Let G be a finite group and H a subgroup of G.The normal index of H in G is defined as the order of K/H_(G),where K is a normal supplement of H in G such that|K|is minimal and H_(G)≤K■G.Let p be a prime which divide...Let G be a finite group and H a subgroup of G.The normal index of H in G is defined as the order of K/H_(G),where K is a normal supplement of H in G such that|K|is minimal and H_(G)≤K■G.Let p be a prime which divides the order of a group G.In this paper,some characterizations of G being p-solvable or p-supersolvable were obtained by analyzing the normal index of certain subgroups of G.These results can be viewed as local version of recent results in the literature.展开更多
On-device Artificial Intelligence(AI)accelerators capable of not only inference but also training neural network models are in increasing demand in the industrial AI field,where frequent retraining is crucial due to f...On-device Artificial Intelligence(AI)accelerators capable of not only inference but also training neural network models are in increasing demand in the industrial AI field,where frequent retraining is crucial due to frequent production changes.Batch normalization(BN)is fundamental to training convolutional neural networks(CNNs),but its implementation in compact accelerator chips remains challenging due to computational complexity,particularly in calculating statistical parameters and gradients across mini-batches.Existing accelerator architectures either compromise the training accuracy of CNNs through approximations or require substantial computational resources,limiting their practical deployment.We present a hardware-optimized BN accelerator that maintains training accuracy while significantly reducing computational overhead through three novel techniques:(1)resourcesharing for efficient resource utilization across forward and backward passes,(2)interleaved buffering for reduced dynamic random-access memory(DRAM)access latencies,and(3)zero-skipping for minimal gradient computation.Implemented on a VCU118 Field Programmable Gate Array(FPGA)on 100 MHz and validated using You Only Look Once version 2-tiny(YOLOv2-tiny)on the PASCALVisualObjectClasses(VOC)dataset,our normalization accelerator achieves a 72%reduction in processing time and 83%lower power consumption compared to a 2.4 GHz Intel Central Processing Unit(CPU)software normalization implementation,while maintaining accuracy(0.51%mean Average Precision(mAP)drop at floating-point 32 bits(FP32),1.35%at brain floating-point 16 bits(bfloat16)).When integrated into a neural processing unit(NPU),the design demonstrates 63%and 97%performance improvements over AMD CPU and Reduced Instruction Set Computing-V(RISC-V)implementations,respectively.These results confirm that our proposed BN hardware design enables efficient,high-accuracy,and power-saving on-device training for modern CNNs.Our results demonstrate that efficient hardware implementation of standard batch normalization is achievable without sacrificing accuracy,enabling practical on-device CNN training with significantly reduced computational and power requirements.展开更多
Renormalization group analysis has been proposed to eliminate secular terms in perturbation solutions of differential equations and thus expand the domain of their validity.Here we extend the method to treat periodic ...Renormalization group analysis has been proposed to eliminate secular terms in perturbation solutions of differential equations and thus expand the domain of their validity.Here we extend the method to treat periodic orbits or limit cycles.Interesting normal forms could be derived through a generalization of the concept'resonance',which offers nontrivial analytic approximations.Compared with traditional techniques such as multi-scale methods,the current scheme proceeds in a very straightforward and simple way,delivering not only the period and the amplitude but also the transient path to limit cycles.The method is demonstrated with several examples including the Duffing oscillator,van der Pol equation and Lorenz equation.The obtained solutions match well with numerical results and with those derived by traditional analytic methods.展开更多
The rapid rise of cyberattacks and the gradual failure of traditional defense systems and approaches led to using artificial intelligence(AI)techniques(such as machine learning(ML)and deep learning(DL))to build more e...The rapid rise of cyberattacks and the gradual failure of traditional defense systems and approaches led to using artificial intelligence(AI)techniques(such as machine learning(ML)and deep learning(DL))to build more efficient and reliable intrusion detection systems(IDSs).However,the advent of larger IDS datasets has negatively impacted the performance and computational complexity of AI-based IDSs.Many researchers used data preprocessing techniques such as feature selection and normalization to overcome such issues.While most of these researchers reported the success of these preprocessing techniques on a shallow level,very few studies have been performed on their effects on a wider scale.Furthermore,the performance of an IDS model is subject to not only the utilized preprocessing techniques but also the dataset and the ML/DL algorithm used,which most of the existing studies give little emphasis on.Thus,this study provides an in-depth analysis of feature selection and normalization effects on IDS models built using three IDS datasets:NSL-KDD,UNSW-NB15,and CSE–CIC–IDS2018,and various AI algorithms.A wrapper-based approach,which tends to give superior performance,and min-max normalization methods were used for feature selection and normalization,respectively.Numerous IDS models were implemented using the full and feature-selected copies of the datasets with and without normalization.The models were evaluated using popular evaluation metrics in IDS modeling,intra-and inter-model comparisons were performed between models and with state-of-the-art works.Random forest(RF)models performed better on NSL-KDD and UNSW-NB15 datasets with accuracies of 99.86%and 96.01%,respectively,whereas artificial neural network(ANN)achieved the best accuracy of 95.43%on the CSE–CIC–IDS2018 dataset.The RF models also achieved an excellent performance compared to recent works.The results show that normalization and feature selection positively affect IDS modeling.Furthermore,while feature selection benefits simpler algorithms(such as RF),normalization is more useful for complex algorithms like ANNs and deep neural networks(DNNs),and algorithms such as Naive Bayes are unsuitable for IDS modeling.The study also found that the UNSW-NB15 and CSE–CIC–IDS2018 datasets are more complex and more suitable for building and evaluating modern-day IDS than the NSL-KDD dataset.Our findings suggest that prioritizing robust algorithms like RF,alongside complex models such as ANN and DNN,can significantly enhance IDS performance.These insights provide valuable guidance for managers to develop more effective security measures by focusing on high detection rates and low false alert rates.展开更多
