Because the hydraulic directional valve usually works in a bad working environment and is disturbed by multi-factor noise,the traditional single sensor monitoring technology is difficult to use for an accurate diagnos...Because the hydraulic directional valve usually works in a bad working environment and is disturbed by multi-factor noise,the traditional single sensor monitoring technology is difficult to use for an accurate diagnosis of it.Therefore,a fault diagnosis method based on multi-sensor information fusion is proposed in this paper to reduce the inaccuracy and uncertainty of traditional single sensor information diagnosis technology and to realize accurate monitoring for the location or diagnosis of early faults in such valves in noisy environments.Firstly,the statistical features of signals collected by the multi-sensor are extracted and the depth features are obtained by a convolutional neural network(CNN)to form a complete and stable multi-dimensional feature set.Secondly,to obtain a weighted multi-dimensional feature set,the multi-dimensional feature sets of similar sensors are combined,and the entropy weight method is used to weight these features to reduce the interference of insensitive features.Finally,the attention mechanism is introduced to improve the dual-channel CNN,which is used to adaptively fuse the weighted multi-dimensional feature sets of heterogeneous sensors,to flexibly select heterogeneous sensor information so as to achieve an accurate diagnosis.Experimental results show that the weighted multi-dimensional feature set obtained by the proposed method has a high fault-representation ability and low information redundancy.It can diagnose simultaneously internal wear faults of the hydraulic directional valve and electromagnetic faults of actuators that are difficult to diagnose by traditional methods.This proposed method can achieve high fault-diagnosis accuracy under severe working conditions.展开更多
This article explores the design of a wireless fire alarm system supported by advanced data fusion technology.It includes discussions on the basic design ideas of the wireless fire alarm system,hardware design analysi...This article explores the design of a wireless fire alarm system supported by advanced data fusion technology.It includes discussions on the basic design ideas of the wireless fire alarm system,hardware design analysis,software design analysis,and simulation analysis,all supported by data fusion technology.Hopefully,this analysis can provide some reference for the rational application of data fusion technology to meet the actual design and application requirements of the system.展开更多
As a complex engineering problem,the satellite module layout design (SMLD) is difficult to resolve by using conventional computation-based approaches. The challenges stem from three aspects:computational complexity,en...As a complex engineering problem,the satellite module layout design (SMLD) is difficult to resolve by using conventional computation-based approaches. The challenges stem from three aspects:computational complexity,engineering complexity,and engineering practicability. Engineers often finish successful satellite designs by way of their plenty of experience and wisdom,lessons learnt from the past practices,as well as the assistance of the advanced computational techniques. Enlightened by the ripe patterns,th...展开更多
Metal additive manufacturing(AM)has been extensively studied in recent decades.Despite the significant progress achieved in manufacturing complex shapes and structures,challenges such as severe cracking when using exi...Metal additive manufacturing(AM)has been extensively studied in recent decades.Despite the significant progress achieved in manufacturing complex shapes and structures,challenges such as severe cracking when using existing alloys for laser powder bed fusion(L-PBF)AM have persisted.These challenges arise because commercial alloys are primarily designed for conventional casting or forging processes,overlooking the fast cooling rates,steep temperature gradients and multiple thermal cycles of L-PBF.To address this,there is an urgent need to develop novel alloys specifically tailored for L-PBF technologies.This review provides a comprehensive summary of the strategies employed in alloy design for L-PBF.It aims to guide future research on designing novel alloys dedicated to L-PBF instead of adapting existing alloys.The review begins by discussing the features of the L-PBF processes,focusing on rapid solidification and intrinsic heat treatment.Next,the printability of the four main existing alloys(Fe-,Ni-,Al-and Ti-based alloys)is critically assessed,with a comparison of their conventional weldability.It was found that the weldability criteria are not always applicable in estimating printability.Furthermore,the review presents recent advances in alloy development and associated strategies,categorizing them into crack mitigation-oriented,microstructure manipulation-oriented and machine learning-assisted approaches.Lastly,an outlook and suggestions are given to highlight the issues that need to be addressed in future work.展开更多
