A new analytical model for geometric size and forming force prediction in incremental flanging(IF)is presented in this work.The complex deformation characteristics of IF are considered in the modeling process,which ca...A new analytical model for geometric size and forming force prediction in incremental flanging(IF)is presented in this work.The complex deformation characteristics of IF are considered in the modeling process,which can accurately describe the strain and stress states in IF.Based on strain analysis,the model can predict the material thickness distribution and neck height after IF.By considering contact area,strain characteristics,material thickness changes,and friction,the model can predict specific moments and corresponding values of maximum axial forming force and maximum horizontal forming force during IF.In addition,an IF experiment involving different tool diameters,flanging diameters,and opening hole diameters is conducted.On the basis of the experimental strain paths,the strain characteristics of different deformation zones are studied,and the stable strain ratio is quantitatively described through two dimensionless parameters:relative tool diameter and relative hole diameter.Then,the changing of material thickness and forming force in IF,and the variation of minimum material thickness,neck height,maximum axial forming force,and maximum horizontal forming force with flanging parameters are studied,and the reliability of the analytical model is verified in this process.Finally,the influence of the horizontal forming force on the tool design and the fluctuation of the forming force are explained.展开更多
This paper presents the design of an asymmetrically variable wingtip anhedral angles morphing aircraft,inspired by biomimetic mechanisms,to enhance lateral maneuver capability.Firstly,we establish a lateral dynamic mo...This paper presents the design of an asymmetrically variable wingtip anhedral angles morphing aircraft,inspired by biomimetic mechanisms,to enhance lateral maneuver capability.Firstly,we establish a lateral dynamic model considering additional forces and moments resulting during the morphing process,and convert it into a Multiple Input Multiple Output(MIMO)virtual control system by importing virtual inputs.Secondly,a classical dynamics inversion controller is designed for the outer-loop system.A new Global Fast Terminal Incremental Sliding Mode Controller(NDO-GFTISMC)is proposed for the inner-loop system,in which an adaptive law is implemented to weaken control surface chattering,and a Nonlinear Disturbance Observer(NDO)is integrated to compensate for unknown disturbances.The whole control system is proven semiglobally uniformly ultimately bounded based on the multi-Lyapunov function method.Furthermore,we consider tracking errors and self-characteristics of actuators,a quadratic programmingbased dynamic control allocation law is designed,which allocates virtual control inputs to the asymmetrically deformed wingtip and rudder.Actuator dynamic models are incorporated to ensure physical realizability of designed allocation law.Finally,comparative experimental results validate the effectiveness of the designed control system and control allocation law.The NDO-GFTISMC features faster convergence,stronger robustness,and 81.25%and 75.0%reduction in maximum state tracking error under uncertainty compared to the Incremental Nonlinear Dynamic Inversion Controller based on NDO(NDO-INDI)and Incremental Sliding Mode Controller based on NDO(NDO-ISMC),respectively.The design of the morphing aircraft significantly enhances lateral maneuver capability,maintaining a substantial control margin during lateral maneuvering,reducing the burden of the rudder surface,and effectively solving the actuator saturation problem of traditional aircraft during lateral maneuvering.展开更多
The field of biometric identification has seen significant advancements over the years,with research focusing on enhancing the accuracy and security of these systems.One of the key developments is the integration of d...The field of biometric identification has seen significant advancements over the years,with research focusing on enhancing the accuracy and security of these systems.One of the key developments is the integration of deep learning techniques in biometric systems.However,despite these advancements,certain challenges persist.One of the most significant challenges is scalability over growing complexity.Traditional methods either require maintaining and securing a growing database,introducing serious security challenges,or relying on retraining the entiremodelwhen new data is introduced-a process that can be computationally expensive and complex.This challenge underscores the need for more efficient methods to scale securely.To this end,we introduce a novel approach that addresses these challenges by integrating multimodal biometrics,cancelable biometrics,and incremental learning techniques.This work is among the first attempts to seamlessly incorporate deep cancelable biometrics with dynamic architectural updates,applied incrementally to the deep learning model as new users are enrolled,achieving high performance with minimal catastrophic forgetting.By leveraging a One-Dimensional Convolutional Neural Network(1D-CNN)architecture combined with a hybrid incremental learning approach,our system achieves high recognition accuracy,averaging 98.98% over incrementing datasets,while ensuring user privacy through cancelable templates generated via a pre-trained CNN model and random projection.The approach demonstrates remarkable adaptability,utilizing the least intrusive biometric traits like facial features and fingerprints,ensuring not only robust performance but also long-term serviceability.展开更多
Semantic communication(SemCom)aims to achieve high-fidelity information delivery under low communication consumption by only guaranteeing semantic accuracy.Nevertheless,semantic communication still suffers from unexpe...Semantic communication(SemCom)aims to achieve high-fidelity information delivery under low communication consumption by only guaranteeing semantic accuracy.Nevertheless,semantic communication still suffers from unexpected channel volatility and thus developing a re-transmission mechanism(e.g.,hybrid automatic repeat request[HARQ])becomes indispensable.In that regard,instead of discarding previously transmitted information,the incremental knowledge-based HARQ(IK-HARQ)is deemed as a more effective mechanism that could sufficiently utilize the information semantics.However,considering the possible existence of semantic ambiguity in image transmission,a simple bit-level cyclic redundancy check(CRC)might compromise the performance of IK-HARQ.Therefore,there emerges a strong incentive to revolutionize the CRC mechanism,thus more effectively reaping the benefits of both SemCom and HARQ.In this paper,built on top of swin transformer-based joint source-channel coding(JSCC)and IK-HARQ,we propose a semantic image transmission framework SC-TDA-HARQ.In particular,different from the conventional CRC,we introduce a topological data analysis(TDA)-based error detection method,which capably digs out the inner topological and geometric information of images,to capture semantic information and determine the necessity for re-transmission.Extensive numerical results validate the effectiveness and efficiency of the proposed SC-TDA-HARQ framework,especially under the limited bandwidth condition,and manifest the superiority of TDA-based error detection method in image transmission.展开更多
The incremental capacity analysis(ICA)technique is notably limited by its sensitivity to variations in charging conditions,which constrains its practical applicability in real-world scenarios.This paper introduces an ...The incremental capacity analysis(ICA)technique is notably limited by its sensitivity to variations in charging conditions,which constrains its practical applicability in real-world scenarios.This paper introduces an ICA-compensation technique to address this limitation and propose a generalized framework for assessing the state of health(SOH)of batteries based on ICA that is applicable under differing charging conditions.This novel approach calculates the voltage profile under quasi-static conditions by subtracting the voltage increase attributable to the additional polarization effects at high currents from the measured voltage profile.This approach's efficacy is contingent upon precisely acquiring the equivalent impedance.To obtain the equivalent impedance throughout the batteries'lifespan while minimizing testing costs,this study employs a current interrupt technique in conjunction with a long short-term memory(LSTM)network to develop a predictive model for equivalent impedance.Following the derivation of ICA curves using voltage profiles under quasi-static conditions,the research explores two scenarios for SOH estimation:one utilizing only incremental capacity(IC)features and the other incorporating both IC features and IC sampling.A genetic algorithm-optimized backpropagation neural network(GABPNN)is employed for the SOH estimation.The proposed generalized framework is validated using independent training and test datasets.Variable test conditions are applied for the test set to rigorously evaluate the methodology under challenging conditions.These evaluation results demonstrate that the proposed framework achieves an estimation accuracy of 1.04%for RMSE and 0.90%for MAPE across a spectrum of charging rates ranging from 0.1 C to 1 C and starting SOCs between 0%and 70%,which constitutes a major advancement compared to established ICA methods.It also significantly enhances the applicability of conventional ICA techniques in varying charging conditions and negates the necessity for separate testing protocols for each charging scenario.展开更多
