp optimizer(CO),the artificial ecosystem-based optimizer(AEO),the seagull optimization algorithm(SOA),the sooty tern optimization algorithm(STOA),and the coyote optimization algorithm(COA).Two main metrics are used as a benchmark for the comparison:the minimum consumed hydrogen and the efficiency of the system.The main findings confirm that the minimum amount of hydrogen consumption and maximum efficiency are achieved by the proposed SHO based EMS.展开更多
In computer vision,emotion recognition using facial expression images is considered an important research issue.Deep learning advances in recent years have aided in attaining improved results in this issue.According t...In computer vision,emotion recognition using facial expression images is considered an important research issue.Deep learning advances in recent years have aided in attaining improved results in this issue.According to recent studies,multiple facial expressions may be included in facial photographs representing a particular type of emotion.It is feasible and useful to convert face photos into collections of visual words and carry out global expression recognition.The main contribution of this paper is to propose a facial expression recognitionmodel(FERM)depending on an optimized Support Vector Machine(SVM).To test the performance of the proposed model(FERM),AffectNet is used.AffectNet uses 1250 emotion-related keywords in six different languages to search three major search engines and get over 1,000,000 facial photos online.The FERM is composed of three main phases:(i)the Data preparation phase,(ii)Applying grid search for optimization,and(iii)the categorization phase.Linear discriminant analysis(LDA)is used to categorize the data into eight labels(neutral,happy,sad,surprised,fear,disgust,angry,and contempt).Due to using LDA,the performance of categorization via SVM has been obviously enhanced.Grid search is used to find the optimal values for hyperparameters of SVM(C and gamma).The proposed optimized SVM algorithm has achieved an accuracy of 99%and a 98%F1 score.展开更多
Recently, gears of high strength, reliability, and surface-damage-resistant under severe service conditions are required to achieve the weight saving and downsizing of a product. For the high-speed condition in partic...Recently, gears of high strength, reliability, and surface-damage-resistant under severe service conditions are required to achieve the weight saving and downsizing of a product. For the high-speed condition in particular, it is important to understand the influence of the surface properties on the scuffing resistance. If the effective surface profile to improve the lubrication property was found, the metal surfaces could be obtained with both surface strength and surface lubricity. Herein, the influence of surface properties modified with fine shot peening, which can form the arbitrary surface profile, on the scuffing resistance in the rolling-sliding contact machine element, was investigated. The scuffing test was performed using a two-cylinder rolling contact test machine. In a specific sliding, a faster roller of 60% and a sliding velocity of 1.75 m/s were utilized. The scuffing test results with shot-peened test rollers and those with non-shot-peened test roller were compared. The influence of the surface roughness of the shot-peened test roller was also discussed. We found that the shot-peened roller had a better scuffing resistance compared with the roller without the shot-peening process.展开更多
The electric power infrastructure that has served huge loads for so long is rapidly running up against many limitations. Out of many challenges it is to operate the power system in secure manner so that the operation ...The electric power infrastructure that has served huge loads for so long is rapidly running up against many limitations. Out of many challenges it is to operate the power system in secure manner so that the operation constraints are fulfilled under both normal and contingent conditions. Smart grid technology offers valuable techniques that can be deployed within the very near future or which are already deployed nowadays. Flexible AC Transmission Systems (FACTS) devices have been introduced to solve various power system problems. In literature, most of the methods proposed for sizing the FACTS devices only consider the normal operating conditions of power systems. Consequently, some transmission lines are heavily loaded in contingency case and the system voltage stability becomes a power transfer-limiting factor. This paper presents a technique for determining the proper rating/size of FACTS devices, namely the Static Synchronous Compensator (STATCOM), while considering contingency cases. The paper also verifies that the weakest bus determined by eigenvalue and eigenvectors method is the best location for STATCOM. The rating of STATCOM is specified according to the required reactive power needed to improve voltage stability under normal and contingency cases. Two case system studies are investigated: a simple 5-bus system and the IEEE 14-bus system. The obtained results verify that the rating of STATCOM can be determined according to the worst contingency case, and through proper control it can still be effective for normal and other contingency cases.展开更多
In this paper, in order to design a cam mechanism be up to the mustard, a set of methods are put forward that using the Visual Basic programming language based on solidworks to draw cam contour line and then get its 3...In this paper, in order to design a cam mechanism be up to the mustard, a set of methods are put forward that using the Visual Basic programming language based on solidworks to draw cam contour line and then get its 3D models and generate the cam motion simulation by the solidworks motion. In the end, it’s proved that the cam designed though this method met the requirement.展开更多