When evaluating the seismic safety and reliability of complex engineering structures,it is a critical problem to reasonably consider the randomness and multi-dimensional nature of ground motions.To this end,a proposed...When evaluating the seismic safety and reliability of complex engineering structures,it is a critical problem to reasonably consider the randomness and multi-dimensional nature of ground motions.To this end,a proposed modeling strategy of multi-dimensional stochastic earthquakes is addressed in this study.This improved seismic model has several merits that enable it to better provide seismic analyses of structures.Specifically,at first,the ground motion model is compatible with the design response spectrum.Secondly,the evolutionary power spectrum involved in the model and the design response spectrum are constructed accordingly with sufficient consideration of the correlation between different seismic components.Thirdly,the random function-based dimension-reduction representation is applied,by which seismic modeling is established,with three elementary random variables.Numerical simulations of multi-dimensional stochastic ground motions in a specific design scenario indicate the effectiveness of the proposed modeling strategy.Moreover,the multi-dimensional seismic response and the global reliability of a high-rise frame-core tube structure is discussed in detail to further illustrate the engineering applicability of the proposed method.The analytical investigations demonstrate that the suggested stochastic model of multi-dimensional ground motion is available for accurate seismic response analysis and dynamic reliability assessment of complex engineering structures for performance-based seismic resistance design.展开更多
Intelligence and perception are two operative technologies in 6G scenarios.The intelligent wireless network and information perception require a deep fusion of artificial intelligence(AI)and wireless communications in...Intelligence and perception are two operative technologies in 6G scenarios.The intelligent wireless network and information perception require a deep fusion of artificial intelligence(AI)and wireless communications in 6G systems.Therefore,fusion is becoming a typical feature and key challenge of 6G wireless communication systems.In this paper,we focus on the critical issues and propose three application scenarios in 6G wireless systems.Specifically,we first discuss the fusion of AI and 6G networks for the enhancement of 5G-advanced technology and future wireless communication systems.Then,we introduce the wireless AI technology architecture with 6G multidimensional information perception,which includes the physical layer technology of multi-dimensional feature information perception,full spectrum fusion technology,and intelligent wireless resource management.The discussion of key technologies for intelligent 6G wireless network networks is expected to provide a guideline for future research.展开更多
By analyzing the shortage of reliability test design and thinking over the producer's risk and consumer's risk, the information fusion technology is used to set up a reliability test design model( RTDM). By an...By analyzing the shortage of reliability test design and thinking over the producer's risk and consumer's risk, the information fusion technology is used to set up a reliability test design model( RTDM). By analyzing the demands and constraint conditions of the RTDM and with applications of Bayesian approach and Monte Carlo method( MCM),this paper puts forward the exponential distributed subsystems and the information fusion technology among them. According to the posteriori risk criteria,formulas of producer's risk and consumer's risk were also inferred,and with the help of Matlab software,selection of the optimum test plan was solved. Finally,validity of the model had been proved by a test of series parallel system.展开更多
An effective breeding blanket is critical to support tritium self-sufficiency for future fusion reactors.The difficulty is to achieve tritium breeding ratio(TBR)target of 1.05 or more.This paper presents a new design ...An effective breeding blanket is critical to support tritium self-sufficiency for future fusion reactors.The difficulty is to achieve tritium breeding ratio(TBR)target of 1.05 or more.This paper presents a new design approach to the blanket design process.It indicates that fusion blanket design is affected by universal functions based on iterations.Three aspects are worth more attention from fusion engineers in the future.The first factor is that the iterations on the material fractions affect not only structure scheme but also TBR variation.The second factor is the cooling condition affecting final TBR due to the change of the structure material proportion.The third factor is temperature field related to the tritium release.In particular,it is suggested that the statistical calculation of effective TBR must be under reasonable control of the blanket temperature field.This approach is novel for blanket engineering in development of a fusion reactor.展开更多