Ever since the research in machine learning gained traction in recent years,it has been employed to address challenges in a wide variety of domains,including mechanical devices.Most of the machine learning models are ...Ever since the research in machine learning gained traction in recent years,it has been employed to address challenges in a wide variety of domains,including mechanical devices.Most of the machine learning models are built on the assumption of a static learning environment,but in practical situations,the data generated by the process is dynamic.This evolution of the data is termed concept drift.This research paper presents an approach for predictingmechanical failure in real-time using incremental learning based on the statistically calculated parameters of mechanical equipment.The method proposed here is applicable to allmechanical devices that are susceptible to failure or operational degradation.The proposed method in this paper is equipped with the capacity to detect the drift in data generation and adaptation.The proposed approach evaluates the machine learning and deep learning models for their efficacy in handling the errors related to industrial machines due to their dynamic nature.It is observed that,in the settings without concept drift in the data,methods like SVM and Random Forest performed better compared to deep neural networks.However,this resulted in poor sensitivity for the smallest drift in the machine data reported as a drift.In this perspective,DNN generated the stable drift detection method;it reported an accuracy of 84%and an AUC of 0.87 while detecting only a single drift point,indicating the stability to performbetter in detecting and adapting to new data in the drifting environments under industrial measurement settings.展开更多
The influence of geometric configuration on the friction characteristics during incremental sheet forming of AA5052 was analyzed by integrating surface morphology and its characteristic parameters,along with plastic s...The influence of geometric configuration on the friction characteristics during incremental sheet forming of AA5052 was analyzed by integrating surface morphology and its characteristic parameters,along with plastic strain,contact pressure,and area.The interface promotes lubrication and support when wall angles were≤40°,a 0.5 mm-thin sheet was used,and a 10 mm-large tool radius was employed.This mainly results in micro-plowing and plastic extrusion flow,leading to lower friction coefficient.However,when wall angles exceed 40°,significant plastic strain roughening occurs,leading to inadequate lubrication on the newly formed surface.Increased sheet thickness and decreased tool radius elevate contact pressure.These actions trigger micro-cutting and adhesion,potentially leading to localized scuffing and dimple tears,and higher friction coefficient.The friction mechanisms remain unaffected by the part’s plane curve features.As the forming process progresses,abrasive wear intensifies,and surface morphology evolves unfavorably for lubrication and friction reduction.展开更多
In this paper,an incremental contact model is developed for the elastic self-affine fractal rough surfaces under plane strain condition.The contact between a rough surface and a rigid plane is simplified by the accumu...In this paper,an incremental contact model is developed for the elastic self-affine fractal rough surfaces under plane strain condition.The contact between a rough surface and a rigid plane is simplified by the accumulation of identical line contacts with half-width given by the truncated area divided by the contact patch number at varying heights.Based on the contact stiffness of two-dimensional flat punch,the total stiffness of rough surface is estimated,and then the normal load is calculated by an incremental method.For various rough surfaces,the approximately linear load-area relationships predicted by the proposed model agree well with the results of finite element simulations.It is found that the real average contact pressure depends significantly on profile properties.展开更多
To improve the prediction accuracy of chaotic time series and reconstruct a more reasonable phase space structure of the prediction network,we propose a convolutional neural network-long short-term memory(CNN-LSTM)pre...To improve the prediction accuracy of chaotic time series and reconstruct a more reasonable phase space structure of the prediction network,we propose a convolutional neural network-long short-term memory(CNN-LSTM)prediction model based on the incremental attention mechanism.Firstly,a traversal search is conducted through the traversal layer for finite parameters in the phase space.Then,an incremental attention layer is utilized for parameter judgment based on the dimension weight criteria(DWC).The phase space parameters that best meet DWC are selected and fed into the input layer.Finally,the constructed CNN-LSTM network extracts spatio-temporal features and provides the final prediction results.The model is verified using Logistic,Lorenz,and sunspot chaotic time series,and the performance is compared from the two dimensions of prediction accuracy and network phase space structure.Additionally,the CNN-LSTM network based on incremental attention is compared with long short-term memory(LSTM),convolutional neural network(CNN),recurrent neural network(RNN),and support vector regression(SVR)for prediction accuracy.The experiment results indicate that the proposed composite network model possesses enhanced capability in extracting temporal features and achieves higher prediction accuracy.Also,the algorithm to estimate the phase space parameter is compared with the traditional CAO,false nearest neighbor,and C-C,three typical methods for determining the chaotic phase space parameters.The experiments reveal that the phase space parameter estimation algorithm based on the incremental attention mechanism is superior in prediction accuracy compared with the traditional phase space reconstruction method in five networks,including CNN-LSTM,LSTM,CNN,RNN,and SVR.展开更多
Hyperspectral images typically have high spectral resolution but low spatial resolution,which impacts the reliability and accuracy of subsequent applications,for example,remote sensingclassification and mineral identi...Hyperspectral images typically have high spectral resolution but low spatial resolution,which impacts the reliability and accuracy of subsequent applications,for example,remote sensingclassification and mineral identification.But in traditional methods via deep convolution neural net-works,indiscriminately extracting and fusing spectral and spatial features makes it challenging toutilize the differentiated information across adjacent spectral channels.Thus,we proposed a multi-branch interleaved iterative upsampling hyperspectral image super-resolution reconstruction net-work(MIIUSR)to address the above problems.We reinforce spatial feature extraction by integrat-ing detailed features from different receptive fields across adjacent channels.Furthermore,we pro-pose an interleaved iterative upsampling process during the reconstruction stage,which progres-sively fuses incremental information among adjacent frequency bands.Additionally,we add twoparallel three dimensional(3D)feature extraction branches to the backbone network to extractspectral and spatial features of varying granularity.We further enhance the backbone network’sconstruction results by leveraging the difference between two dimensional(2D)channel-groupingspatial features and 3D multi-granularity features.The results obtained by applying the proposednetwork model to the CAVE test set show that,at a scaling factor of×4,the peak signal to noiseratio,spectral angle mapping,and structural similarity are 37.310 dB,3.525 and 0.9438,respec-tively.Besides,extensive experiments conducted on the Harvard and Foster datasets demonstratethe superior potential of the proposed model in hyperspectral super-resolution reconstruction.展开更多