Background:Isotonic crystalloids are recommended as the first choice for fluid therapy in acute pan-creatitis(AP),with normal saline(NS)and lactate Ringer’s(LR)used most often.Evidence based recom-mendations on the t...Background:Isotonic crystalloids are recommended as the first choice for fluid therapy in acute pan-creatitis(AP),with normal saline(NS)and lactate Ringer’s(LR)used most often.Evidence based recom-mendations on the type of fluid are conflicting and generally come from small single-center randomized controlled trials(RCTs).We therefore conducted a systematic review and meta-analysis to compare the effect of balanced solutions(BS)versus NS on patient-centered clinical outcomes in AP.Methods:From four databases searched up to October 2024,we included only RCTs of adult patients with AP that compared the use of BS(including LR,acetate Ringer’s,etc.)with NS.The primary out-come was the disease advances from AP to moderately severe and severe AP(MSAP/SAP).Trial sequential analyses(TSA)were conducted to control for type-I and type-II errors and Grading of Recommendations Assessment,Development,and Evaluation(GRADE)was used to assess the quality of evidence.Results:Six RCTs were identified and included,involving 260 patients treated with BS and 298 patients with NS.Patients who received the BS had less MSAP/SAP[odds ratio(OR)=0.50,95%confidence in-terval(CI):0.29 to 0.85,P=0.01,I^(2)=0%;5 studies,299 patients],reduced the need of ICU admission(OR=0.60,95%CI:0.39 to 0.93,P=0.02,I^(2)=0%;5 studies,507 patients)and shorter length of hospital stay[mean difference(MD)=-0.88,95%CI:-1.48 to-0.28,P=0.004,I^(2)=0%;6 studies,558 patients;confirmed by TSA with high certainty]compared with those who received NS.The evidence for most of the clinical outcomes was rated as moderate to low due to the risk of bias,imprecision and inconsistency.Conclusions:BS,compared with NS,was associated with improved clinical outcomes in patients with AP.However,given the moderate to low quality of evidence for most of the outcomes assessed,further trials are warranted.展开更多
A conceptual model of intermittent joints is introduced to the cyclic shear test in the laboratory to explore the effects of loading parameters on its shear behavior under cyclic shear loading.The results show that th...A conceptual model of intermittent joints is introduced to the cyclic shear test in the laboratory to explore the effects of loading parameters on its shear behavior under cyclic shear loading.The results show that the loading parameters(initial normal stress,normal stiffness,and shear velocity)determine propagation paths of the wing and secondary cracks in rock bridges during the initial shear cycle,creating different morphologies of macroscopic step-path rupture surfaces and asperities on them.The differences in stress state and rupture surface induce different cyclic shear responses.It shows that high initial normal stress accelerates asperity degradation,raises shear resistance,and promotes compression of intermittent joints.In addition,high normal stiffness provides higher normal stress and shear resistance during the initial cycles and inhibits the dilation and compression of intermittent joints.High shear velocity results in a higher shear resistance,greater dilation,and greater compression.Finally,shear strength is most sensitive to initial normal stress,followed by shear velocity and normal stiffness.Moreover,average dilation angle is most sensitive to initial normal stress,followed by normal stiffness and shear velocity.During the shear cycles,frictional coefficient is affected by asperity degradation,backfilling of rock debris,and frictional area,exhibiting a non-monotonic behavior.展开更多
In this paper,we mainly focus on a type of nonlinear Choquard equations with nonconstant potential.Under appropriate hypotheses on potential function and nonlinear terms,we prove that the above Choquard equation with ...In this paper,we mainly focus on a type of nonlinear Choquard equations with nonconstant potential.Under appropriate hypotheses on potential function and nonlinear terms,we prove that the above Choquard equation with prescribed 2-norm has some normalized solutions by introducing variational methods.展开更多
This article studies a class of nonlinear Kirchhoff equations with exponential critical growth,trapping potential,and perturbation.Under appropriate assumptions about f and h,the article obtained the existence of norm...This article studies a class of nonlinear Kirchhoff equations with exponential critical growth,trapping potential,and perturbation.Under appropriate assumptions about f and h,the article obtained the existence of normalized positive solutions for this equation via the Trudinger-Moser inequality and variational methods.Moreover,these solutions are also ground state solutions.Additionally,the article also characterized the asymptotic behavior of solutions.The results of this article expand the research of relevant literature.展开更多
Due to differences in the distribution of scores for different trials, the performance of a speaker verification system will be seriously diminished if raw scores are directly used for detection with a unified thresho...Due to differences in the distribution of scores for different trials, the performance of a speaker verification system will be seriously diminished if raw scores are directly used for detection with a unified threshold value. As such, the scores must be normalized. To tackle the shortcomings of score normalization methods, we propose a speaker verification system based on log-likelihood normalization (LLN). Without a priori knowledge, LLN increases the separation between scores of target and non-target speaker models, so as to improve score aliasing of “same-speaker” and “different-speaker” trials corresponding to the same test speech, enabling better discrimination and decision capability. The experiment shows that LLN is an effective method of scoring normalization.展开更多
Edge detection is an image processing technique for finding the boundaries of objects within images. It is typically used to interpret gravity and magnetic data, and find the horizontal boundaries of geological bodies...Edge detection is an image processing technique for finding the boundaries of objects within images. It is typically used to interpret gravity and magnetic data, and find the horizontal boundaries of geological bodies. Large deviations between model and true edges are common because of the interference of depth and errors in computing the derivatives; thus, edge detection methods cannot provide information about the depth of the source. To simultaneously obtain the horizontal extent and depth of geophysical anomalies, we use normalized edge detection filters, which normalize the edge detection function at different depths, and the maxima that correspond to the location of the source. The errors between model and actual edges are minimized as the depth of the source decreases and the normalized edge detection method recognizes the extent of the source based on the maxima, allowing for reliable model results. We demonstrate the applicability of the normalized edge detection filters in defining the horizontal extent and depth using synthetic and actual aeromagnetic data.展开更多
The present study focuses on simulating supercavitating projectile tail-slaps with an analytical method.A model of 3σ-normal distribution tail-slaps for a supercavitating projectile is established.Meanwhile,theσ-κe...The present study focuses on simulating supercavitating projectile tail-slaps with an analytical method.A model of 3σ-normal distribution tail-slaps for a supercavitating projectile is established.Meanwhile,theσ-κequation is derived,which is included in this model.Next,the supercavitating projectile tail-slaps are simulated by combining the proposed model and the Logvinovich supercavity section expansion equation.The results show that the number of tail-slaps depends on where the initial several tail-slaps are under the same initial condition.If the distances between the initial several tail-slap positions are large,the number of tail-slaps will considerably decrease,and vice versa.Furthermore,a series of simulations is employed to analyze the influence of the initial angular velocity and the centroid.Analysis of variance is used to evaluate simulation results.The evaluation results suggest that the projectile’s initial angular velocity and centroid have a major impact on the tail-slap number.The larger the value of initial angular velocity,the higher the probability of an increase in tail-slap number.Additionally,the closer the centroid is to the projectile head,the less likely a tail-slap number increase.This study offers important insights into supercavitating projectile tail-slap research.展开更多