Although existing style transfer techniques have made significant progress in the field of image generation,there are still some challenges in the field of exhibition hall design.The existing style transfer methods ma...Although existing style transfer techniques have made significant progress in the field of image generation,there are still some challenges in the field of exhibition hall design.The existing style transfer methods mainly focus on the transformation of single dimensional features,but ignore the deep integration of content and style features in exhibition hall design.In addition,existing methods are deficient in detail retention,especially in accurately capturing and reproducing local textures and details while preserving the content image structure.In addition,point-based attention mechanisms tend to ignore the complexity and diversity of image features in multi-dimensional space,resulting in alignment problems between features in different semantic areas,resulting in inconsistent stylistic features in content areas.In this context,this paper proposes a semantic-enhanced multimodal style transfer algorithm tailored for exhibition hall design.The proposed approach leverages a multimodal encoder architecture to integrate information from text,source images,and style images,using separate encoder modules for each modality to capture shallow,deep,and semantic features.A novel Style Transfer Convolution(STConv)convolutional kernel,based on the Visual Geometry Group(VGG)19 network,is introduced to improve feature extraction in style transfer.Additionally,an enhanced Transformer encoder is incorporated to capture contextual semantic information within images,while the CLIP model is employed for text data processing.A hybrid attention module is designed to precisely capture style features,achieving multimodal feature fusion via a diffusion model that generates exhibition hall design images aligned with stylistic requirements.Quantitative experiments show that compared with the most advanced algorithms,the proposed method has achieved significant performance improvement on both Fréchet Inception Distance(FID)and Kernel Inception Distance(KID)indexes.For example,on the ExpoArchive dataset,the proposed method has a FID value of 87.9 and a KID value of 1.98,which is significantly superior to other methods.展开更多
The task of student action recognition in the classroom is to precisely capture and analyze the actions of students in classroom videos,providing a foundation for realizing intelligent and accurate teaching.However,th...The task of student action recognition in the classroom is to precisely capture and analyze the actions of students in classroom videos,providing a foundation for realizing intelligent and accurate teaching.However,the complex nature of the classroom environment has added challenges and difficulties in the process of student action recognition.In this research article,with regard to the circumstances where students are prone to be occluded and classroom computing resources are restricted in real classroom scenarios,a lightweight multi-modal fusion action recognition approach is put forward.This proposed method is capable of enhancing the accuracy of student action recognition while concurrently diminishing the number of parameters of the model and the Computation Amount,thereby achieving a more efficient and accurate recognition performance.In the feature extraction stage,this method fuses the keypoint heatmap with the RGB(Red-Green-Blue color model)image.In order to fully utilize the unique information of different modalities for feature complementarity,a Feature Fusion Module(FFE)is introduced.The FFE encodes and fuses the unique features of the two modalities during the feature extraction process.This fusion strategy not only achieves fusion and complementarity between modalities,but also improves the overall model performance.Furthermore,to reduce the computational load and parameter scale of the model,we use keypoint information to crop RGB images.At the same time,the first three networks of the lightweight feature extraction network X3D are used to extract dual-branch features.These methods significantly reduce the computational load and parameter scale.The number of parameters of the model is 1.40 million,and the computation amount is 5.04 billion floating-point operations per second(GFLOPs),achieving an efficient lightweight design.In the Student Classroom Action Dataset(SCAD),the accuracy of the model is 88.36%.In NTU 60(Nanyang Technological University Red-Green-Blue-Depth RGB+Ddataset with 60 categories),the accuracies on X-Sub(The people in the training set are different from those in the test set)and X-View(The perspectives of the training set and the test set are different)are 95.76%and 98.82%,respectively.On the NTU 120 dataset(Nanyang Technological University Red-Green-Blue-Depth dataset with 120 categories),RGB+Dthe accuracies on X-Sub and X-Set(the perspectives of the training set and the test set are different)are 91.97%and 93.45%,respectively.The model has achieved a balance in terms of accuracy,computation amount,and the number of parameters.展开更多