Knowledge spillover via collaboration is essential to innovation,with proximity being a vital factor.Nevertheless,little consensus has been achieved on which form of proximity is more critical for innovation.Instead o...Knowledge spillover via collaboration is essential to innovation,with proximity being a vital factor.Nevertheless,little consensus has been achieved on which form of proximity is more critical for innovation.Instead of reaching a definitive conclusion,we highlight the potential of addressing the argument through the lens of innovation heterogeneity.This work thus contributes to current literature by integrating two forms of innovation,radical and incremental,into the discourse of geographical and organizational proximity in knowledge spillover via collaboration.Utilizing a dataset of patents from China’s listed firms between 2001 and 2017,we first categorize radical and incremental innovation according to the characteristics of knowledge combination,encompassing the familiarity of combined knowledge and maturity of combination ways.We further investigate the heterogenous effects of intra-region and intra-group knowledge spillovers,linked to geographical and organizational proximity in collaboration,on radical and incremental innovation.Empirical findings demonstrate that innovation relies on knowledge spillover both within groups and within regions.Moreover,intra-region spillover is essential for fostering radical innovation,while intra-group spillover only facilitates incremental innovation.Our findings provide both theoretical and practical implications,suggesting that multilocational enterprises should enhance their collaborator selection to leverage diverse knowledge spillovers,thereby fostering radical and incremental innovation in distinct ways.展开更多
Thin-walled metal parts with functional micro-featured surface have broad application prospects in the fields of resistance reduction,noise reduction,etc.In this study,a novel micro-rolling&incremental sheet formi...Thin-walled metal parts with functional micro-featured surface have broad application prospects in the fields of resistance reduction,noise reduction,etc.In this study,a novel micro-rolling&incremental sheet forming hybrid process(μR-ISF)is proposed to fabricate thin-walled metal parts with microgroove arrays.An analytical model which relates the rolling force and microgroove depth in the micro-rolling stage was first established.Then,the formation mechanism of microgroove morphology during both micro-rolling stage and macro-shape forming stage are investigated.After the micro-grooved sheet being incrementally formed,a significant reduction(between 21%to nearly 60%)is occurred in the depth of both transverse and longitudinal grooves compared to the flat sheet.Meanwhile,the width of transverse grooves decreases slightly by about 10%on average,while the width of longitudinal microgrooves increases significantly by more than 30%on average.After micro-rolling,85°{102}tensile twins appear on the micro-grooved sheet and the percentage of 65°{112}compressive twins increases.After incremental forming,the percentage of low-angle grain boundaries and the density of geometrically necessary dislocations in the formed part increase significantly,and the grain size distribution becomes more uniform.The present work provides a new strategy for the fabrication of 3D metal thin-walled components with surface micro-features.展开更多
The scale and complexity of big data are growing continuously,posing severe challenges to traditional data processing methods,especially in the field of clustering analysis.To address this issue,this paper introduces ...The scale and complexity of big data are growing continuously,posing severe challenges to traditional data processing methods,especially in the field of clustering analysis.To address this issue,this paper introduces a new method named Big Data Tensor Multi-Cluster Distributed Incremental Update(BDTMCDIncreUpdate),which combines distributed computing,storage technology,and incremental update techniques to provide an efficient and effective means for clustering analysis.Firstly,the original dataset is divided into multiple subblocks,and distributed computing resources are utilized to process the sub-blocks in parallel,enhancing efficiency.Then,initial clustering is performed on each sub-block using tensor-based multi-clustering techniques to obtain preliminary results.When new data arrives,incremental update technology is employed to update the core tensor and factor matrix,ensuring that the clustering model can adapt to changes in data.Finally,by combining the updated core tensor and factor matrix with historical computational results,refined clustering results are obtained,achieving real-time adaptation to dynamic data.Through experimental simulation on the Aminer dataset,the BDTMCDIncreUpdate method has demonstrated outstanding performance in terms of accuracy(ACC)and normalized mutual information(NMI)metrics,achieving an accuracy rate of 90%and an NMI score of 0.85,which outperforms existing methods such as TClusInitUpdate and TKLClusUpdate in most scenarios.Therefore,the BDTMCDIncreUpdate method offers an innovative solution to the field of big data analysis,integrating distributed computing,incremental updates,and tensor-based multi-clustering techniques.It not only improves the efficiency and scalability in processing large-scale high-dimensional datasets but also has been validated for its effectiveness and accuracy through experiments.This method shows great potential in real-world applications where dynamic data growth is common,and it is of significant importance for advancing the development of data analysis technology.展开更多
We investigated the parametric optimization on incremental sheet forming of stainless steel using Grey Relational Analysis(GRA) coupled with Principal Component Analysis(PCA). AISI 316L stainless steel sheets were use...We investigated the parametric optimization on incremental sheet forming of stainless steel using Grey Relational Analysis(GRA) coupled with Principal Component Analysis(PCA). AISI 316L stainless steel sheets were used to develop double wall angle pyramid with aid of tungsten carbide tool. GRA coupled with PCA was used to plan the experiment conditions. Control factors such as Tool Diameter(TD), Step Depth(SD), Bottom Wall Angle(BWA), Feed Rate(FR) and Spindle Speed(SS) on Top Wall Angle(TWA) and Top Wall Angle Surface Roughness(TWASR) have been studied. Wall angle increases with increasing tool diameter due to large contact area between tool and workpiece. As the step depth, feed rate and spindle speed increase,TWASR decreases with increasing tool diameter. As the step depth increasing, the hydrostatic stress is raised causing severe cracks in the deformed surface. Hence it was concluded that the proposed hybrid method was suitable for optimizing the factors and response.展开更多
The continuous emergence of new targets in open scenarios leads to a substantial decrease in the performance of Inverse Synthetic Aperture Radar(ISAR)recognition systems.Also,data scarcity further exacerbates the chal...The continuous emergence of new targets in open scenarios leads to a substantial decrease in the performance of Inverse Synthetic Aperture Radar(ISAR)recognition systems.Also,data scarcity further exacerbates the challenge of identifying new classes of ISAR targets.In this paper,a few-shot incremental target recognition framework based on Scattering-Topology Properties(STPIL)is proposed.Specifically,STPIL extracts scattering-topology properties of ISAR targets as recognition features.Meanwhile,the pseudo-incremental training strategy effectively alleviates the algorithm’s forgetting of old knowledge,and improves compatibility with new classes.Besides,a feature embedding network,with few parameters,is designed based on the graph neural network.This embedding network is highly adaptable to changes in data distribution.Additionally,STPIL fully considers the joint distribution and marginal distribution in scattering features,and uses the Brownian distance metric module to make the scattering-topology features more discriminative.Experimental results on both the simulation dataset and the public measured data indicate that STPIL can effectively balance new classes with old classes,and has superior performance to other advanced methods in the incremental recognition of targets.展开更多
Time-Sensitive Network(TSN)with deterministic transmission capability is increasingly used in many emerging fields.It mainly guarantees the Quality of Service(QoS)of applications with strict requirements on time and s...Time-Sensitive Network(TSN)with deterministic transmission capability is increasingly used in many emerging fields.It mainly guarantees the Quality of Service(QoS)of applications with strict requirements on time and security.One of the core features of TSN is traffic scheduling with bounded low delay in the network.However,traffic scheduling schemes in TSN are usually synthesized offline and lack dynamism.To implement incremental scheduling of newly arrived traffic in TSN,we propose a Dynamic Response Incremental Scheduling(DR-IS)method for time-sensitive traffic and deploy it on a software-defined time-sensitive network architecture.Under the premise of meeting the traffic scheduling requirements,we adopt two modes,traffic shift and traffic exchange,to dynamically adjust the time slot injection position of the traffic in the original scheme,and determine the sending offset time of the new timesensitive traffic to minimize the global traffic transmission jitter.The evaluation results show that DRIS method can effectively control the large increase of traffic transmission jitter in incremental scheduling without affecting the transmission delay,thus realizing the dynamic incremental scheduling of time-sensitive traffic in TSN.展开更多