Utilities around the world have been considering Demand Side Management (DSM) in their strategic planning. The costs of constructing and operating a new capacity generation unit are increasing everyday as well as Tran...Utilities around the world have been considering Demand Side Management (DSM) in their strategic planning. The costs of constructing and operating a new capacity generation unit are increasing everyday as well as Transmission and distribution and land issues for new generation plants, which force the utilities to search for another alternatives without any additional constraints on customers comfort level or quality of delivered product. De can be defined as the selection, planning, and implementation of measures intended to have an influence on the demand or customer-side of the electric meter, either caused directly or stimulated indirectly by the utility. DSM programs are peak clipping, Valley filling, Load shifting, Load building, energy conservation and flexible load shape. The main Target of this paper is to show the relation between DSM and Load Forecasting. Moreover, it highlights on the effect of applying DSM on Forecasted demands and how this affects the planning strategies for utility companies. This target will be clearly illustrated through applying the developed algorithm in this paper on an existing residential compound in Cairo-Egypt.展开更多
It is well known that the hot spot temperature represents the most critical parameter identifying the power rating of a transformer. This paper investigates the effect of the degradation of core magnetic properties on...It is well known that the hot spot temperature represents the most critical parameter identifying the power rating of a transformer. This paper investigates the effect of the degradation of core magnetic properties on temperature variation of aged oil-cooled transformers. Within this work, 2D accurate assessment of time average flux density distribution in an oil insulated-cooled 25 MVA transformer has been computed using finite-element analysis taking into account ageing and stress-induced non-uniform core permeability values. Knowing the core material specific loss and winding details, local core and winding losses are converted into heat. Based upon the ambient temperature outside the transformer tank and thermal heat transfer related factors, the detailed thermal modeling and analysis have then been carried out to determine temperature distribution everywhere. Analytical details and simulation results demonstrating effects of core magnetic properties degradation on hot spot temperatures of the transformer’s components are given in the paper.展开更多
This study presents the development of an ultrasonic transducer with a radius horn for an ultrasonic milling spindle(UMS)system.The ultrasonic transducer was intended to have a working frequency of approximately 30 kH...This study presents the development of an ultrasonic transducer with a radius horn for an ultrasonic milling spindle(UMS)system.The ultrasonic transducer was intended to have a working frequency of approximately 30 kHz.Two different materials were considered in the study:stainless steel(SS 316L)and titanium alloy(Ti-6Al-4V).Titanium alloy gave a higher resonance frequency(33 kHz)than stainless steel(30 kHz)under the same preload compression stress.An electromechanical impedance simulation was carried out to predict the impedance resonance frequency for both materials,and the effect of the overhanging toolbar was investigated.According to the electromechanical impedance simulation,the overhanging toolbar length affected the resonance frequency,and the error was less than 3%.Harmonic analysis confirmed that the damping ratio helps determine the resonance amplitude.Therefore,damping ratios of 0.015-0.020 and 0.005-0.020 were selected for stainless steel and titanium alloy,respectively,with an error of less than 1.5%.Experimental machining was also performed to assess the feasibility of ultrasonic-assisted milling;the result was a lesser cutting force and better surface topography of Al 6061.展开更多
The economic emission dispatch (EED) problem minimizes two competing objective functions, fuel cost and emission, while satisfying several equality and inequality constraints. Since the availability of wind power (WP)...The economic emission dispatch (EED) problem minimizes two competing objective functions, fuel cost and emission, while satisfying several equality and inequality constraints. Since the availability of wind power (WP) is highly dependent on the weather conditions, the inclusion of a significant amount of WP into EED will result in additional constraints to accommodate the intermittent nature of the output. In this paper, a new correlated bivariate Weibull probability distribution model is proposed to analytically remove the assumption that the total WP is characterized by a single random variable. This probability distribution is used as chance constraint. The inclusion of the probability distribution of stochastic WP in the EED problem is defined as the here-and-now strategy. Non-dominated sorting genetic algorithm built in MATLAB is used to handle the EED problem as a multi-objective optimization problem. A 69-bus ten-unit test system with non-smooth cost function is used to test the effectiveness of the proposed model.展开更多