Flow and transport properties of fractured crystalline rock are of great interest for different geotechnical ap-plications,such as storage of carbon dioxide,extraction of geothermal energy,or geologic storage of hazar...Flow and transport properties of fractured crystalline rock are of great interest for different geotechnical ap-plications,such as storage of carbon dioxide,extraction of geothermal energy,or geologic storage of hazardous waste.For the long-term safety assessment of geological storage of hazardous waste,the understanding of flow and transport properties through the network of fractures is essential.The flow and transport behaviour can be explored using numerical models to investigate what parameters that affect the results.In this work a pilot study is carried out for multiple realizations of single realistic fractures,using fractal theories,which then are nu-merically sheared using a semi-analytical algorithm.The aperture field is calculated using the average distance of the volume integral of the void that a 1 mm lateral displacement of the sheared surfaces generates.The flow field through the aperture field is solved using Reynolds lubrication equation in linear triangular finite elements.The transport properties,travel length,travel time,transport resistance and specific flow wetted surface,are calculated in a Lagrangian framework using 10,000 particles for each of the 128 flow fields.Evaluating these four metrics,varying initial roughness,4<JRC<10 and normal stress between 0.2 and 20 MPa during shearing,it is concluded that an increase in normal stress generally results in longer travel paths,longer travel times,higher transport resistance and larger specific flow wetted surface while an increase of initial roughness will generally result in longer travel paths,shorter travel times,lower transport resistance and smaller specific flow wetted surface.展开更多
In subsurface projects where the host rock is of low permeability,fractures play an important role in fluid circulation.Both the geometrical and mechanical properties of the fracture are relevant to the permeability o...In subsurface projects where the host rock is of low permeability,fractures play an important role in fluid circulation.Both the geometrical and mechanical properties of the fracture are relevant to the permeability of the fracture.To evaluate this relationship,we numerically generated self-affine fractures reproducing the scaling relationship of the power spectral density(PSD)of the measured fracture surfaces.The fractures were then subjected to a uniform and stepwise increase in normal stress.A fast Fourier transform(FFT)-based elastic contact model was used to simulate the fracture closure.The evolution of fracture contact area,fracture closure,and fracture normal stiffness were determined throughout the whole process.In addition,the fracture permeability at each step was calculated by the local cubic law(LCL).The influences of roughness exponent and correlation length on the fracture hydraulic and mechanical behaviors were investigated.Based on the power law of normal stiffness versus normal stress,the corrected cubic law and the linear relationship between fracture closure and mechanical aperture were obtained from numerical modeling of a set of fractures.Then,we derived a fracture normal stiffness-permeability equation which incorporates fracture geometric parameters such as the root-mean-square(RMS),roughness exponent,and correlation length,which can describe the fracture flow under an effective medium regime and a percolation regime.Finally,we interpreted the flow transition behavior from the effective medium regime to the percolation regime during fracture closure with the established stiffness-permeability function.展开更多
The distinctive characteristics exhibited by the aftershocks of Ms6.0 induced earthquakes in Changning,Sichuan,China,have attracted significant attention.The prevalence of salt rock(halite)in this area is closely asso...The distinctive characteristics exhibited by the aftershocks of Ms6.0 induced earthquakes in Changning,Sichuan,China,have attracted significant attention.The prevalence of salt rock(halite)in this area is closely associated with induced seismic events.The present study was conducted to examine the role of halite in frictional properties.To this end,laboratory measurements were taken for simulated fault gouge composed of halite.Slide-hold-slide(SHS)shear experiments were performed on gouges with grain size<106 mm at constant normal stress from 5 MPa to 30 MPa and constant shear velocity in the range of 1-10 mm/s.Halite gouge shows higher frictional strength and frictional healing rate than most minerals.The results reveal that the fault within halite can potentially generate intense seismic events and more significant aftershocks.An increase in normal stress leads to a reduction in frictional healing,with frictional strength initially increasing and then decreasing.The elevated shear velocity following fault activation facilitates fault dilation,diminishes the frictional strength of the fault,and contributes to fault healing during the inter-seismic period.The aforementioned findings will contribute to a comprehensive understanding of the potential for the healing property of induced seismicity on faults containing halite,particularly in the Changning region of China.展开更多
With the deep advancement of modern educational informatization,the micro-video teaching model has gradually become an effective approach for promoting the innovation and reform of experimental teaching,owing to its a...With the deep advancement of modern educational informatization,the micro-video teaching model has gradually become an effective approach for promoting the innovation and reform of experimental teaching,owing to its advantages such as intuitive visualization,repeatability,and flexible learning.This paper addressed the limitations of the traditional zoology experiment teaching model,which include insufficiently clear teacher demonstrations,limited class time,and the difficulty of accommodating individual student differences.Accordingly,we systematically analyzed the main characteristics,implementation models,and effectiveness of the micro-video teaching model in the Zoology Experiment course.We also discussed the primary challenges encountered during its teaching practice and proposed corresponding recommendations for improvement.This analysis aimed to provide a theoretical reference for the teaching reform of Zoology Experiment in normal universities.展开更多