The objective of this study is to construct a multi-department symptom-based automatic diagnosis model.However,it is dificult to establish a model to classify plenty of diseases and collect thousands of disease-sympto...The objective of this study is to construct a multi-department symptom-based automatic diagnosis model.However,it is dificult to establish a model to classify plenty of diseases and collect thousands of disease-symptom datasets simultaneously.Inspired by the thought of"knowledge graph is model",this study proposes to build an experience-infused knowledge model by continuously learning the experiential knowledge from data,and incrementally injecting it into the knowledge graph.Therefore,incremental learning and injection are used to solve the data collection problem,and the knowledge graph is modeled and containerized to solve the large-scale multi-classification problems.First,an entity linking method is designed and a heterogeneous knowledge graph is constructed by graph fusion.Then,an adaptive neural network model is constructed for each dataset,and the data is used for statistical initialization and model training.Finally,the weights and biases of the learned neural network model are updated to the knowledge graph.It is worth noting that for the incremental process,we consider both the data and class increments.We evaluate the diagnostic effectiveness of the model on the current dataset and the anti-forgetting ability on the historical dataset after class increment on three public datasets.Compared with the classical model,the proposed model improves the diagnostic accuracy of the three datasets by 5%,2%,and 15%on average,respectively.Meanwhile,the model under incremental learning has a better ability to resist forgetting.展开更多
Currently,distributed routing protocols are constrained by offering a single path between any pair of nodes,thereby limiting the potential throughput and overall network performance.This approach not only restricts th...Currently,distributed routing protocols are constrained by offering a single path between any pair of nodes,thereby limiting the potential throughput and overall network performance.This approach not only restricts the flow of data but also makes the network susceptible to failures in case the primary path is disrupted.In contrast,routing protocols that leverage multiple paths within the network offer a more resilient and efficient solution.Multipath routing,as a fundamental concept,surpasses the limitations of traditional shortest path first protocols.It not only redirects traffic to unused resources,effectively mitigating network congestion,but also ensures load balancing across the network.This optimization significantly improves network utilization and boosts the overall performance,making it a widely recognized efficient method for enhancing network reliability.To further strengthen network resilience against failures,we introduce a routing scheme known as Multiple Nodes with at least Two Choices(MNTC).This innovative approach aims to significantly enhance network availability by providing each node with at least two routing choices.By doing so,it not only reduces the dependency on a single path but also creates redundant paths that can be utilized in case of failures,thereby enhancing the overall resilience of the network.To ensure the optimal placement of nodes,we propose three incremental deployment algorithms.These algorithms carefully select the most suitable set of nodes for deployment,taking into account various factors such as node connectivity,traffic patterns,and network topology.By deployingMNTCon a carefully chosen set of nodes,we can significantly enhance network reliability without the need for a complete overhaul of the existing infrastructure.We have conducted extensive evaluations of MNTC in diverse topological spaces,demonstrating its effectiveness in maintaining high network availability with minimal path stretch.The results are impressive,showing that even when implemented on just 60%of nodes,our incremental deployment method significantly boosts network availability.This underscores the potential of MNTC in enhancing network resilience and performance,making it a viable solution for modern networks facing increasing demands and complexities.The algorithms OSPF,TBFH,DC and LFC perform fast rerouting based on strict conditions,while MNTC is not restricted by these conditions.In five real network topologies,the average network availability ofMNTCis improved by 14.68%,6.28%,4.76%and 2.84%,respectively,compared with OSPF,TBFH,DC and LFC.展开更多
The visions of Industry 4.0 and 5.0 have reinforced the industrial environment.They have also made artificial intelligence incorporated as a major facilitator.Diagnosing machine faults has become a solid foundation fo...The visions of Industry 4.0 and 5.0 have reinforced the industrial environment.They have also made artificial intelligence incorporated as a major facilitator.Diagnosing machine faults has become a solid foundation for automatically recognizing machine failure,and thus timely maintenance can ensure safe operations.Transfer learning is a promising solution that can enhance the machine fault diagnosis model by borrowing pre-trained knowledge from the source model and applying it to the target model,which typically involves two datasets.In response to the availability of multiple datasets,this paper proposes using selective and adaptive incremental transfer learning(SA-ITL),which fuses three algorithms,namely,the hybrid selective algorithm,the transferability enhancement algorithm,and the incremental transfer learning algorithm.It is a selective algorithm that enables selecting and ordering appropriate datasets for transfer learning and selecting useful knowledge to avoid negative transfer.The algorithm also adaptively adjusts the portion of training data to balance the learning rate and training time.The proposed algorithm is evaluated and analyzed using ten benchmark datasets.Compared with other algorithms from existing works,SA-ITL improves the accuracy of all datasets.Ablation studies present the accuracy enhancements of the SA-ITL,including the hybrid selective algorithm(1.22%-3.82%),transferability enhancement algorithm(1.91%-4.15%),and incremental transfer learning algorithm(0.605%-2.68%).These also show the benefits of enhancing the target model with heterogeneous image datasets that widen the range of domain selection between source and target domains.展开更多
Humans are experiencing the inclusion of artificial agents in their lives,such as unmanned vehicles,service robots,voice assistants,and intelligent medical care.If the artificial agents cannot align with social values...Humans are experiencing the inclusion of artificial agents in their lives,such as unmanned vehicles,service robots,voice assistants,and intelligent medical care.If the artificial agents cannot align with social values or make ethical decisions,they may not meet the expectations of humans.Traditionally,an ethical decision-making framework is constructed by rule-based or statistical approaches.In this paper,we propose an ethical decision-making framework based on incremental ILP(Inductive Logic Programming),which can overcome the brittleness of rule-based approaches and little interpretability of statistical approaches.As the current incremental ILP makes it difficult to solve conflicts,we propose a novel ethical decision-making framework considering conflicts in this paper,which adopts our proposed incremental ILP system.The framework consists of two processes:the learning process and the deduction process.The first process records bottom clauses with their score functions and learns rules guided by the entailment and the score function.The second process obtains an ethical decision based on the rules.In an ethical scenario about chatbots for teenagers’mental health,we verify that our framework can learn ethical rules and make ethical decisions.Besides,we extract incremental ILP from the framework and compare it with the state-of-the-art ILP systems based on ASP(Answer Set Programming)focusing on conflict resolution.The results of comparisons show that our proposed system can generate better-quality rules than most other systems.展开更多
基金supported in part by financial support from the National Key R&D Program of China(No.2023YFB3407003)the National Natural Science Foundation of China(No.52375378).