This paper presents a novel Simulink models with an evaluation study of more widely used On-Line Maximum Power Point tracking(MPPT)techniques for Photo-Voltaic based Battery Storage Systems(PV-BSS).To have a full comp...This paper presents a novel Simulink models with an evaluation study of more widely used On-Line Maximum Power Point tracking(MPPT)techniques for Photo-Voltaic based Battery Storage Systems(PV-BSS).To have a full comparative study in terms of the dynamic response,battery state of charge(SOC),and oscillations around the Maximum Power Point(MPP)of the PV-BSS to variations in climate conditions,these techniques are simulated in Matlab/Simulink.The introduced methodologies are classified into two types;the first type is conventional hill-climbing techniques which are based on instantaneous PV data measurements such as Perturb&Observe and Incremental Conductance techniques.The second type is a novel proposed methodology is based on using solar irradiance and cell temperature measurements with pre-build Adaptive Neuro-Fuzzy Inference System(ANFIS)model to predict DC–DC converter optimum duty cycle to track MPP.Then evaluation study is introduced for conventional and proposed On-Line MPPT techniques.This comparative study can be useful in specifying the appropriateness of the MPPT techniques for PV-BSS.Also the introduced model can be used as a valued reference model for future research related to Soft Computing(SC)MPPT techniques.A significant improvement of SOC is achieved by the proposed model and methodology with high accuracy and lower oscillations.展开更多
Grease life refers to the time it takes for the grease to lose its ability to keep the lubrication due to grease degradation. As grease life is generally shorter than fatigue life of bearing, the service life of greas...Grease life refers to the time it takes for the grease to lose its ability to keep the lubrication due to grease degradation. As grease life is generally shorter than fatigue life of bearing, the service life of grease-lubricated rolling bearings is often dominated by grease life. When designing a bearing systemolecular weightith grease lubrication, it is necessary to define the operating conditions limits of the bearing, for which grease life becomes a determining factor. Prolongation of grease life becomes an especially important challenge when the bearing is to be operated trader high-speed, high-temperature, and other severe conditions. Selecting a number of commercially sold greases composed of varying base oils, the author evaluated their properties and analyzed how each property affected the grease life by performing a multiple regression analysis. The optimum grease composition to best exploit each property was also examined. The results revealed among others that one would need to first determine the base oil type and then maximize ultimate bleeding while minimizing the evaporation rate.展开更多
This research proposes a component to restrict dust from entering an oil hydraulic system through the rod-seal clearance of an oil hydraulic cylinder.The oil hydraulic cylinder as one of main parts of the hydraulic sy...This research proposes a component to restrict dust from entering an oil hydraulic system through the rod-seal clearance of an oil hydraulic cylinder.The oil hydraulic cylinder as one of main parts of the hydraulic system,controls position of load by reciprocation.For example,on construction machines such as excavators and graders,the cylinder controls position of folk lift,crane and bucket.However,during operation,dust enters the cylinder,wears seals,causes fluid degradation and affects lubrication of valves,pumps and other parts of hydraulic system.This increases breakdown rate of cylinder and entire system.Thus,it seems necessary to reduce on intrusion of dust into the system via the hydraulic cylinder.In this research,we made an experimental apparatus to simulate intrusion of the dust into system.Results proved that the apparatus is a suitable simulator to realize the intrusion.The proposed component to restrict dust from entering cylinder was fabricated and its performance tested when inserted with various elastic rings.The component gave tremendous results when inserted with O-ring seal and a plastic nylon washer,and can be retrofitted on new and old hydraulic cylinders.It is an appropriate technology especially in developing countries where dust is still a major concern.展开更多
The aim of the present work is to illustrate the application of mixed H2/H∞ control theory with Pole-Placement in de- signing controller for semi-active suspension system. It is well known that the ride comfort is im...The aim of the present work is to illustrate the application of mixed H2/H∞ control theory with Pole-Placement in de- signing controller for semi-active suspension system. It is well known that the ride comfort is improved by reducing vehicle body acceleration generated by road disturbance. In order to study this phenomenon, Two Degrees of Freedom (DOF) in state space vehicle model was built in. However, the role of H is to minimize the disturbance effect on the output while H2 is used to improve the input of controller. Linear Matrix Inequality (LMI) technique is used to calculate the dynamic controller parameters. The simulation results show that the H2 and H techniques can effectively control the vibration of vehicle system where the reduction of suspension working space, dynamic tire load and body acceleration. Moreover, the simulation results show that the (RMS) of suspension working space was reduced by 44.5%, body acceleration and dynamic tire load are reduced by 18.5% and 20% respectively.展开更多
基金funded by the Centre for Advanced Modelling and Geospatial Information Systems(CAMGIS),Faculty of Engineering and IT,University of Technology Sydneysupported by the Research Funding Program,King Saud University,Riyadh,Saudi Arabia,under Project Ongoing Research Funding program(ORF-2025-14).
