The operational demands of a wide range significantly exacerbate combustion instability issues within ramjet combustor.To suppress combustion oscillations,an open-loop control system utilizing Linear Genetic Programmi...The operational demands of a wide range significantly exacerbate combustion instability issues within ramjet combustor.To suppress combustion oscillations,an open-loop control system utilizing Linear Genetic Programming(LGP)has been developed for a full-scale annular ramjet combustor.The LGP is used to generate control laws that include multi-frequency forcing.These laws are then transformed into square waves to actuate the solenoid valve,which modulates the kerosene supply for open-loop control.The results show that the duty cycle has little effect on instability amplitude,whereas an increase in frequency leads to a remarked reduction in combustion amplitude.After five generations evolvements,the pressure amplitude is reduced by 40.6% under the optimal control law generated by LGP.Furthermore,the machine learning process is depicted using a proximity map of control law similarity,with the search pathway visualized by the steepest descent.All individuals go forward to the upper left corner of the map with the evolution process,terminating at the optimal individual of the fifth generation.展开更多
Many spore-forming Bacillus species can cause serious human diseases,because of accidental Bacillusspore infection.Thus,developing an identification strategy with both high sensitivity and specificity is greatly in de...Many spore-forming Bacillus species can cause serious human diseases,because of accidental Bacillusspore infection.Thus,developing an identification strategy with both high sensitivity and specificity is greatly in demand.In this work,we proposed a novel approach named multi-head self-attention mechanism-guided neural network Raman platform to identify living Bacillus spores within a single-cell resolution.The multi-head self-attention mechanism-guided neural network Raman platform was created by combining single-cell Raman spectroscopy,convolutional neural network(CNN),and multi-head self-attention mechanism.To address the limited size of the original spectra dataset,Gaussian noise-based spectra augmentation was employed to increase the number of single-cell Raman spectra datasets for CNN training.Owing to the assistance of both spectra augmentation and multi-head self-attention mechanism,the obtained prediction accuracy of five Bacillus spore species was further improved from 92.29±0.82%to 99.43±0.15%.To figure out the spectra differences covered by the multi-head self-attention mechanism-guided CNN,the relative classification weight from typical Raman bands was visualized via multi-head self-attention mechanism curve.In the process of spectra augmentation from 0 to 1000,the distribution of relative classification weight varied from a discrete state to a more concentrated phase.More importantly,these highlighted four Raman bands(1017,1449,1576,and 1660 cm^(-1))were assigned large weights,showing that the spectra differences in the Raman bands produced the largest contribution to prediction accuracy.It can be foreseen that,our proposed sorting platform has great potential in accurately identifying Bacillus and its related genera species at a single-cell level.展开更多
Abnormal network traffic, as a frequent security risk, requires a series of techniques to categorize and detect it. Existing network traffic anomaly detection still faces challenges: the inability to fully extract loc...Abnormal network traffic, as a frequent security risk, requires a series of techniques to categorize and detect it. Existing network traffic anomaly detection still faces challenges: the inability to fully extract local and global features, as well as the lack of effective mechanisms to capture complex interactions between features;Additionally, when increasing the receptive field to obtain deeper feature representations, the reliance on increasing network depth leads to a significant increase in computational resource consumption, affecting the efficiency and performance of detection. Based on these issues, firstly, this paper proposes a network traffic anomaly detection model based on parallel dilated convolution and residual learning (Res-PDC). To better explore the interactive relationships between features, the traffic samples are converted into two-dimensional matrix. A module combining parallel dilated convolutions and residual learning (res-pdc) was designed to extract local and global features of traffic at different scales. By utilizing res-pdc modules with different dilation rates, we can effectively capture spatial features at different scales and explore feature dependencies spanning wider regions without increasing computational resources. Secondly, to focus and integrate the information in different feature subspaces, further enhance and extract the interactions among the features, multi-head attention is added to Res-PDC, resulting in the final model: multi-head attention enhanced parallel dilated convolution and residual learning (MHA-Res-PDC) for network traffic anomaly detection. Finally, comparisons with other machine learning and deep learning algorithms are conducted on the NSL-KDD and CIC-IDS-2018 datasets. The experimental results demonstrate that the proposed method in this paper can effectively improve the detection performance.展开更多