Anomaly detection in wind turbines involves emphasizing its ability to improve operational efficiency,reduce maintenance costs,extend their lifespan,and enhance reliability in the wind energy sector.This is particular...Anomaly detection in wind turbines involves emphasizing its ability to improve operational efficiency,reduce maintenance costs,extend their lifespan,and enhance reliability in the wind energy sector.This is particularly necessary in offshore wind,currently one of the most critical assets for achieving sustainable energy generation goals,due to the harsh marine environment and the difficulty of maintenance tasks.To address this problem,this work proposes a data-driven methodology for detecting power generation anomalies in offshore wind turbines,using normalized and linearized operational data.The proposed framework transforms heterogeneous wind speed and power measurements into a unified scale,enabling the development of a new wind power index(WPi)that quantifies deviations from expected performance.Additionally,spatial and temporal coherence analyses of turbines within a wind farm ensure the validity of these normalized measurements across different wind turbine models and operating conditions.Furthermore,a Support Vector Machine(SVM)refines the classification process,effectively distinguishing measurement errors from actual power generation failures.Validation of this strategy using real-world data from the Alpha Ventus wind farm demonstrates that the proposed approach not only improves predictive maintenance but also optimizes energy production,highlighting its potential for broad application in offshore wind installations.展开更多
Urban spaces are becoming increasingly congested,and excavations are frequently performed close to existing underground structures such as tunnels.Understanding the mechanical response of proximal soil and tunnels to ...Urban spaces are becoming increasingly congested,and excavations are frequently performed close to existing underground structures such as tunnels.Understanding the mechanical response of proximal soil and tunnels to these excavations is important for efficient and safe underground construction.However,previous investigations of this issue have predominantly made assumptions of plane-strain conditions and normal gravity states,and focused on the performance of tunnels affected by excavation and unloading in sandy strata.In this study,a 3D centrifuge model test is conducted to investigate the influence of excavation on an adjacent existing tunnel in normally consolidated clay.The testing results indicate that the excavation has a significant impact on the horizontal deformation of the retaining wall and tunnel.Moreover,the settlements of the ground surface and the tunnel are mainly affected by the long-term period after excavation.The excavation is found to induce ground movement towards the pit,resulting in prolonged fluctuations in pore water pressure and lateral earth pressure.The testing results are compared with numerical simulations,achieving consistency.A numerical parametric study on the tunnel location shows that when the tunnel is closer to the retaining wall,the decreases in lateral earth pressure and pore water pressure during excavation are more pronounced.展开更多
A grain-oriented silicon steel was normalized with a novel high magnetic field using one-stage cooling process.The effect of high-magnetic-field normalizing on the microstructures and textures was studied with a hot-r...A grain-oriented silicon steel was normalized with a novel high magnetic field using one-stage cooling process.The effect of high-magnetic-field normalizing on the microstructures and textures was studied with a hot-rolled sheet as initial material.It was found that recrystallization and the grain growth were enhanced owing to the external magnetic field driving force.The angle between Goss orientation and magnetic field direction was small,resulting in a high nucleation rate of Goss grains,and hence,the intensity of Goss texture was increased and the deviation angle of Goss grains was reduced after high-magnetic-field normalizing.Furthermore,the migration of dislocation was promoted with an external magnetic field driving force and the density of dislocation decreased,reducing the proportion of low-angle grain boundaries around the Goss grains.The enhancement of recrystallization process and grain growth increased the proportion of high-energy grain boundaries and high-angle grain boundaries,providing a favorable condition for the growth of Goss grains.展开更多
The cultivation of the educator spirit among normal university students provides dual guidance in theory and practice for higher education institutions to train future teachers with noble educational sentiments and pr...The cultivation of the educator spirit among normal university students provides dual guidance in theory and practice for higher education institutions to train future teachers with noble educational sentiments and professional qualities.This study elaborates on the essential connotation of the educator spirit from six perspectives:guiding philosophy,spiritual cultivation,internal requirements,essential requirements,the soul of teacher ethics,and sources of motivation.On this basis,it explains the three major dilemmas in cultivating the educator spirit among normal university students:the cognitive dilemma regarding the understanding of the value of the educator spirit and educational identity,the integration dilemma caused by the disconnection between educational curriculum design and the penetration of the educator spirit,and the alienation dilemma arising from improper connection between educational practice and the educator spirit.Therefore,it is proposed to incorporate the educator spirit into the evaluation system for normal university students,by deepening the understanding of values,optimizing curriculum design,strengthening practice orientation,and comprehensively motivating and evaluating normal university students’recognition and practice of the educator spirit,in order to promote the inheritance and innovation of the educator spirit.展开更多
Deep learning has emerged as a powerful tool for predicting the remaining useful life(RUL)of batteries,contingent upon access to ample data.However,the inherent limitations of data availability from traditional or acc...Deep learning has emerged as a powerful tool for predicting the remaining useful life(RUL)of batteries,contingent upon access to ample data.However,the inherent limitations of data availability from traditional or accelerated life testing pose significant challenges.To mitigate the prediction accuracy issues arising from small sample sizes in existing intelligent methods,we introduce a novel data augmentation framework for RUL prediction.This framework harnesses the inherent high coincidence of degradation patterns exhibited by lithium-ion batteries to pinpoint the knee point,a critical juncture marking a significant shift in the degradation trajectory.By focusing on this critical knee point,we leverage the power of normalizing flow models to generate virtual data,effectively augmenting the training sample size.Additionally,we integrate a Bayesian Long Short-Term Memory network,optimized with Box-Cox transformation,to address the inherent uncertainty associated with predictions based on augmented data.This integration allows for a more nuanced understanding of RUL prediction uncertainties,offering valuable confidence intervals.The efficacy and superiority of the proposed framework are validated through extensive experiments on the CS2 dataset from the University of Maryland and the CrFeMnNiCo dataset from our laboratory.The results clearly demonstrate a substantial improvement in the confidence interval of RUL predictions compared to pre-optimization,highlighting the ability of the framework to achieve high-precision RUL predictions even with limited data.展开更多
基金Supported by the National Natural Science Foundation of China(Grant No.12071092)Guangdong Basic and Applied Basic Research Foundation(Grant No.2025A1515012072)+1 种基金the Natural Science Research Project of Anhui Educational Committee(Grant No.2024AH051298)the Scientific Research Foundation of Bozhou University(Grant No.BYKQ202419).