文摘Diabetic Retinopathy(DR)is a critical disorder that affects the retina due to the constant rise in diabetics and remains the major cause of blindness across the world.Early detection and timely treatment are essential to mitigate the effects of DR,such as retinal damage and vision impairment.Several conventional approaches have been proposed to detect DR early and accurately,but they are limited by data imbalance,interpretability,overfitting,convergence time,and other issues.To address these drawbacks and improve DR detection accurately,a distributed Explainable Convolutional Neural network-enabled Light Gradient Boosting Machine(DE-ExLNN)is proposed in this research.The model combines an explainable Convolutional Neural Network(CNN)and Light Gradient Boosting Machine(LightGBM),achieving highly accurate outcomes in DR detection.LightGBM serves as the detection model,and the inclusion of an explainable CNN addresses issues that conventional CNN classifiers could not resolve.A custom dataset was created for this research,containing both fundus and OCTA images collected from a realtime environment,providing more accurate results compared to standard conventional DR datasets.The custom dataset demonstrates notable accuracy,sensitivity,specificity,and Matthews Correlation Coefficient(MCC)scores,underscoring the effectiveness of this approach.Evaluations against other standard datasets achieved an accuracy of 93.94%,sensitivity of 93.90%,specificity of 93.99%,and MCC of 93.88%for fundus images.For OCTA images,the results obtained an accuracy of 95.30%,sensitivity of 95.50%,specificity of 95.09%,andMCC of 95%.Results prove that the combination of explainable CNN and LightGBMoutperforms othermethods.The inclusion of distributed learning enhances the model’s efficiency by reducing time consumption and complexity while facilitating feature extraction.
基金funded by The National Natural Science Foundation of China under Grant(Nos.62273108 and 62306081)The Youth Project of Guangdong Artificial Intelligence and Digital Economy Laboratory(Guangzhou)(PZL2022KF0006)+6 种基金The National Key Research and Development Program-Research on Key technology of High Frequency broadband mobile communication credit Filter and its Industrialization application-Subproject Circuit Design and Simulation of high frequency broadband Filter(2022YFB3604502)‘New Generation Information Technology’Major Science and Technology Project of Guangzhou Key Field R&D Plan(202206070001)the Special Fund Project of Guangzhou Science and Technology Innovation Development(202201011307)the Guangdong Provincial Department of Education Key construction discipline Scientific research ability Improvement Project,Introduction of Talents Project of Guangdong Polytechnic Normal University of China(99166990222)the Special Projects in Key Fields of General Colleges and Universities in Guangdong Province(2021ZDZX1016)the Natural Science Foundation of Guangdong Province(2024A1515010120)the Special Fund Project of Guangzhou Science and Technology Innovation Development(202201011307).
文摘Underwater acoustic target recognition(UATR)has become increasingly prevalent for ocean detection,localisation,and identification.However,due to the complexity and variability of underwater environments,especially in multi ship event environments,where multiple acoustic signals coexist,practical applications face significant challenges.These challenges hinder single-category acoustic recognition algorithms,particularly in extracting time series features and achieving fine-grained or multi-scale feature fusion.This paper innovatively introduce the SKANN framework,which achieve precise submarine sound recognition in underwater mixed ship events environments through timing data enhancement and sampling training module and selective kernel feature extraction module.The timing data enhancement and sampling training module improves time sequence feature extraction through progressive acoustic sampling.The selective kernel feature extraction module effectively fuses multi-scale features by integrating selective kernel(SK)technology.To simulate concurrent ship events,we constructed the mixed ship noise dataset(MDeepShip),providing an experimental basis and test platform for underwater mixed ship event detection.This dataset ensures that the model encounters diverse audio samples during training and validation,improving its ability to extract temporal features.Experimental results show that SKANN achieves a 93.6%recognition rate on the M-DeepShip dataset,demonstrating its effectiveness in recognising underwater mixed ship events.Given the complexity of real underwater environments,this work lays a crucial foundation for the sound recognition of submarine vessels.Future research will focus on real marine environments to validate and refine the models and methods for practical applications.
基金supported by Open Research Projects of Zhejiang Lab(No.2022QA0AB02)Natural Science Foundation of Sichuan Province(2022NSFSC0913)Sichuan Province Selected Funding for Postdoctoral Research Projects(TB2022032).