Powder bed fusion(PBF)in metallic additive manufacturing offers the ability to produce intricate geometries,high-strength components,and reliable products.However,powder processing before energy-based binding signific...Powder bed fusion(PBF)in metallic additive manufacturing offers the ability to produce intricate geometries,high-strength components,and reliable products.However,powder processing before energy-based binding significantly impacts the final product’s integrity.Processing maps guide efficient process design to minimize defects,but creating them through experimentation alone is challenging due to the wide range of parameters,necessitating a comprehensive computational parametric analysis.In this study,we used the discrete element method to parametrically analyze the powder processing design space in PBF of stainless steel 316L powders.Uniform lattice parameter sweeps are often used for parametric analysis,but are computationally intensive.We find that non-uniform parameter sweep based on the low discrepancy sequence(LDS)algorithm is ten times more efficient at exploring the design space while accurately capturing the relationship between powder flow dynamics and bed packing density.We introduce a multi-layer perceptron(MLP)model to interpolate parametric causalities within the LDS parameter space.With over 99%accuracy,it effectively captures these causalities while requiring fewer simulations.Finally,we generate processing design maps for machine setups and powder selections for efficient process design.We find that recoating speed has the highest impact on powder processing quality,followed by recoating layer thickness,particle size,and inter-particle friction.展开更多
This paper proposes a class of novel progressive edge growth-based codebooks for downlink sparse code multiple access(SCMA)systems.In the first scheme,we propose to progressively design the codebooks of each resource ...This paper proposes a class of novel progressive edge growth-based codebooks for downlink sparse code multiple access(SCMA)systems.In the first scheme,we propose to progressively design the codebooks of each resource node(RN)instead of rotating a mother constellation(MC)as in the conventional SCMA works.In the other one,based on the MC,a multi-resources rotated codebooks are proposed to improve the performance of the superimposed constellations.The resultant codebooks are respectively referred to as the resource edge multidimensional codebooks(REMC)and the user edge multi-dimensional codebooks(UEMC).Additionally,we delve into the detailed design of the MC and the superimposed constellation.Then,we pay special attention to the application of the proposed schemes to challenging design cases,particularly for the high dimensional,high rate,and irregular codebooks,where the corresponding simplified schemes are proposed to reduce the complexity of codebook design.Finally,simulation results are presented to demonstrate the superiority of our progressive edge growth-based schemes.The numerical results indicate that the proposed codebooks significantly outperform the stateof-the-art codebooks.In addition,we also show that the proposed REMC codebooks outperform in the lower signal-to-noise ratio(SNR)regime,whereas the UEMC codebooks exhibit better performance at higher SNRs.展开更多
目的/意义分析多模态融合技术在人工智能(artificial intelligence,AI)制药核心场景应用的关键价值,为制药领域创新发展提供理论与实践参考。方法/过程采用文献综述方法,在中国知网、PubMed和Web of Science检索相关文献,系统归纳多模...目的/意义分析多模态融合技术在人工智能(artificial intelligence,AI)制药核心场景应用的关键价值,为制药领域创新发展提供理论与实践参考。方法/过程采用文献综述方法,在中国知网、PubMed和Web of Science检索相关文献,系统归纳多模态融合技术在AI制药领域的研究进展与应用实践。结果/结论 多模态融合技术在靶点识别、从头药物设计等核心场景的应用成效显著,但仍面临模态对齐精度不足、模型解释性较弱等挑战,未来将向高效模型架构与实际应用场景深度融合方向发展。展开更多
基金supported by the National Natural Science Foundation of China(Nos.51805376 and U1709208)the Zhejiang Provincial Natural Science Foundation of China(Nos.LY20E050028 and LD21E050001)。
文摘Because the hydraulic directional valve usually works in a bad working environment and is disturbed by multi-factor noise,the traditional single sensor monitoring technology is difficult to use for an accurate diagnosis of it.Therefore,a fault diagnosis method based on multi-sensor information fusion is proposed in this paper to reduce the inaccuracy and uncertainty of traditional single sensor information diagnosis technology and to realize accurate monitoring for the location or diagnosis of early faults in such valves in noisy environments.Firstly,the statistical features of signals collected by the multi-sensor are extracted and the depth features are obtained by a convolutional neural network(CNN)to form a complete and stable multi-dimensional feature set.Secondly,to obtain a weighted multi-dimensional feature set,the multi-dimensional feature sets of similar sensors are combined,and the entropy weight method is used to weight these features to reduce the interference of insensitive features.Finally,the attention mechanism is introduced to improve the dual-channel CNN,which is used to adaptively fuse the weighted multi-dimensional feature sets of heterogeneous sensors,to flexibly select heterogeneous sensor information so as to achieve an accurate diagnosis.Experimental results show that the weighted multi-dimensional feature set obtained by the proposed method has a high fault-representation ability and low information redundancy.It can diagnose simultaneously internal wear faults of the hydraulic directional valve and electromagnetic faults of actuators that are difficult to diagnose by traditional methods.This proposed method can achieve high fault-diagnosis accuracy under severe working conditions.