文摘Let G be a finite group and H a subgroup of G.The normal index of H in G is defined as the order of K/H_(G),where K is a normal supplement of H in G such that|K|is minimal and H_(G)≤K■G.Let p be a prime which divides the order of a group G.In this paper,some characterizations of G being p-solvable or p-supersolvable were obtained by analyzing the normal index of certain subgroups of G.These results can be viewed as local version of recent results in the literature.
基金supported by the National Research Foundation of Korea(NRF)grant for RLRC funded by the Korea government(MSIT)(No.2022R1A5A8026986,RLRC)supported by Institute of Information&Communications Technology Planning&Evaluation(IITP)grant funded by the Korea government(MSIT)(No.2020-0-01304,Development of Self-Learnable Mobile Recursive Neural Network Processor Technology)+3 种基金supported by the MSIT(Ministry of Science and ICT),Republic of Korea,under the Grand Information Technology Research Center support program(IITP-2024-2020-0-01462,Grand-ICT)supervised by the IITP(Institute for Information&Communications Technology Planning&Evaluation)supported by the Korea Technology and Information Promotion Agency for SMEs(TIPA)supported by the Korean government(Ministry of SMEs and Startups)’s Smart Manufacturing Innovation R&D(RS-2024-00434259).
文摘On-device Artificial Intelligence(AI)accelerators capable of not only inference but also training neural network models are in increasing demand in the industrial AI field,where frequent retraining is crucial due to frequent production changes.Batch normalization(BN)is fundamental to training convolutional neural networks(CNNs),but its implementation in compact accelerator chips remains challenging due to computational complexity,particularly in calculating statistical parameters and gradients across mini-batches.Existing accelerator architectures either compromise the training accuracy of CNNs through approximations or require substantial computational resources,limiting their practical deployment.We present a hardware-optimized BN accelerator that maintains training accuracy while significantly reducing computational overhead through three novel techniques:(1)resourcesharing for efficient resource utilization across forward and backward passes,(2)interleaved buffering for reduced dynamic random-access memory(DRAM)access latencies,and(3)zero-skipping for minimal gradient computation.Implemented on a VCU118 Field Programmable Gate Array(FPGA)on 100 MHz and validated using You Only Look Once version 2-tiny(YOLOv2-tiny)on the PASCALVisualObjectClasses(VOC)dataset,our normalization accelerator achieves a 72%reduction in processing time and 83%lower power consumption compared to a 2.4 GHz Intel Central Processing Unit(CPU)software normalization implementation,while maintaining accuracy(0.51%mean Average Precision(mAP)drop at floating-point 32 bits(FP32),1.35%at brain floating-point 16 bits(bfloat16)).When integrated into a neural processing unit(NPU),the design demonstrates 63%and 97%performance improvements over AMD CPU and Reduced Instruction Set Computing-V(RISC-V)implementations,respectively.These results confirm that our proposed BN hardware design enables efficient,high-accuracy,and power-saving on-device training for modern CNNs.Our results demonstrate that efficient hardware implementation of standard batch normalization is achievable without sacrificing accuracy,enabling practical on-device CNN training with significantly reduced computational and power requirements.
文摘Renormalization group analysis has been proposed to eliminate secular terms in perturbation solutions of differential equations and thus expand the domain of their validity.Here we extend the method to treat periodic orbits or limit cycles.Interesting normal forms could be derived through a generalization of the concept'resonance',which offers nontrivial analytic approximations.Compared with traditional techniques such as multi-scale methods,the current scheme proceeds in a very straightforward and simple way,delivering not only the period and the amplitude but also the transient path to limit cycles.The method is demonstrated with several examples including the Duffing oscillator,van der Pol equation and Lorenz equation.The obtained solutions match well with numerical results and with those derived by traditional analytic methods.
文摘The rapid rise of cyberattacks and the gradual failure of traditional defense systems and approaches led to using artificial intelligence(AI)techniques(such as machine learning(ML)and deep learning(DL))to build more efficient and reliable intrusion detection systems(IDSs).However,the advent of larger IDS datasets has negatively impacted the performance and computational complexity of AI-based IDSs.Many researchers used data preprocessing techniques such as feature selection and normalization to overcome such issues.While most of these researchers reported the success of these preprocessing techniques on a shallow level,very few studies have been performed on their effects on a wider scale.Furthermore,the performance of an IDS model is subject to not only the utilized preprocessing techniques but also the dataset and the ML/DL algorithm used,which most of the existing studies give little emphasis on.Thus,this study provides an in-depth analysis of feature selection and normalization effects on IDS models built using three IDS datasets:NSL-KDD,UNSW-NB15,and CSE–CIC–IDS2018,and various AI algorithms.A wrapper-based approach,which tends to give superior performance,and min-max normalization methods were used for feature selection and normalization,respectively.Numerous IDS models were implemented using the full and feature-selected copies of the datasets with and without normalization.The models were evaluated using popular evaluation metrics in IDS modeling,intra-and inter-model comparisons were performed between models and with state-of-the-art works.Random forest(RF)models performed better on NSL-KDD and UNSW-NB15 datasets with accuracies of 99.86%and 96.01%,respectively,whereas artificial neural network(ANN)achieved the best accuracy of 95.43%on the CSE–CIC–IDS2018 dataset.The RF models also achieved an excellent performance compared to recent works.The results show that normalization and feature selection positively affect IDS modeling.Furthermore,while feature selection benefits simpler algorithms(such as RF),normalization is more useful for complex algorithms like ANNs and deep neural networks(DNNs),and algorithms such as Naive Bayes are unsuitable for IDS modeling.The study also found that the UNSW-NB15 and CSE–CIC–IDS2018 datasets are more complex and more suitable for building and evaluating modern-day IDS than the NSL-KDD dataset.Our findings suggest that prioritizing robust algorithms like RF,alongside complex models such as ANN and DNN,can significantly enhance IDS performance.These insights provide valuable guidance for managers to develop more effective security measures by focusing on high detection rates and low false alert rates.