文摘Vertical Federated Learning(VFL),which draws attention because of its ability to evaluate individuals based on features spread across multiple institutions,encounters numerous privacy and security threats.Existing solutions often suffer from centralized architectures,and exorbitant costs.To mitigate these issues,in this paper,we propose SecureVFL,a decentralized multi-party VFL scheme designed to enhance efficiency and trustworthiness while guaranteeing privacy.SecureVFL uses a permissioned blockchain and introduces a novel consensus algorithm,Proof of Feature Sharing(PoFS),to facilitate decentralized,trustworthy,and high-throughput federated training.SecureVFL introduces a verifiable and lightweight three-party Replicated Secret Sharing(RSS)protocol for feature intersection summation among overlapping users.Furthermore,we propose a(_(2)^(4))-sharing protocol to achieve federated training in a four-party VFL setting.This protocol involves only addition operations and exhibits robustness.SecureVFL not only enables anonymous interactions among participants but also safeguards their real identities,and provides mechanisms to unmask these identities when malicious activities are performed.We illustrate the proposed mechanism through a case study on VFL across four banks.Finally,our theoretical analysis proves the security of SecureVFL.Experiments demonstrated that SecureVFL outperformed existing multi-party VFL privacy-preserving schemes,such as MP-FedXGB,in terms of both overhead and model performance.
基金funded by the Centre for Advanced Modelling and Geospatial Information Systems(CAMGIS),Faculty of Engineering and IT,University of Technology Sydneysupported by the Researchers Supporting Project,King Saud University,Riyadh,Saudi Arabia,under Project RSP2025 R14.
文摘Electronic nose and thermal images are effective ways to diagnose the presence of gases in real-time realtime.Multimodal fusion of these modalities can result in the development of highly accurate diagnostic systems.The low-cost thermal imaging software produces low-resolution thermal images in grayscale format,hence necessitating methods for improving the resolution and colorizing the images.The objective of this paper is to develop and train a super-resolution generative adversarial network for improving the resolution of the thermal images,followed by a sparse autoencoder for colorization of thermal images and amultimodal convolutional neural network for gas detection using electronic nose and thermal images.The dataset used comprises 6400 thermal images and electronic nose measurements for four classes.A multimodal Convolutional Neural Network(CNN)comprising an EfficientNetB2 pre-trainedmodel was developed using both early and late feature fusion.The Super Resolution Generative Adversarial Network(SRGAN)model was developed and trained on low and high-resolution thermal images.Asparse autoencoder was trained on the grayscale and colorized thermal images.The SRGAN was trained on lowand high-resolution thermal images,achieving a Structural Similarity Index(SSIM)of 90.28,a Peak Signal-to-Noise Ratio(PSNR)of 68.74,and a Mean Absolute Error(MAE)of 0.066.The autoencoder model produced an MAE of 0.035,a Mean Squared Error(MSE)of 0.006,and a Root Mean Squared Error(RMSE)of 0.0705.The multimodal CNN,trained on these images and electronic nose measurements using both early and late fusion techniques,achieved accuracies of 97.89% and 98.55%,respectively.Hence,the proposed framework can be of great aid for the integration with low-cost software to generate high quality thermal camera images and highly accurate detection of gases in real-time.
基金funded by the Centre for Advanced Modelling and Geospatial Information Systems(CAMGIS),Faculty of Engineering and IT,University of Technology SydneyMoreover,supported by the Researchers Supporting Project,King Saud University,Riyadh,Saudi Arabia,under Ongoing Research Funding(ORF-2025-14).
文摘The development of generative architectures has resulted in numerous novel deep-learning models that generate images using text inputs.However,humans naturally use speech for visualization prompts.Therefore,this paper proposes an architecture that integrates speech prompts as input to image-generation Generative Adversarial Networks(GANs)model,leveraging Speech-to-Text translation along with the CLIP+VQGAN model.The proposed method involves translating speech prompts into text,which is then used by the Contrastive Language-Image Pretraining(CLIP)+Vector Quantized Generative Adversarial Network(VQGAN)model to generate images.This paper outlines the steps required to implement such a model and describes in detail the methods used for evaluating the model.The GAN model successfully generates artwork from descriptions using speech and text prompts.Experimental outcomes of synthesized images demonstrate that the proposed methodology can produce beautiful abstract visuals containing elements from the input prompts.The model achieved a Frechet Inception Distance(FID)score of 28.75,showcasing its capability to produce high-quality and diverse images.The proposed model can find numerous applications in educational,artistic,and design spaces due to its ability to generate images using speech and the distinct abstract artistry of the output images.This capability is demonstrated by giving the model out-of-the-box prompts to generate never-before-seen images with plausible realistic qualities.
基金supported by the Department of Science and Technology,Science and Engineering Research Board,New Delhi,India,under Grant No.EEQ/2022/000812.