文摘A new analytical model for geometric size and forming force prediction in incremental flanging(IF)is presented in this work.The complex deformation characteristics of IF are considered in the modeling process,which can accurately describe the strain and stress states in IF.Based on strain analysis,the model can predict the material thickness distribution and neck height after IF.By considering contact area,strain characteristics,material thickness changes,and friction,the model can predict specific moments and corresponding values of maximum axial forming force and maximum horizontal forming force during IF.In addition,an IF experiment involving different tool diameters,flanging diameters,and opening hole diameters is conducted.On the basis of the experimental strain paths,the strain characteristics of different deformation zones are studied,and the stable strain ratio is quantitatively described through two dimensionless parameters:relative tool diameter and relative hole diameter.Then,the changing of material thickness and forming force in IF,and the variation of minimum material thickness,neck height,maximum axial forming force,and maximum horizontal forming force with flanging parameters are studied,and the reliability of the analytical model is verified in this process.Finally,the influence of the horizontal forming force on the tool design and the fluctuation of the forming force are explained.
基金supported by the National Natural Science Foundation of China(Nos.62103052 and No.52175214)。
文摘This paper presents the design of an asymmetrically variable wingtip anhedral angles morphing aircraft,inspired by biomimetic mechanisms,to enhance lateral maneuver capability.Firstly,we establish a lateral dynamic model considering additional forces and moments resulting during the morphing process,and convert it into a Multiple Input Multiple Output(MIMO)virtual control system by importing virtual inputs.Secondly,a classical dynamics inversion controller is designed for the outer-loop system.A new Global Fast Terminal Incremental Sliding Mode Controller(NDO-GFTISMC)is proposed for the inner-loop system,in which an adaptive law is implemented to weaken control surface chattering,and a Nonlinear Disturbance Observer(NDO)is integrated to compensate for unknown disturbances.The whole control system is proven semiglobally uniformly ultimately bounded based on the multi-Lyapunov function method.Furthermore,we consider tracking errors and self-characteristics of actuators,a quadratic programmingbased dynamic control allocation law is designed,which allocates virtual control inputs to the asymmetrically deformed wingtip and rudder.Actuator dynamic models are incorporated to ensure physical realizability of designed allocation law.Finally,comparative experimental results validate the effectiveness of the designed control system and control allocation law.The NDO-GFTISMC features faster convergence,stronger robustness,and 81.25%and 75.0%reduction in maximum state tracking error under uncertainty compared to the Incremental Nonlinear Dynamic Inversion Controller based on NDO(NDO-INDI)and Incremental Sliding Mode Controller based on NDO(NDO-ISMC),respectively.The design of the morphing aircraft significantly enhances lateral maneuver capability,maintaining a substantial control margin during lateral maneuvering,reducing the burden of the rudder surface,and effectively solving the actuator saturation problem of traditional aircraft during lateral maneuvering.
基金the Deputyship for Research&Innovation,Ministry of Education in Saudi Arabia for funding this research work through project number RI-44-0833.
文摘The field of biometric identification has seen significant advancements over the years,with research focusing on enhancing the accuracy and security of these systems.One of the key developments is the integration of deep learning techniques in biometric systems.However,despite these advancements,certain challenges persist.One of the most significant challenges is scalability over growing complexity.Traditional methods either require maintaining and securing a growing database,introducing serious security challenges,or relying on retraining the entiremodelwhen new data is introduced-a process that can be computationally expensive and complex.This challenge underscores the need for more efficient methods to scale securely.To this end,we introduce a novel approach that addresses these challenges by integrating multimodal biometrics,cancelable biometrics,and incremental learning techniques.This work is among the first attempts to seamlessly incorporate deep cancelable biometrics with dynamic architectural updates,applied incrementally to the deep learning model as new users are enrolled,achieving high performance with minimal catastrophic forgetting.By leveraging a One-Dimensional Convolutional Neural Network(1D-CNN)architecture combined with a hybrid incremental learning approach,our system achieves high recognition accuracy,averaging 98.98% over incrementing datasets,while ensuring user privacy through cancelable templates generated via a pre-trained CNN model and random projection.The approach demonstrates remarkable adaptability,utilizing the least intrusive biometric traits like facial features and fingerprints,ensuring not only robust performance but also long-term serviceability.
基金supported in part by the National Key Research and Development Program of China under Grant 2024YFE0200600in part by the National Natural Science Foundation of China under Grant 62071425+3 种基金in part by the Zhejiang Key Research and Development Plan under Grant 2022C01093in part by the Zhejiang Provincial Natural Science Foundation of China under Grant LR23F010005in part by the National Key Laboratory of Wireless Communications Foundation under Grant 2023KP01601in part by the Big Data and Intelligent Computing Key Lab of CQUPT under Grant BDIC-2023-B-001.
文摘Semantic communication(SemCom)aims to achieve high-fidelity information delivery under low communication consumption by only guaranteeing semantic accuracy.Nevertheless,semantic communication still suffers from unexpected channel volatility and thus developing a re-transmission mechanism(e.g.,hybrid automatic repeat request[HARQ])becomes indispensable.In that regard,instead of discarding previously transmitted information,the incremental knowledge-based HARQ(IK-HARQ)is deemed as a more effective mechanism that could sufficiently utilize the information semantics.However,considering the possible existence of semantic ambiguity in image transmission,a simple bit-level cyclic redundancy check(CRC)might compromise the performance of IK-HARQ.Therefore,there emerges a strong incentive to revolutionize the CRC mechanism,thus more effectively reaping the benefits of both SemCom and HARQ.In this paper,built on top of swin transformer-based joint source-channel coding(JSCC)and IK-HARQ,we propose a semantic image transmission framework SC-TDA-HARQ.In particular,different from the conventional CRC,we introduce a topological data analysis(TDA)-based error detection method,which capably digs out the inner topological and geometric information of images,to capture semantic information and determine the necessity for re-transmission.Extensive numerical results validate the effectiveness and efficiency of the proposed SC-TDA-HARQ framework,especially under the limited bandwidth condition,and manifest the superiority of TDA-based error detection method in image transmission.
基金funded by the Bavarian State Ministry of ScienceResearch and Art(Grant number:H.2-F1116.WE/52/2)。
文摘The incremental capacity analysis(ICA)technique is notably limited by its sensitivity to variations in charging conditions,which constrains its practical applicability in real-world scenarios.This paper introduces an ICA-compensation technique to address this limitation and propose a generalized framework for assessing the state of health(SOH)of batteries based on ICA that is applicable under differing charging conditions.This novel approach calculates the voltage profile under quasi-static conditions by subtracting the voltage increase attributable to the additional polarization effects at high currents from the measured voltage profile.This approach's efficacy is contingent upon precisely acquiring the equivalent impedance.To obtain the equivalent impedance throughout the batteries'lifespan while minimizing testing costs,this study employs a current interrupt technique in conjunction with a long short-term memory(LSTM)network to develop a predictive model for equivalent impedance.Following the derivation of ICA curves using voltage profiles under quasi-static conditions,the research explores two scenarios for SOH estimation:one utilizing only incremental capacity(IC)features and the other incorporating both IC features and IC sampling.A genetic algorithm-optimized backpropagation neural network(GABPNN)is employed for the SOH estimation.The proposed generalized framework is validated using independent training and test datasets.Variable test conditions are applied for the test set to rigorously evaluate the methodology under challenging conditions.These evaluation results demonstrate that the proposed framework achieves an estimation accuracy of 1.04%for RMSE and 0.90%for MAPE across a spectrum of charging rates ranging from 0.1 C to 1 C and starting SOCs between 0%and 70%,which constitutes a major advancement compared to established ICA methods.It also significantly enhances the applicability of conventional ICA techniques in varying charging conditions and negates the necessity for separate testing protocols for each charging scenario.