文摘Background:Isotonic crystalloids are recommended as the first choice for fluid therapy in acute pan-creatitis(AP),with normal saline(NS)and lactate Ringer’s(LR)used most often.Evidence based recom-mendations on the type of fluid are conflicting and generally come from small single-center randomized controlled trials(RCTs).We therefore conducted a systematic review and meta-analysis to compare the effect of balanced solutions(BS)versus NS on patient-centered clinical outcomes in AP.Methods:From four databases searched up to October 2024,we included only RCTs of adult patients with AP that compared the use of BS(including LR,acetate Ringer’s,etc.)with NS.The primary out-come was the disease advances from AP to moderately severe and severe AP(MSAP/SAP).Trial sequential analyses(TSA)were conducted to control for type-I and type-II errors and Grading of Recommendations Assessment,Development,and Evaluation(GRADE)was used to assess the quality of evidence.Results:Six RCTs were identified and included,involving 260 patients treated with BS and 298 patients with NS.Patients who received the BS had less MSAP/SAP[odds ratio(OR)=0.50,95%confidence in-terval(CI):0.29 to 0.85,P=0.01,I^(2)=0%;5 studies,299 patients],reduced the need of ICU admission(OR=0.60,95%CI:0.39 to 0.93,P=0.02,I^(2)=0%;5 studies,507 patients)and shorter length of hospital stay[mean difference(MD)=-0.88,95%CI:-1.48 to-0.28,P=0.004,I^(2)=0%;6 studies,558 patients;confirmed by TSA with high certainty]compared with those who received NS.The evidence for most of the clinical outcomes was rated as moderate to low due to the risk of bias,imprecision and inconsistency.Conclusions:BS,compared with NS,was associated with improved clinical outcomes in patients with AP.However,given the moderate to low quality of evidence for most of the outcomes assessed,further trials are warranted.
基金financially supported by the National Natural Science Foundation of China(Grant No.42172292)Taishan Scholars Project Special Funding,and Shandong Energy Group(Grant No.SNKJ 2022A01-R26).
文摘A conceptual model of intermittent joints is introduced to the cyclic shear test in the laboratory to explore the effects of loading parameters on its shear behavior under cyclic shear loading.The results show that the loading parameters(initial normal stress,normal stiffness,and shear velocity)determine propagation paths of the wing and secondary cracks in rock bridges during the initial shear cycle,creating different morphologies of macroscopic step-path rupture surfaces and asperities on them.The differences in stress state and rupture surface induce different cyclic shear responses.It shows that high initial normal stress accelerates asperity degradation,raises shear resistance,and promotes compression of intermittent joints.In addition,high normal stiffness provides higher normal stress and shear resistance during the initial cycles and inhibits the dilation and compression of intermittent joints.High shear velocity results in a higher shear resistance,greater dilation,and greater compression.Finally,shear strength is most sensitive to initial normal stress,followed by shear velocity and normal stiffness.Moreover,average dilation angle is most sensitive to initial normal stress,followed by normal stiffness and shear velocity.During the shear cycles,frictional coefficient is affected by asperity degradation,backfilling of rock debris,and frictional area,exhibiting a non-monotonic behavior.
基金Supported by the National Natural Science Foundation of China(11671403,11671236,12101192)Henan Provincial General Natural Science Foundation Project(232300420113)。
文摘In this paper,we mainly focus on a type of nonlinear Choquard equations with nonconstant potential.Under appropriate hypotheses on potential function and nonlinear terms,we prove that the above Choquard equation with prescribed 2-norm has some normalized solutions by introducing variational methods.
基金Supported by National Natural Science Foundation of China(11671403,11671236)Henan Provincial General Natural Science Foundation Project(232300420113)。
文摘This article studies a class of nonlinear Kirchhoff equations with exponential critical growth,trapping potential,and perturbation.Under appropriate assumptions about f and h,the article obtained the existence of normalized positive solutions for this equation via the Trudinger-Moser inequality and variational methods.Moreover,these solutions are also ground state solutions.Additionally,the article also characterized the asymptotic behavior of solutions.The results of this article expand the research of relevant literature.
文摘Due to differences in the distribution of scores for different trials, the performance of a speaker verification system will be seriously diminished if raw scores are directly used for detection with a unified threshold value. As such, the scores must be normalized. To tackle the shortcomings of score normalization methods, we propose a speaker verification system based on log-likelihood normalization (LLN). Without a priori knowledge, LLN increases the separation between scores of target and non-target speaker models, so as to improve score aliasing of “same-speaker” and “different-speaker” trials corresponding to the same test speech, enabling better discrimination and decision capability. The experiment shows that LLN is an effective method of scoring normalization.
基金supported by the China Postdoctoral Science Foundation (No.2014M551188)the Deep Exploration in China Sinoprobe-09-01 (No.201011078)
文摘Edge detection is an image processing technique for finding the boundaries of objects within images. It is typically used to interpret gravity and magnetic data, and find the horizontal boundaries of geological bodies. Large deviations between model and true edges are common because of the interference of depth and errors in computing the derivatives; thus, edge detection methods cannot provide information about the depth of the source. To simultaneously obtain the horizontal extent and depth of geophysical anomalies, we use normalized edge detection filters, which normalize the edge detection function at different depths, and the maxima that correspond to the location of the source. The errors between model and actual edges are minimized as the depth of the source decreases and the normalized edge detection method recognizes the extent of the source based on the maxima, allowing for reliable model results. We demonstrate the applicability of the normalized edge detection filters in defining the horizontal extent and depth using synthetic and actual aeromagnetic data.
基金Supported by the National Natural Science Foundation of China(Grant No.62101590).