文摘Landslide hazard detection is a prevalent problem in remote sensing studies,particularly with the technological advancement of computer vision.With the continuous and exceptional growth of the computational environment,the manual and partially automated procedure of landslide detection from remotely sensed images has shifted toward automatic methods with deep learning.Furthermore,attention models,driven by human visual procedures,have become vital in natural hazard-related studies.Hence,this paper proposes an enhanced YOLOv5(You Only Look Once version 5)network for improved satellite-based landslide detection,embedded with two popular attention modules:CBAM(Convolutional Block Attention Module)and ECA(Efficient Channel Attention).These attention mechanisms are incorporated into the backbone and neck of the YOLOv5 architecture,distinctly,and evaluated across three YOLOv5 variants:nano(n),small(s),and medium(m).The experiments use opensource satellite images from three distinct regions with complex terrain.The standard metrics,including F-score,precision,recall,and mean average precision(mAP),are computed for quantitative assessment.The YOLOv5n+CBAM demonstrates the most optimal results with an F-score of 77.2%,confirming its effectiveness.The suggested attention-driven architecture augments detection accuracy,supporting post-landslide event assessment and recovery.
基金supported by the Deanship of Scientific Research at Prince Sattam Bin Abdulaziz University under the research project No.2020/01/11742.
文摘Hydrocarbons,carbon monoxide and other pollutants from the transportation sector harm human health in many ways.Fuel cell(FC)has been evolving rapidly over the past two decades due to its efficient mechanism to transform the chemical energy in hydrogen-rich compounds into electrical energy.The main drawback of the standalone FC is its slow dynamic response and its inability to supply rapid variations in the load demand.Therefore,adding energy storage systems is necessary.However,to manage and distribute the power-sharing among the hybrid proton exchange membrane(PEM)fuel cell(FC),battery storage(BS),and supercapacitor(SC),an energy management strategy(EMS)is essential.In this research work,an optimal EMS based on a spotted hyena optimizer(SHO)for hybrid PEM fuel cell/BS/SC is proposed.The main goal of an EMS is to improve the performance of hybrid FC/BS/SC and to reduce the amount of hydrogen consumption.To prove the superiority of the SHO method,the obtained results are compared with the chimp optimizer(CO),the artificial ecosystem-based optimizer(AEO),the seagull optimization algorithm(SOA),the sooty tern optimization algorithm(STOA),and the coyote optimization algorithm(COA).Two main metrics are used as a benchmark for the comparison:the minimum consumed hydrogen and the efficiency of the system.The main findings confirm that the minimum amount of hydrogen consumption and maximum efficiency are achieved by the proposed SHO based EMS.
文摘In computer vision,emotion recognition using facial expression images is considered an important research issue.Deep learning advances in recent years have aided in attaining improved results in this issue.According to recent studies,multiple facial expressions may be included in facial photographs representing a particular type of emotion.It is feasible and useful to convert face photos into collections of visual words and carry out global expression recognition.The main contribution of this paper is to propose a facial expression recognitionmodel(FERM)depending on an optimized Support Vector Machine(SVM).To test the performance of the proposed model(FERM),AffectNet is used.AffectNet uses 1250 emotion-related keywords in six different languages to search three major search engines and get over 1,000,000 facial photos online.The FERM is composed of three main phases:(i)the Data preparation phase,(ii)Applying grid search for optimization,and(iii)the categorization phase.Linear discriminant analysis(LDA)is used to categorize the data into eight labels(neutral,happy,sad,surprised,fear,disgust,angry,and contempt).Due to using LDA,the performance of categorization via SVM has been obviously enhanced.Grid search is used to find the optimal values for hyperparameters of SVM(C and gamma).The proposed optimized SVM algorithm has achieved an accuracy of 99%and a 98%F1 score.
文摘Recently, gears of high strength, reliability, and surface-damage-resistant under severe service conditions are required to achieve the weight saving and downsizing of a product. For the high-speed condition in particular, it is important to understand the influence of the surface properties on the scuffing resistance. If the effective surface profile to improve the lubrication property was found, the metal surfaces could be obtained with both surface strength and surface lubricity. Herein, the influence of surface properties modified with fine shot peening, which can form the arbitrary surface profile, on the scuffing resistance in the rolling-sliding contact machine element, was investigated. The scuffing test was performed using a two-cylinder rolling contact test machine. In a specific sliding, a faster roller of 60% and a sliding velocity of 1.75 m/s were utilized. The scuffing test results with shot-peened test rollers and those with non-shot-peened test roller were compared. The influence of the surface roughness of the shot-peened test roller was also discussed. We found that the shot-peened roller had a better scuffing resistance compared with the roller without the shot-peening process.