文摘Ever since the research in machine learning gained traction in recent years,it has been employed to address challenges in a wide variety of domains,including mechanical devices.Most of the machine learning models are built on the assumption of a static learning environment,but in practical situations,the data generated by the process is dynamic.This evolution of the data is termed concept drift.This research paper presents an approach for predictingmechanical failure in real-time using incremental learning based on the statistically calculated parameters of mechanical equipment.The method proposed here is applicable to allmechanical devices that are susceptible to failure or operational degradation.The proposed method in this paper is equipped with the capacity to detect the drift in data generation and adaptation.The proposed approach evaluates the machine learning and deep learning models for their efficacy in handling the errors related to industrial machines due to their dynamic nature.It is observed that,in the settings without concept drift in the data,methods like SVM and Random Forest performed better compared to deep neural networks.However,this resulted in poor sensitivity for the smallest drift in the machine data reported as a drift.In this perspective,DNN generated the stable drift detection method;it reported an accuracy of 84%and an AUC of 0.87 while detecting only a single drift point,indicating the stability to performbetter in detecting and adapting to new data in the drifting environments under industrial measurement settings.
基金the support of the Key Research and Development Program of Shaanxi Province,China(No.2021GXLH-Z-049)。
文摘The influence of geometric configuration on the friction characteristics during incremental sheet forming of AA5052 was analyzed by integrating surface morphology and its characteristic parameters,along with plastic strain,contact pressure,and area.The interface promotes lubrication and support when wall angles were≤40°,a 0.5 mm-thin sheet was used,and a 10 mm-large tool radius was employed.This mainly results in micro-plowing and plastic extrusion flow,leading to lower friction coefficient.However,when wall angles exceed 40°,significant plastic strain roughening occurs,leading to inadequate lubrication on the newly formed surface.Increased sheet thickness and decreased tool radius elevate contact pressure.These actions trigger micro-cutting and adhesion,potentially leading to localized scuffing and dimple tears,and higher friction coefficient.The friction mechanisms remain unaffected by the part’s plane curve features.As the forming process progresses,abrasive wear intensifies,and surface morphology evolves unfavorably for lubrication and friction reduction.
基金supported by the National Natural Science Foundation of China(Grant Nos.12372100,12302126,and 12302141)the China Postdoctoral Science Foundation(Grant No.2023M732799)+1 种基金the Fundamental Research Funds for the Central Universities(Grant No.xzy012024020)Sihe Wang also thanks the support from the China Scholarship Council(CSC).
文摘In this paper,an incremental contact model is developed for the elastic self-affine fractal rough surfaces under plane strain condition.The contact between a rough surface and a rigid plane is simplified by the accumulation of identical line contacts with half-width given by the truncated area divided by the contact patch number at varying heights.Based on the contact stiffness of two-dimensional flat punch,the total stiffness of rough surface is estimated,and then the normal load is calculated by an incremental method.For various rough surfaces,the approximately linear load-area relationships predicted by the proposed model agree well with the results of finite element simulations.It is found that the real average contact pressure depends significantly on profile properties.
文摘To improve the prediction accuracy of chaotic time series and reconstruct a more reasonable phase space structure of the prediction network,we propose a convolutional neural network-long short-term memory(CNN-LSTM)prediction model based on the incremental attention mechanism.Firstly,a traversal search is conducted through the traversal layer for finite parameters in the phase space.Then,an incremental attention layer is utilized for parameter judgment based on the dimension weight criteria(DWC).The phase space parameters that best meet DWC are selected and fed into the input layer.Finally,the constructed CNN-LSTM network extracts spatio-temporal features and provides the final prediction results.The model is verified using Logistic,Lorenz,and sunspot chaotic time series,and the performance is compared from the two dimensions of prediction accuracy and network phase space structure.Additionally,the CNN-LSTM network based on incremental attention is compared with long short-term memory(LSTM),convolutional neural network(CNN),recurrent neural network(RNN),and support vector regression(SVR)for prediction accuracy.The experiment results indicate that the proposed composite network model possesses enhanced capability in extracting temporal features and achieves higher prediction accuracy.Also,the algorithm to estimate the phase space parameter is compared with the traditional CAO,false nearest neighbor,and C-C,three typical methods for determining the chaotic phase space parameters.The experiments reveal that the phase space parameter estimation algorithm based on the incremental attention mechanism is superior in prediction accuracy compared with the traditional phase space reconstruction method in five networks,including CNN-LSTM,LSTM,CNN,RNN,and SVR.
基金the National Natural Science Foun-dation of China(Nos.61471263,61872267 and U21B2024)the Natural Science Foundation of Tianjin,China(No.16JCZDJC31100)Tianjin University Innovation Foundation(No.2021XZC0024).
文摘Hyperspectral images typically have high spectral resolution but low spatial resolution,which impacts the reliability and accuracy of subsequent applications,for example,remote sensingclassification and mineral identification.But in traditional methods via deep convolution neural net-works,indiscriminately extracting and fusing spectral and spatial features makes it challenging toutilize the differentiated information across adjacent spectral channels.Thus,we proposed a multi-branch interleaved iterative upsampling hyperspectral image super-resolution reconstruction net-work(MIIUSR)to address the above problems.We reinforce spatial feature extraction by integrat-ing detailed features from different receptive fields across adjacent channels.Furthermore,we pro-pose an interleaved iterative upsampling process during the reconstruction stage,which progres-sively fuses incremental information among adjacent frequency bands.Additionally,we add twoparallel three dimensional(3D)feature extraction branches to the backbone network to extractspectral and spatial features of varying granularity.We further enhance the backbone network’sconstruction results by leveraging the difference between two dimensional(2D)channel-groupingspatial features and 3D multi-granularity features.The results obtained by applying the proposednetwork model to the CAVE test set show that,at a scaling factor of×4,the peak signal to noiseratio,spectral angle mapping,and structural similarity are 37.310 dB,3.525 and 0.9438,respec-tively.Besides,extensive experiments conducted on the Harvard and Foster datasets demonstratethe superior potential of the proposed model in hyperspectral super-resolution reconstruction.