基金Chongqing Engineering University Undergraduate Innovation and Entrepreneurship Training Program Project:Wireless Fire Automatic Alarm System(Project No.:CXCY2024017)Chongqing Municipal Education Commission Science and Technology Research Project:Development and Research of Chongqing Wireless Fire Automatic Alarm System(Project No.:KJQN202401906)。
文摘This article explores the design of a wireless fire alarm system supported by advanced data fusion technology.It includes discussions on the basic design ideas of the wireless fire alarm system,hardware design analysis,software design analysis,and simulation analysis,all supported by data fusion technology.Hopefully,this analysis can provide some reference for the rational application of data fusion technology to meet the actual design and application requirements of the system.
基金National Natural Science Foundation of China (50575031, 50275019)National High-tech Research and Development Program (2006AA04Z109)
文摘As a complex engineering problem,the satellite module layout design (SMLD) is difficult to resolve by using conventional computation-based approaches. The challenges stem from three aspects:computational complexity,engineering complexity,and engineering practicability. Engineers often finish successful satellite designs by way of their plenty of experience and wisdom,lessons learnt from the past practices,as well as the assistance of the advanced computational techniques. Enlightened by the ripe patterns,th...
基金financially supported by the National Key Research and Development Program of China(2022YFB4600302)National Natural Science Foundation of China(52090041)+1 种基金National Natural Science Foundation of China(52104368)National Major Science and Technology Projects of China(J2019-VII-0010-0150)。
文摘Metal additive manufacturing(AM)has been extensively studied in recent decades.Despite the significant progress achieved in manufacturing complex shapes and structures,challenges such as severe cracking when using existing alloys for laser powder bed fusion(L-PBF)AM have persisted.These challenges arise because commercial alloys are primarily designed for conventional casting or forging processes,overlooking the fast cooling rates,steep temperature gradients and multiple thermal cycles of L-PBF.To address this,there is an urgent need to develop novel alloys specifically tailored for L-PBF technologies.This review provides a comprehensive summary of the strategies employed in alloy design for L-PBF.It aims to guide future research on designing novel alloys dedicated to L-PBF instead of adapting existing alloys.The review begins by discussing the features of the L-PBF processes,focusing on rapid solidification and intrinsic heat treatment.Next,the printability of the four main existing alloys(Fe-,Ni-,Al-and Ti-based alloys)is critically assessed,with a comparison of their conventional weldability.It was found that the weldability criteria are not always applicable in estimating printability.Furthermore,the review presents recent advances in alloy development and associated strategies,categorizing them into crack mitigation-oriented,microstructure manipulation-oriented and machine learning-assisted approaches.Lastly,an outlook and suggestions are given to highlight the issues that need to be addressed in future work.
基金National Natural Science Foundation of China under Grant Nos.51978543,52108444,and 51778343Plan of Outstanding Young and Middle-aged Scientific and Technological Innovation Team in the Universities of Hubei Province with Project No.T2020010Natural Science Foundation of Hebei Province under Grant No.E2021512001。
文摘When evaluating the seismic safety and reliability of complex engineering structures,it is a critical problem to reasonably consider the randomness and multi-dimensional nature of ground motions.To this end,a proposed modeling strategy of multi-dimensional stochastic earthquakes is addressed in this study.This improved seismic model has several merits that enable it to better provide seismic analyses of structures.Specifically,at first,the ground motion model is compatible with the design response spectrum.Secondly,the evolutionary power spectrum involved in the model and the design response spectrum are constructed accordingly with sufficient consideration of the correlation between different seismic components.Thirdly,the random function-based dimension-reduction representation is applied,by which seismic modeling is established,with three elementary random variables.Numerical simulations of multi-dimensional stochastic ground motions in a specific design scenario indicate the effectiveness of the proposed modeling strategy.Moreover,the multi-dimensional seismic response and the global reliability of a high-rise frame-core tube structure is discussed in detail to further illustrate the engineering applicability of the proposed method.The analytical investigations demonstrate that the suggested stochastic model of multi-dimensional ground motion is available for accurate seismic response analysis and dynamic reliability assessment of complex engineering structures for performance-based seismic resistance design.
文摘Intelligence and perception are two operative technologies in 6G scenarios.The intelligent wireless network and information perception require a deep fusion of artificial intelligence(AI)and wireless communications in 6G systems.Therefore,fusion is becoming a typical feature and key challenge of 6G wireless communication systems.In this paper,we focus on the critical issues and propose three application scenarios in 6G wireless systems.Specifically,we first discuss the fusion of AI and 6G networks for the enhancement of 5G-advanced technology and future wireless communication systems.Then,we introduce the wireless AI technology architecture with 6G multidimensional information perception,which includes the physical layer technology of multi-dimensional feature information perception,full spectrum fusion technology,and intelligent wireless resource management.The discussion of key technologies for intelligent 6G wireless network networks is expected to provide a guideline for future research.