文摘The electric power infrastructure that has served huge loads for so long is rapidly running up against many limitations. Out of many challenges it is to operate the power system in secure manner so that the operation constraints are fulfilled under both normal and contingent conditions. Smart grid technology offers valuable techniques that can be deployed within the very near future or which are already deployed nowadays. Flexible AC Transmission Systems (FACTS) devices have been introduced to solve various power system problems. In literature, most of the methods proposed for sizing the FACTS devices only consider the normal operating conditions of power systems. Consequently, some transmission lines are heavily loaded in contingency case and the system voltage stability becomes a power transfer-limiting factor. This paper presents a technique for determining the proper rating/size of FACTS devices, namely the Static Synchronous Compensator (STATCOM), while considering contingency cases. The paper also verifies that the weakest bus determined by eigenvalue and eigenvectors method is the best location for STATCOM. The rating of STATCOM is specified according to the required reactive power needed to improve voltage stability under normal and contingency cases. Two case system studies are investigated: a simple 5-bus system and the IEEE 14-bus system. The obtained results verify that the rating of STATCOM can be determined according to the worst contingency case, and through proper control it can still be effective for normal and other contingency cases.
文摘In this paper, in order to design a cam mechanism be up to the mustard, a set of methods are put forward that using the Visual Basic programming language based on solidworks to draw cam contour line and then get its 3D models and generate the cam motion simulation by the solidworks motion. In the end, it’s proved that the cam designed though this method met the requirement.
文摘Utilities around the world have been considering Demand Side Management (DSM) in their strategic planning. The costs of constructing and operating a new capacity generation unit are increasing everyday as well as Transmission and distribution and land issues for new generation plants, which force the utilities to search for another alternatives without any additional constraints on customers comfort level or quality of delivered product. De can be defined as the selection, planning, and implementation of measures intended to have an influence on the demand or customer-side of the electric meter, either caused directly or stimulated indirectly by the utility. DSM programs are peak clipping, Valley filling, Load shifting, Load building, energy conservation and flexible load shape. The main Target of this paper is to show the relation between DSM and Load Forecasting. Moreover, it highlights on the effect of applying DSM on Forecasted demands and how this affects the planning strategies for utility companies. This target will be clearly illustrated through applying the developed algorithm in this paper on an existing residential compound in Cairo-Egypt.
文摘It is well known that the hot spot temperature represents the most critical parameter identifying the power rating of a transformer. This paper investigates the effect of the degradation of core magnetic properties on temperature variation of aged oil-cooled transformers. Within this work, 2D accurate assessment of time average flux density distribution in an oil insulated-cooled 25 MVA transformer has been computed using finite-element analysis taking into account ageing and stress-induced non-uniform core permeability values. Knowing the core material specific loss and winding details, local core and winding losses are converted into heat. Based upon the ambient temperature outside the transformer tank and thermal heat transfer related factors, the detailed thermal modeling and analysis have then been carried out to determine temperature distribution everywhere. Analytical details and simulation results demonstrating effects of core magnetic properties degradation on hot spot temperatures of the transformer’s components are given in the paper.
基金supported by Korea Electrotechnology Research Institute(KERI)Primary Research Program through the National Research Council of Science&Technology(NST)funded by the Ministry of Science and ICT(MSIT)in 2023(No.23A01021)the National Research Foundation of Korea(NRF)grant funded by the Korean Government(MSIT)(No.RS-2023-00278890).
文摘This study presents the development of an ultrasonic transducer with a radius horn for an ultrasonic milling spindle(UMS)system.The ultrasonic transducer was intended to have a working frequency of approximately 30 kHz.Two different materials were considered in the study:stainless steel(SS 316L)and titanium alloy(Ti-6Al-4V).Titanium alloy gave a higher resonance frequency(33 kHz)than stainless steel(30 kHz)under the same preload compression stress.An electromechanical impedance simulation was carried out to predict the impedance resonance frequency for both materials,and the effect of the overhanging toolbar was investigated.According to the electromechanical impedance simulation,the overhanging toolbar length affected the resonance frequency,and the error was less than 3%.Harmonic analysis confirmed that the damping ratio helps determine the resonance amplitude.Therefore,damping ratios of 0.015-0.020 and 0.005-0.020 were selected for stainless steel and titanium alloy,respectively,with an error of less than 1.5%.Experimental machining was also performed to assess the feasibility of ultrasonic-assisted milling;the result was a lesser cutting force and better surface topography of Al 6061.