文摘The present study focuses on simulating supercavitating projectile tail-slaps with an analytical method.A model of 3σ-normal distribution tail-slaps for a supercavitating projectile is established.Meanwhile,theσ-κequation is derived,which is included in this model.Next,the supercavitating projectile tail-slaps are simulated by combining the proposed model and the Logvinovich supercavity section expansion equation.The results show that the number of tail-slaps depends on where the initial several tail-slaps are under the same initial condition.If the distances between the initial several tail-slap positions are large,the number of tail-slaps will considerably decrease,and vice versa.Furthermore,a series of simulations is employed to analyze the influence of the initial angular velocity and the centroid.Analysis of variance is used to evaluate simulation results.The evaluation results suggest that the projectile’s initial angular velocity and centroid have a major impact on the tail-slap number.The larger the value of initial angular velocity,the higher the probability of an increase in tail-slap number.Additionally,the closer the centroid is to the projectile head,the less likely a tail-slap number increase.This study offers important insights into supercavitating projectile tail-slap research.
文摘Flow and transport properties of fractured crystalline rock are of great interest for different geotechnical ap-plications,such as storage of carbon dioxide,extraction of geothermal energy,or geologic storage of hazardous waste.For the long-term safety assessment of geological storage of hazardous waste,the understanding of flow and transport properties through the network of fractures is essential.The flow and transport behaviour can be explored using numerical models to investigate what parameters that affect the results.In this work a pilot study is carried out for multiple realizations of single realistic fractures,using fractal theories,which then are nu-merically sheared using a semi-analytical algorithm.The aperture field is calculated using the average distance of the volume integral of the void that a 1 mm lateral displacement of the sheared surfaces generates.The flow field through the aperture field is solved using Reynolds lubrication equation in linear triangular finite elements.The transport properties,travel length,travel time,transport resistance and specific flow wetted surface,are calculated in a Lagrangian framework using 10,000 particles for each of the 128 flow fields.Evaluating these four metrics,varying initial roughness,4<JRC<10 and normal stress between 0.2 and 20 MPa during shearing,it is concluded that an increase in normal stress generally results in longer travel paths,longer travel times,higher transport resistance and larger specific flow wetted surface while an increase of initial roughness will generally result in longer travel paths,shorter travel times,lower transport resistance and smaller specific flow wetted surface.
基金supported by the China Postdoctoral Science Foundation Funded Project(Grant No.2023M740385)the Postdoctoral Fellowship Program of CPSF(Grant No.GZC20233326)the support by the Helmholtz Association's Initiative and Networking Fund for the Helmholtz Young Investigator Group ARES(Contract No.VH-NG-1516).
文摘In subsurface projects where the host rock is of low permeability,fractures play an important role in fluid circulation.Both the geometrical and mechanical properties of the fracture are relevant to the permeability of the fracture.To evaluate this relationship,we numerically generated self-affine fractures reproducing the scaling relationship of the power spectral density(PSD)of the measured fracture surfaces.The fractures were then subjected to a uniform and stepwise increase in normal stress.A fast Fourier transform(FFT)-based elastic contact model was used to simulate the fracture closure.The evolution of fracture contact area,fracture closure,and fracture normal stiffness were determined throughout the whole process.In addition,the fracture permeability at each step was calculated by the local cubic law(LCL).The influences of roughness exponent and correlation length on the fracture hydraulic and mechanical behaviors were investigated.Based on the power law of normal stiffness versus normal stress,the corrected cubic law and the linear relationship between fracture closure and mechanical aperture were obtained from numerical modeling of a set of fractures.Then,we derived a fracture normal stiffness-permeability equation which incorporates fracture geometric parameters such as the root-mean-square(RMS),roughness exponent,and correlation length,which can describe the fracture flow under an effective medium regime and a percolation regime.Finally,we interpreted the flow transition behavior from the effective medium regime to the percolation regime during fracture closure with the established stiffness-permeability function.
基金supported by the National Key Research and Development Project(Grant No.2023YFE0110900)the National Natural Science Foundation of China(Grant Nos.42320104003 and 42077247).
文摘The distinctive characteristics exhibited by the aftershocks of Ms6.0 induced earthquakes in Changning,Sichuan,China,have attracted significant attention.The prevalence of salt rock(halite)in this area is closely associated with induced seismic events.The present study was conducted to examine the role of halite in frictional properties.To this end,laboratory measurements were taken for simulated fault gouge composed of halite.Slide-hold-slide(SHS)shear experiments were performed on gouges with grain size<106 mm at constant normal stress from 5 MPa to 30 MPa and constant shear velocity in the range of 1-10 mm/s.Halite gouge shows higher frictional strength and frictional healing rate than most minerals.The results reveal that the fault within halite can potentially generate intense seismic events and more significant aftershocks.An increase in normal stress leads to a reduction in frictional healing,with frictional strength initially increasing and then decreasing.The elevated shear velocity following fault activation facilitates fault dilation,diminishes the frictional strength of the fault,and contributes to fault healing during the inter-seismic period.The aforementioned findings will contribute to a comprehensive understanding of the potential for the healing property of induced seismicity on faults containing halite,particularly in the Changning region of China.
基金Supported by Undergraduate Higher Education Teaching Quality and Reform Projects of Guangdong Province(Yuejiao Gao Han[2024]No.9)Yuejiao Gao Han[2024]No.30)+5 种基金Curriculum Ideological and Political Reform Demonstration Project of Zhaoqing University(Zhao Xue Yuan[2024]No.83)Key Research Project of Zhaoqing University(ZD202407)Quality Engineering and Teaching Reform Projects of Zhaoqing University(zlgc202207zlgc2024005zlgc202239zlgc2024038).
文摘With the deep advancement of modern educational informatization,the micro-video teaching model has gradually become an effective approach for promoting the innovation and reform of experimental teaching,owing to its advantages such as intuitive visualization,repeatability,and flexible learning.This paper addressed the limitations of the traditional zoology experiment teaching model,which include insufficiently clear teacher demonstrations,limited class time,and the difficulty of accommodating individual student differences.Accordingly,we systematically analyzed the main characteristics,implementation models,and effectiveness of the micro-video teaching model in the Zoology Experiment course.We also discussed the primary challenges encountered during its teaching practice and proposed corresponding recommendations for improvement.This analysis aimed to provide a theoretical reference for the teaching reform of Zoology Experiment in normal universities.
基金supported by the Spanish Ministry of Science and Innovation under the MCI/AEI/FEDER project number PID2021-123543OBC21.