基金National Natural Science Foundation of China,No.42122006,No.42471187。
文摘Knowledge spillover via collaboration is essential to innovation,with proximity being a vital factor.Nevertheless,little consensus has been achieved on which form of proximity is more critical for innovation.Instead of reaching a definitive conclusion,we highlight the potential of addressing the argument through the lens of innovation heterogeneity.This work thus contributes to current literature by integrating two forms of innovation,radical and incremental,into the discourse of geographical and organizational proximity in knowledge spillover via collaboration.Utilizing a dataset of patents from China’s listed firms between 2001 and 2017,we first categorize radical and incremental innovation according to the characteristics of knowledge combination,encompassing the familiarity of combined knowledge and maturity of combination ways.We further investigate the heterogenous effects of intra-region and intra-group knowledge spillovers,linked to geographical and organizational proximity in collaboration,on radical and incremental innovation.Empirical findings demonstrate that innovation relies on knowledge spillover both within groups and within regions.Moreover,intra-region spillover is essential for fostering radical innovation,while intra-group spillover only facilitates incremental innovation.Our findings provide both theoretical and practical implications,suggesting that multilocational enterprises should enhance their collaborator selection to leverage diverse knowledge spillovers,thereby fostering radical and incremental innovation in distinct ways.
基金This work is supported by the National Natural Science Foundation of China(Nos.51975328,52275348)Taishan Scholar Project of Shandong Province(No.tsqn202306006)Youth Innovation Technology Support Program of Shandong Provincial Universities(No.2022KJ041).
文摘Thin-walled metal parts with functional micro-featured surface have broad application prospects in the fields of resistance reduction,noise reduction,etc.In this study,a novel micro-rolling&incremental sheet forming hybrid process(μR-ISF)is proposed to fabricate thin-walled metal parts with microgroove arrays.An analytical model which relates the rolling force and microgroove depth in the micro-rolling stage was first established.Then,the formation mechanism of microgroove morphology during both micro-rolling stage and macro-shape forming stage are investigated.After the micro-grooved sheet being incrementally formed,a significant reduction(between 21%to nearly 60%)is occurred in the depth of both transverse and longitudinal grooves compared to the flat sheet.Meanwhile,the width of transverse grooves decreases slightly by about 10%on average,while the width of longitudinal microgrooves increases significantly by more than 30%on average.After micro-rolling,85°{102}tensile twins appear on the micro-grooved sheet and the percentage of 65°{112}compressive twins increases.After incremental forming,the percentage of low-angle grain boundaries and the density of geometrically necessary dislocations in the formed part increase significantly,and the grain size distribution becomes more uniform.The present work provides a new strategy for the fabrication of 3D metal thin-walled components with surface micro-features.
基金sponsored by the National Natural Science Foundation of China(Nos.61972208,62102194 and 62102196)National Natural Science Foundation of China(Youth Project)(No.62302237)+3 种基金Six Talent Peaks Project of Jiangsu Province(No.RJFW-111),China Postdoctoral Science Foundation Project(No.2018M640509)Postgraduate Research and Practice Innovation Program of Jiangsu Province(Nos.KYCX22_1019,KYCX23_1087,KYCX22_1027,KYCX23_1087,SJCX24_0339 and SJCX24_0346)Innovative Training Program for College Students of Nanjing University of Posts and Telecommunications(No.XZD2019116)Nanjing University of Posts and Telecommunications College Students Innovation Training Program(Nos.XZD2019116,XYB2019331).
文摘The scale and complexity of big data are growing continuously,posing severe challenges to traditional data processing methods,especially in the field of clustering analysis.To address this issue,this paper introduces a new method named Big Data Tensor Multi-Cluster Distributed Incremental Update(BDTMCDIncreUpdate),which combines distributed computing,storage technology,and incremental update techniques to provide an efficient and effective means for clustering analysis.Firstly,the original dataset is divided into multiple subblocks,and distributed computing resources are utilized to process the sub-blocks in parallel,enhancing efficiency.Then,initial clustering is performed on each sub-block using tensor-based multi-clustering techniques to obtain preliminary results.When new data arrives,incremental update technology is employed to update the core tensor and factor matrix,ensuring that the clustering model can adapt to changes in data.Finally,by combining the updated core tensor and factor matrix with historical computational results,refined clustering results are obtained,achieving real-time adaptation to dynamic data.Through experimental simulation on the Aminer dataset,the BDTMCDIncreUpdate method has demonstrated outstanding performance in terms of accuracy(ACC)and normalized mutual information(NMI)metrics,achieving an accuracy rate of 90%and an NMI score of 0.85,which outperforms existing methods such as TClusInitUpdate and TKLClusUpdate in most scenarios.Therefore,the BDTMCDIncreUpdate method offers an innovative solution to the field of big data analysis,integrating distributed computing,incremental updates,and tensor-based multi-clustering techniques.It not only improves the efficiency and scalability in processing large-scale high-dimensional datasets but also has been validated for its effectiveness and accuracy through experiments.This method shows great potential in real-world applications where dynamic data growth is common,and it is of significant importance for advancing the development of data analysis technology.
文摘We investigated the parametric optimization on incremental sheet forming of stainless steel using Grey Relational Analysis(GRA) coupled with Principal Component Analysis(PCA). AISI 316L stainless steel sheets were used to develop double wall angle pyramid with aid of tungsten carbide tool. GRA coupled with PCA was used to plan the experiment conditions. Control factors such as Tool Diameter(TD), Step Depth(SD), Bottom Wall Angle(BWA), Feed Rate(FR) and Spindle Speed(SS) on Top Wall Angle(TWA) and Top Wall Angle Surface Roughness(TWASR) have been studied. Wall angle increases with increasing tool diameter due to large contact area between tool and workpiece. As the step depth, feed rate and spindle speed increase,TWASR decreases with increasing tool diameter. As the step depth increasing, the hydrostatic stress is raised causing severe cracks in the deformed surface. Hence it was concluded that the proposed hybrid method was suitable for optimizing the factors and response.
文摘The continuous emergence of new targets in open scenarios leads to a substantial decrease in the performance of Inverse Synthetic Aperture Radar(ISAR)recognition systems.Also,data scarcity further exacerbates the challenge of identifying new classes of ISAR targets.In this paper,a few-shot incremental target recognition framework based on Scattering-Topology Properties(STPIL)is proposed.Specifically,STPIL extracts scattering-topology properties of ISAR targets as recognition features.Meanwhile,the pseudo-incremental training strategy effectively alleviates the algorithm’s forgetting of old knowledge,and improves compatibility with new classes.Besides,a feature embedding network,with few parameters,is designed based on the graph neural network.This embedding network is highly adaptable to changes in data distribution.Additionally,STPIL fully considers the joint distribution and marginal distribution in scattering features,and uses the Brownian distance metric module to make the scattering-topology features more discriminative.Experimental results on both the simulation dataset and the public measured data indicate that STPIL can effectively balance new classes with old classes,and has superior performance to other advanced methods in the incremental recognition of targets.