基金National Natural Science Foundation of China(No.70971133)
文摘By analyzing the shortage of reliability test design and thinking over the producer's risk and consumer's risk, the information fusion technology is used to set up a reliability test design model( RTDM). By analyzing the demands and constraint conditions of the RTDM and with applications of Bayesian approach and Monte Carlo method( MCM),this paper puts forward the exponential distributed subsystems and the information fusion technology among them. According to the posteriori risk criteria,formulas of producer's risk and consumer's risk were also inferred,and with the help of Matlab software,selection of the optimum test plan was solved. Finally,validity of the model had been proved by a test of series parallel system.
基金supported by the Project for Scientific Research of West Anhui University(No.00701092282)。
文摘An effective breeding blanket is critical to support tritium self-sufficiency for future fusion reactors.The difficulty is to achieve tritium breeding ratio(TBR)target of 1.05 or more.This paper presents a new design approach to the blanket design process.It indicates that fusion blanket design is affected by universal functions based on iterations.Three aspects are worth more attention from fusion engineers in the future.The first factor is that the iterations on the material fractions affect not only structure scheme but also TBR variation.The second factor is the cooling condition affecting final TBR due to the change of the structure material proportion.The third factor is temperature field related to the tritium release.In particular,it is suggested that the statistical calculation of effective TBR must be under reasonable control of the blanket temperature field.This approach is novel for blanket engineering in development of a fusion reactor.
文摘Although existing style transfer techniques have made significant progress in the field of image generation,there are still some challenges in the field of exhibition hall design.The existing style transfer methods mainly focus on the transformation of single dimensional features,but ignore the deep integration of content and style features in exhibition hall design.In addition,existing methods are deficient in detail retention,especially in accurately capturing and reproducing local textures and details while preserving the content image structure.In addition,point-based attention mechanisms tend to ignore the complexity and diversity of image features in multi-dimensional space,resulting in alignment problems between features in different semantic areas,resulting in inconsistent stylistic features in content areas.In this context,this paper proposes a semantic-enhanced multimodal style transfer algorithm tailored for exhibition hall design.The proposed approach leverages a multimodal encoder architecture to integrate information from text,source images,and style images,using separate encoder modules for each modality to capture shallow,deep,and semantic features.A novel Style Transfer Convolution(STConv)convolutional kernel,based on the Visual Geometry Group(VGG)19 network,is introduced to improve feature extraction in style transfer.Additionally,an enhanced Transformer encoder is incorporated to capture contextual semantic information within images,while the CLIP model is employed for text data processing.A hybrid attention module is designed to precisely capture style features,achieving multimodal feature fusion via a diffusion model that generates exhibition hall design images aligned with stylistic requirements.Quantitative experiments show that compared with the most advanced algorithms,the proposed method has achieved significant performance improvement on both Fréchet Inception Distance(FID)and Kernel Inception Distance(KID)indexes.For example,on the ExpoArchive dataset,the proposed method has a FID value of 87.9 and a KID value of 1.98,which is significantly superior to other methods.
基金supported by the National Natural Science Foundation of China under Grant 62107034the Major Science and Technology Project of Yunnan Province(202402AD080002)Yunnan International Joint R&D Center of China-Laos-Thailand Educational Digitalization(202203AP140006).