1.Challenges Thermoacoustic instability in combustors arises from the interaction between sound waves and unsteady heat release,commonly found in systems like gas turbines and aeroengines.This instability leads to und...1.Challenges Thermoacoustic instability in combustors arises from the interaction between sound waves and unsteady heat release,commonly found in systems like gas turbines and aeroengines.This instability leads to undesirable consequences such as structural damage and performance deterioration.The challenge lies in predicting and mitigating these instabilities due to the complex interplay of various physical phenomena like acoustic propagation,turbulent flow,and combustion chemistry,which are summarized in detail in Aimee S.Morgans and Dong Yang's published article.展开更多
The spatiotemporal distribution of soot concentration in aero-engine combustor is important for assessing its combustion performance.Here,we report experimental measurements of soot concentration in terms of Soot Volu...The spatiotemporal distribution of soot concentration in aero-engine combustor is important for assessing its combustion performance.Here,we report experimental measurements of soot concentration in terms of Soot Volume Fraction(SVF)and its spatiotemporal distribution in a single-sector dual-swirl aero-engine combustor using Two-Color Laser-Induced Incandescence(2C-LII).It is shown that soot predominantly forms in the symmetrical vortices of the primary combustion zone,exhibiting a V-type distribution with higher concentration in the lower half of the zone than the upper half,with a small amount distributed in the secondary recirculation zone.Soot emissions at the combustor outlet are relatively low under three typical operating conditions by LII experiments,which is aligned with Smoke Number(SN)from gas analysis.The effect of inlet air temperature on SVF distribution and dynamics in the primary combustion zone is studied,which suggests that the SVF level in the primary combustion zone monotonically increases with the temperature.Meanwhile,the SVF distribution becomes more symmetrical as the inlet temperature increases,although the overall SVF level in the lower half of the zone is still higher.We also investigate the influence of the inlet air pressure on the SVF distribution at the combustor outlet.The soot concentration at the combustor outlet increases with inlet pressure,mainly distributed irregularly across both sides and the center.On both sides,the distribution is continuous,while the center exhibits dot-like and linear patterns.Numerical simulations correlated SVF distribution with the flow field in the primary combustion zone,qualitatively explaining the observed SVF distribution behavior.These results under various conditions can provide valuable insights for improving the performance of this specific combustor and designing high-temperature-rise combustors in the future.展开更多
Sarcasm detection is a complex and challenging task,particularly in the context of Chinese social media,where it exhibits strong contextual dependencies and cultural specificity.To address the limitations of existing ...Sarcasm detection is a complex and challenging task,particularly in the context of Chinese social media,where it exhibits strong contextual dependencies and cultural specificity.To address the limitations of existing methods in capturing the implicit semantics and contextual associations in sarcastic expressions,this paper proposes an event-aware model for Chinese sarcasm detection,leveraging a multi-head attention(MHA)mechanism and contrastive learning(CL)strategies.The proposed model employs a dual-path Bidirectional Encoder Representations from Transformers(BERT)encoder to process comment text and event context separately and integrates an MHA mechanism to facilitate deep interactions between the two,thereby capturing multidimensional semantic associations.Additionally,a CL strategy is introduced to enhance feature representation capabilities,further improving the model’s performance in handling class imbalance and complex contextual scenarios.The model achieves state-of-the-art performance on the Chinese sarcasm dataset,with significant improvements in accuracy(79.55%),F1-score(84.22%),and an area under the curve(AUC,84.35%).展开更多