文摘The economic emission dispatch (EED) problem minimizes two competing objective functions, fuel cost and emission, while satisfying several equality and inequality constraints. Since the availability of wind power (WP) is highly dependent on the weather conditions, the inclusion of a significant amount of WP into EED will result in additional constraints to accommodate the intermittent nature of the output. In this paper, a new correlated bivariate Weibull probability distribution model is proposed to analytically remove the assumption that the total WP is characterized by a single random variable. This probability distribution is used as chance constraint. The inclusion of the probability distribution of stochastic WP in the EED problem is defined as the here-and-now strategy. Non-dominated sorting genetic algorithm built in MATLAB is used to handle the EED problem as a multi-objective optimization problem. A 69-bus ten-unit test system with non-smooth cost function is used to test the effectiveness of the proposed model.
基金The Deanship of Scientific Research at Najran University has supported this work,under the General Research Funding program grant code(NU/-/SERC/10/650).
文摘This paper presents a novel Simulink models with an evaluation study of more widely used On-Line Maximum Power Point tracking(MPPT)techniques for Photo-Voltaic based Battery Storage Systems(PV-BSS).To have a full comparative study in terms of the dynamic response,battery state of charge(SOC),and oscillations around the Maximum Power Point(MPP)of the PV-BSS to variations in climate conditions,these techniques are simulated in Matlab/Simulink.The introduced methodologies are classified into two types;the first type is conventional hill-climbing techniques which are based on instantaneous PV data measurements such as Perturb&Observe and Incremental Conductance techniques.The second type is a novel proposed methodology is based on using solar irradiance and cell temperature measurements with pre-build Adaptive Neuro-Fuzzy Inference System(ANFIS)model to predict DC–DC converter optimum duty cycle to track MPP.Then evaluation study is introduced for conventional and proposed On-Line MPPT techniques.This comparative study can be useful in specifying the appropriateness of the MPPT techniques for PV-BSS.Also the introduced model can be used as a valued reference model for future research related to Soft Computing(SC)MPPT techniques.A significant improvement of SOC is achieved by the proposed model and methodology with high accuracy and lower oscillations.
文摘Grease life refers to the time it takes for the grease to lose its ability to keep the lubrication due to grease degradation. As grease life is generally shorter than fatigue life of bearing, the service life of grease-lubricated rolling bearings is often dominated by grease life. When designing a bearing systemolecular weightith grease lubrication, it is necessary to define the operating conditions limits of the bearing, for which grease life becomes a determining factor. Prolongation of grease life becomes an especially important challenge when the bearing is to be operated trader high-speed, high-temperature, and other severe conditions. Selecting a number of commercially sold greases composed of varying base oils, the author evaluated their properties and analyzed how each property affected the grease life by performing a multiple regression analysis. The optimum grease composition to best exploit each property was also examined. The results revealed among others that one would need to first determine the base oil type and then maximize ultimate bleeding while minimizing the evaporation rate.
文摘This research proposes a component to restrict dust from entering an oil hydraulic system through the rod-seal clearance of an oil hydraulic cylinder.The oil hydraulic cylinder as one of main parts of the hydraulic system,controls position of load by reciprocation.For example,on construction machines such as excavators and graders,the cylinder controls position of folk lift,crane and bucket.However,during operation,dust enters the cylinder,wears seals,causes fluid degradation and affects lubrication of valves,pumps and other parts of hydraulic system.This increases breakdown rate of cylinder and entire system.Thus,it seems necessary to reduce on intrusion of dust into the system via the hydraulic cylinder.In this research,we made an experimental apparatus to simulate intrusion of the dust into system.Results proved that the apparatus is a suitable simulator to realize the intrusion.The proposed component to restrict dust from entering cylinder was fabricated and its performance tested when inserted with various elastic rings.The component gave tremendous results when inserted with O-ring seal and a plastic nylon washer,and can be retrofitted on new and old hydraulic cylinders.It is an appropriate technology especially in developing countries where dust is still a major concern.
文摘The aim of the present work is to illustrate the application of mixed H2/H∞ control theory with Pole-Placement in de- signing controller for semi-active suspension system. It is well known that the ride comfort is improved by reducing vehicle body acceleration generated by road disturbance. In order to study this phenomenon, Two Degrees of Freedom (DOF) in state space vehicle model was built in. However, the role of H is to minimize the disturbance effect on the output while H2 is used to improve the input of controller. Linear Matrix Inequality (LMI) technique is used to calculate the dynamic controller parameters. The simulation results show that the H2 and H techniques can effectively control the vibration of vehicle system where the reduction of suspension working space, dynamic tire load and body acceleration. Moreover, the simulation results show that the (RMS) of suspension working space was reduced by 44.5%, body acceleration and dynamic tire load are reduced by 18.5% and 20% respectively.