文摘The task of student action recognition in the classroom is to precisely capture and analyze the actions of students in classroom videos,providing a foundation for realizing intelligent and accurate teaching.However,the complex nature of the classroom environment has added challenges and difficulties in the process of student action recognition.In this research article,with regard to the circumstances where students are prone to be occluded and classroom computing resources are restricted in real classroom scenarios,a lightweight multi-modal fusion action recognition approach is put forward.This proposed method is capable of enhancing the accuracy of student action recognition while concurrently diminishing the number of parameters of the model and the Computation Amount,thereby achieving a more efficient and accurate recognition performance.In the feature extraction stage,this method fuses the keypoint heatmap with the RGB(Red-Green-Blue color model)image.In order to fully utilize the unique information of different modalities for feature complementarity,a Feature Fusion Module(FFE)is introduced.The FFE encodes and fuses the unique features of the two modalities during the feature extraction process.This fusion strategy not only achieves fusion and complementarity between modalities,but also improves the overall model performance.Furthermore,to reduce the computational load and parameter scale of the model,we use keypoint information to crop RGB images.At the same time,the first three networks of the lightweight feature extraction network X3D are used to extract dual-branch features.These methods significantly reduce the computational load and parameter scale.The number of parameters of the model is 1.40 million,and the computation amount is 5.04 billion floating-point operations per second(GFLOPs),achieving an efficient lightweight design.In the Student Classroom Action Dataset(SCAD),the accuracy of the model is 88.36%.In NTU 60(Nanyang Technological University Red-Green-Blue-Depth RGB+Ddataset with 60 categories),the accuracies on X-Sub(The people in the training set are different from those in the test set)and X-View(The perspectives of the training set and the test set are different)are 95.76%and 98.82%,respectively.On the NTU 120 dataset(Nanyang Technological University Red-Green-Blue-Depth dataset with 120 categories),RGB+Dthe accuracies on X-Sub and X-Set(the perspectives of the training set and the test set are different)are 91.97%and 93.45%,respectively.The model has achieved a balance in terms of accuracy,computation amount,and the number of parameters.
基金supported by the funding provided by Boeing Center for Aviation and Aerospace Safety.
文摘Powder bed fusion(PBF)in metallic additive manufacturing offers the ability to produce intricate geometries,high-strength components,and reliable products.However,powder processing before energy-based binding significantly impacts the final product’s integrity.Processing maps guide efficient process design to minimize defects,but creating them through experimentation alone is challenging due to the wide range of parameters,necessitating a comprehensive computational parametric analysis.In this study,we used the discrete element method to parametrically analyze the powder processing design space in PBF of stainless steel 316L powders.Uniform lattice parameter sweeps are often used for parametric analysis,but are computationally intensive.We find that non-uniform parameter sweep based on the low discrepancy sequence(LDS)algorithm is ten times more efficient at exploring the design space while accurately capturing the relationship between powder flow dynamics and bed packing density.We introduce a multi-layer perceptron(MLP)model to interpolate parametric causalities within the LDS parameter space.With over 99%accuracy,it effectively captures these causalities while requiring fewer simulations.Finally,we generate processing design maps for machine setups and powder selections for efficient process design.We find that recoating speed has the highest impact on powder processing quality,followed by recoating layer thickness,particle size,and inter-particle friction.
文摘This paper proposes a class of novel progressive edge growth-based codebooks for downlink sparse code multiple access(SCMA)systems.In the first scheme,we propose to progressively design the codebooks of each resource node(RN)instead of rotating a mother constellation(MC)as in the conventional SCMA works.In the other one,based on the MC,a multi-resources rotated codebooks are proposed to improve the performance of the superimposed constellations.The resultant codebooks are respectively referred to as the resource edge multidimensional codebooks(REMC)and the user edge multi-dimensional codebooks(UEMC).Additionally,we delve into the detailed design of the MC and the superimposed constellation.Then,we pay special attention to the application of the proposed schemes to challenging design cases,particularly for the high dimensional,high rate,and irregular codebooks,where the corresponding simplified schemes are proposed to reduce the complexity of codebook design.Finally,simulation results are presented to demonstrate the superiority of our progressive edge growth-based schemes.The numerical results indicate that the proposed codebooks significantly outperform the stateof-the-art codebooks.In addition,we also show that the proposed REMC codebooks outperform in the lower signal-to-noise ratio(SNR)regime,whereas the UEMC codebooks exhibit better performance at higher SNRs.
文摘目的/意义分析多模态融合技术在人工智能(artificial intelligence,AI)制药核心场景应用的关键价值,为制药领域创新发展提供理论与实践参考。方法/过程采用文献综述方法,在中国知网、PubMed和Web of Science检索相关文献,系统归纳多模态融合技术在AI制药领域的研究进展与应用实践。结果/结论 多模态融合技术在靶点识别、从头药物设计等核心场景的应用成效显著,但仍面临模态对齐精度不足、模型解释性较弱等挑战,未来将向高效模型架构与实际应用场景深度融合方向发展。