基金supported by the Innovation Scientists and Technicians Troop Construction Projects of Henan Province(224000510002)。
文摘Time-Sensitive Network(TSN)with deterministic transmission capability is increasingly used in many emerging fields.It mainly guarantees the Quality of Service(QoS)of applications with strict requirements on time and security.One of the core features of TSN is traffic scheduling with bounded low delay in the network.However,traffic scheduling schemes in TSN are usually synthesized offline and lack dynamism.To implement incremental scheduling of newly arrived traffic in TSN,we propose a Dynamic Response Incremental Scheduling(DR-IS)method for time-sensitive traffic and deploy it on a software-defined time-sensitive network architecture.Under the premise of meeting the traffic scheduling requirements,we adopt two modes,traffic shift and traffic exchange,to dynamically adjust the time slot injection position of the traffic in the original scheme,and determine the sending offset time of the new timesensitive traffic to minimize the global traffic transmission jitter.The evaluation results show that DRIS method can effectively control the large increase of traffic transmission jitter in incremental scheduling without affecting the transmission delay,thus realizing the dynamic incremental scheduling of time-sensitive traffic in TSN.
基金the National Key Research and Development Program(No.2018YFB1307005)the Smart Medical Project of Shanghai Municipal Commission of Health and Family Planning(No.2018ZHYL0226)。
文摘The objective of this study is to construct a multi-department symptom-based automatic diagnosis model.However,it is dificult to establish a model to classify plenty of diseases and collect thousands of disease-symptom datasets simultaneously.Inspired by the thought of"knowledge graph is model",this study proposes to build an experience-infused knowledge model by continuously learning the experiential knowledge from data,and incrementally injecting it into the knowledge graph.Therefore,incremental learning and injection are used to solve the data collection problem,and the knowledge graph is modeled and containerized to solve the large-scale multi-classification problems.First,an entity linking method is designed and a heterogeneous knowledge graph is constructed by graph fusion.Then,an adaptive neural network model is constructed for each dataset,and the data is used for statistical initialization and model training.Finally,the weights and biases of the learned neural network model are updated to the knowledge graph.It is worth noting that for the incremental process,we consider both the data and class increments.We evaluate the diagnostic effectiveness of the model on the current dataset and the anti-forgetting ability on the historical dataset after class increment on three public datasets.Compared with the classical model,the proposed model improves the diagnostic accuracy of the three datasets by 5%,2%,and 15%on average,respectively.Meanwhile,the model under incremental learning has a better ability to resist forgetting.
基金supported by Fundamental Research Program of Shanxi Province(No.20210302123444)the Research Project at the College Level of China Institute of Labor Relations(No.23XYJS018)+2 种基金the ICH Digitalization and Multi-Source Information Fusion Fujian Provincial University Engineering Research Center 2022 Open Fund Project(G3-KF2207)the China University Industry University Research Innovation Fund(No.2021FNA02009)the Key R&D Program(International Science and Technology Cooperation Project)of Shanxi Province China(No.201903D421003).
文摘Currently,distributed routing protocols are constrained by offering a single path between any pair of nodes,thereby limiting the potential throughput and overall network performance.This approach not only restricts the flow of data but also makes the network susceptible to failures in case the primary path is disrupted.In contrast,routing protocols that leverage multiple paths within the network offer a more resilient and efficient solution.Multipath routing,as a fundamental concept,surpasses the limitations of traditional shortest path first protocols.It not only redirects traffic to unused resources,effectively mitigating network congestion,but also ensures load balancing across the network.This optimization significantly improves network utilization and boosts the overall performance,making it a widely recognized efficient method for enhancing network reliability.To further strengthen network resilience against failures,we introduce a routing scheme known as Multiple Nodes with at least Two Choices(MNTC).This innovative approach aims to significantly enhance network availability by providing each node with at least two routing choices.By doing so,it not only reduces the dependency on a single path but also creates redundant paths that can be utilized in case of failures,thereby enhancing the overall resilience of the network.To ensure the optimal placement of nodes,we propose three incremental deployment algorithms.These algorithms carefully select the most suitable set of nodes for deployment,taking into account various factors such as node connectivity,traffic patterns,and network topology.By deployingMNTCon a carefully chosen set of nodes,we can significantly enhance network reliability without the need for a complete overhaul of the existing infrastructure.We have conducted extensive evaluations of MNTC in diverse topological spaces,demonstrating its effectiveness in maintaining high network availability with minimal path stretch.The results are impressive,showing that even when implemented on just 60%of nodes,our incremental deployment method significantly boosts network availability.This underscores the potential of MNTC in enhancing network resilience and performance,making it a viable solution for modern networks facing increasing demands and complexities.The algorithms OSPF,TBFH,DC and LFC perform fast rerouting based on strict conditions,while MNTC is not restricted by these conditions.In five real network topologies,the average network availability ofMNTCis improved by 14.68%,6.28%,4.76%and 2.84%,respectively,compared with OSPF,TBFH,DC and LFC.
文摘The visions of Industry 4.0 and 5.0 have reinforced the industrial environment.They have also made artificial intelligence incorporated as a major facilitator.Diagnosing machine faults has become a solid foundation for automatically recognizing machine failure,and thus timely maintenance can ensure safe operations.Transfer learning is a promising solution that can enhance the machine fault diagnosis model by borrowing pre-trained knowledge from the source model and applying it to the target model,which typically involves two datasets.In response to the availability of multiple datasets,this paper proposes using selective and adaptive incremental transfer learning(SA-ITL),which fuses three algorithms,namely,the hybrid selective algorithm,the transferability enhancement algorithm,and the incremental transfer learning algorithm.It is a selective algorithm that enables selecting and ordering appropriate datasets for transfer learning and selecting useful knowledge to avoid negative transfer.The algorithm also adaptively adjusts the portion of training data to balance the learning rate and training time.The proposed algorithm is evaluated and analyzed using ten benchmark datasets.Compared with other algorithms from existing works,SA-ITL improves the accuracy of all datasets.Ablation studies present the accuracy enhancements of the SA-ITL,including the hybrid selective algorithm(1.22%-3.82%),transferability enhancement algorithm(1.91%-4.15%),and incremental transfer learning algorithm(0.605%-2.68%).These also show the benefits of enhancing the target model with heterogeneous image datasets that widen the range of domain selection between source and target domains.
基金This work was funded by the National Natural Science Foundation of China Nos.U22A2099,61966009,62006057the Graduate Innovation Program No.YCSW2022286.
文摘Humans are experiencing the inclusion of artificial agents in their lives,such as unmanned vehicles,service robots,voice assistants,and intelligent medical care.If the artificial agents cannot align with social values or make ethical decisions,they may not meet the expectations of humans.Traditionally,an ethical decision-making framework is constructed by rule-based or statistical approaches.In this paper,we propose an ethical decision-making framework based on incremental ILP(Inductive Logic Programming),which can overcome the brittleness of rule-based approaches and little interpretability of statistical approaches.As the current incremental ILP makes it difficult to solve conflicts,we propose a novel ethical decision-making framework considering conflicts in this paper,which adopts our proposed incremental ILP system.The framework consists of two processes:the learning process and the deduction process.The first process records bottom clauses with their score functions and learns rules guided by the entailment and the score function.The second process obtains an ethical decision based on the rules.In an ethical scenario about chatbots for teenagers’mental health,we verify that our framework can learn ethical rules and make ethical decisions.Besides,we extract incremental ILP from the framework and compare it with the state-of-the-art ILP systems based on ASP(Answer Set Programming)focusing on conflict resolution.The results of comparisons show that our proposed system can generate better-quality rules than most other systems.