文摘Anomaly detection in wind turbines involves emphasizing its ability to improve operational efficiency,reduce maintenance costs,extend their lifespan,and enhance reliability in the wind energy sector.This is particularly necessary in offshore wind,currently one of the most critical assets for achieving sustainable energy generation goals,due to the harsh marine environment and the difficulty of maintenance tasks.To address this problem,this work proposes a data-driven methodology for detecting power generation anomalies in offshore wind turbines,using normalized and linearized operational data.The proposed framework transforms heterogeneous wind speed and power measurements into a unified scale,enabling the development of a new wind power index(WPi)that quantifies deviations from expected performance.Additionally,spatial and temporal coherence analyses of turbines within a wind farm ensure the validity of these normalized measurements across different wind turbine models and operating conditions.Furthermore,a Support Vector Machine(SVM)refines the classification process,effectively distinguishing measurement errors from actual power generation failures.Validation of this strategy using real-world data from the Alpha Ventus wind farm demonstrates that the proposed approach not only improves predictive maintenance but also optimizes energy production,highlighting its potential for broad application in offshore wind installations.
基金supported by the National Natural Science Foundation of China(Nos.52378341,51938005,and 52090082).
文摘Urban spaces are becoming increasingly congested,and excavations are frequently performed close to existing underground structures such as tunnels.Understanding the mechanical response of proximal soil and tunnels to these excavations is important for efficient and safe underground construction.However,previous investigations of this issue have predominantly made assumptions of plane-strain conditions and normal gravity states,and focused on the performance of tunnels affected by excavation and unloading in sandy strata.In this study,a 3D centrifuge model test is conducted to investigate the influence of excavation on an adjacent existing tunnel in normally consolidated clay.The testing results indicate that the excavation has a significant impact on the horizontal deformation of the retaining wall and tunnel.Moreover,the settlements of the ground surface and the tunnel are mainly affected by the long-term period after excavation.The excavation is found to induce ground movement towards the pit,resulting in prolonged fluctuations in pore water pressure and lateral earth pressure.The testing results are compared with numerical simulations,achieving consistency.A numerical parametric study on the tunnel location shows that when the tunnel is closer to the retaining wall,the decreases in lateral earth pressure and pore water pressure during excavation are more pronounced.
基金supported by the National Natural Science Foundation of China(Nos.52274393,52074200 and 12102310)the Key R&D Program of Hubei Province(No.2023BAB141).
文摘A grain-oriented silicon steel was normalized with a novel high magnetic field using one-stage cooling process.The effect of high-magnetic-field normalizing on the microstructures and textures was studied with a hot-rolled sheet as initial material.It was found that recrystallization and the grain growth were enhanced owing to the external magnetic field driving force.The angle between Goss orientation and magnetic field direction was small,resulting in a high nucleation rate of Goss grains,and hence,the intensity of Goss texture was increased and the deviation angle of Goss grains was reduced after high-magnetic-field normalizing.Furthermore,the migration of dislocation was promoted with an external magnetic field driving force and the density of dislocation decreased,reducing the proportion of low-angle grain boundaries around the Goss grains.The enhancement of recrystallization process and grain growth increased the proportion of high-energy grain boundaries and high-angle grain boundaries,providing a favorable condition for the growth of Goss grains.
基金Study on 2024 Teacher Education Curriculum Reform in Henan Province(2024-JSJYYB-010)。
文摘The cultivation of the educator spirit among normal university students provides dual guidance in theory and practice for higher education institutions to train future teachers with noble educational sentiments and professional qualities.This study elaborates on the essential connotation of the educator spirit from six perspectives:guiding philosophy,spiritual cultivation,internal requirements,essential requirements,the soul of teacher ethics,and sources of motivation.On this basis,it explains the three major dilemmas in cultivating the educator spirit among normal university students:the cognitive dilemma regarding the understanding of the value of the educator spirit and educational identity,the integration dilemma caused by the disconnection between educational curriculum design and the penetration of the educator spirit,and the alienation dilemma arising from improper connection between educational practice and the educator spirit.Therefore,it is proposed to incorporate the educator spirit into the evaluation system for normal university students,by deepening the understanding of values,optimizing curriculum design,strengthening practice orientation,and comprehensively motivating and evaluating normal university students’recognition and practice of the educator spirit,in order to promote the inheritance and innovation of the educator spirit.
基金supported by the National Natural Science Foundation of China(Grant No.62227814,52205040,22279070,and U21A20170)the Natural Science Basic Research Program of Shaanxi(2023-JC-QN-0140)+3 种基金the Young Talent Fund of Xi’an Association for Science and Technology(Grant No.959202313096)the Key Projects of the Shaanxi Province Natural Science Foundation(Grant No.2025JC-QYXQ-038)the Open Foundation of the State Key Laboratory of Fluid Power and Mechatronic Systems(Grant No.GZKF-202430)the National Key Research and Development Program of China(Grant No.2024YFB3311204)。
文摘Deep learning has emerged as a powerful tool for predicting the remaining useful life(RUL)of batteries,contingent upon access to ample data.However,the inherent limitations of data availability from traditional or accelerated life testing pose significant challenges.To mitigate the prediction accuracy issues arising from small sample sizes in existing intelligent methods,we introduce a novel data augmentation framework for RUL prediction.This framework harnesses the inherent high coincidence of degradation patterns exhibited by lithium-ion batteries to pinpoint the knee point,a critical juncture marking a significant shift in the degradation trajectory.By focusing on this critical knee point,we leverage the power of normalizing flow models to generate virtual data,effectively augmenting the training sample size.Additionally,we integrate a Bayesian Long Short-Term Memory network,optimized with Box-Cox transformation,to address the inherent uncertainty associated with predictions based on augmented data.This integration allows for a more nuanced understanding of RUL prediction uncertainties,offering valuable confidence intervals.The efficacy and superiority of the proposed framework are validated through extensive experiments on the CS2 dataset from the University of Maryland and the CrFeMnNiCo dataset from our laboratory.The results clearly demonstrate a substantial improvement in the confidence interval of RUL predictions compared to pre-optimization,highlighting the ability of the framework to achieve high-precision RUL predictions even with limited data.