As the group-buying model shows significant progress in attracting new users,enhancing user engagement,and increasing platform profitability,providing personalized recommendations for group-buying users has emerged as...As the group-buying model shows significant progress in attracting new users,enhancing user engagement,and increasing platform profitability,providing personalized recommendations for group-buying users has emerged as a new challenge in the field of recommendation systems.This paper introduces a group-buying recommendation model based on multi-head attention mechanisms and multi-task learning,termed the Multi-head Attention Mechanisms and Multi-task Learning Group-Buying Recommendation(MAMGBR)model,specifically designed to optimize group-buying recommendations on e-commerce platforms.The core dataset of this study comes from the Chinese maternal and infant e-commerce platform“Beibei,”encompassing approximately 430,000 successful groupbuying actions and over 120,000 users.Themodel focuses on twomain tasks:recommending items for group organizers(Task Ⅰ)and recommending participants for a given group-buying event(Task Ⅱ).In model evaluation,MAMGBR achieves an MRR@10 of 0.7696 for Task I,marking a 20.23%improvement over baseline models.Furthermore,in Task II,where complex interaction patterns prevail,MAMGBR utilizes auxiliary loss functions to effectively model the multifaceted roles of users,items,and participants,leading to a 24.08%increase in MRR@100 under a 1:99 sample ratio.Experimental results show that compared to benchmark models,such as NGCF and EATNN,MAMGBR’s integration ofmulti-head attentionmechanisms,expert networks,and gating mechanisms enables more accurate modeling of user preferences and social associations within group-buying scenarios,significantly enhancing recommendation accuracy and platform group-buying success rates.展开更多
Knowing the optimal operating parameters of Stirling engines is important for efficient combustion through adaptability to changed pressures and oxygen atmospheres. In this study, the optimum operating conditions for ...Knowing the optimal operating parameters of Stirling engines is important for efficient combustion through adaptability to changed pressures and oxygen atmospheres. In this study, the optimum operating conditions for efficient combustion in a singular Stirling engine combustor at different oxygen atmospheres were investigated and determined. Numerical simulations were performed to investigate the effects of ejection ratio and pressure on combustion performance. In an oxygen/carbon dioxide atmosphere, the results show that increasing the ejection ratio substantially alters the flame distribution in the Stirling engine combustor, increasing heat transfer and external combustion efficiency. In contrast, increasing the ejection ratio reduces the average and maximum temperatures of the Stirling engine combustor. Increased pressure affects the flame distribution in the Stirling engine combustor and impedes the flow and convective heat transfer in the combustor, reducing the overall external combustion efficiency at pressures above 6.5 MPa. In an air/carbon dioxide atmosphere, an increased ejection ratio reduces the average and maximum temperatures in the Stirling engine combustor. However, the overall flame distribution does not change substantially. The external combustion efficiency tends to increase and then decrease because of two opposing factors: the increase in the convective heat transfer coefficient and the decrease in the temperature difference. Increasing pressure inhibits forced convection heat transfer in the Stirling engine combustor, reducing external combustion efficiency, which drops from 78% to 65% when pressure increases from 0.2 MPa to 0.5 MPa.展开更多
Safety maintenance of power equipment is of great importance in power grids,in which image-processing-based defect recognition is supposed to classify abnormal conditions during daily inspection.However,owing to the b...Safety maintenance of power equipment is of great importance in power grids,in which image-processing-based defect recognition is supposed to classify abnormal conditions during daily inspection.However,owing to the blurred features of defect images,the current defect recognition algorithm has poor fine-grained recognition ability.Visual attention can achieve fine-grained recognition with its abil-ity to model long-range dependencies while introducing extra computational complexity,especially for multi-head attention in vision transformer structures.Under these circumstances,this paper proposes a self-reduction multi-head attention module that can reduce computational complexity and be easily combined with a Convolutional Neural Network(CNN).In this manner,local and global fea-tures can be calculated simultaneously in our proposed structure,aiming to improve the defect recognition performance.Specifically,the proposed self-reduction multi-head attention can reduce redundant parameters,thereby solving the problem of limited computational resources.Experimental results were obtained based on the defect dataset collected from the substation.The results demonstrated the efficiency and superiority of the proposed method over other advanced algorithms.展开更多
A numerical and experimental study was conducted to investigate the Laser Ablation(LA)ignition mode in an ethylene-fueled supersonic combustor with a cavity flameholder.Theexperiments were operated under a Mach number...A numerical and experimental study was conducted to investigate the Laser Ablation(LA)ignition mode in an ethylene-fueled supersonic combustor with a cavity flameholder.Theexperiments were operated under a Mach number 2.92 supersonic inflow,with stagnation pressureof 2.4 MPa and stagnation temperature of 1600 K.Reynolds-averaged Navier-Stokes simulationswere conducted to characterize the mixing process and flow field structure.This study identifiedfour distinct LA ignition modes.Under the specified condition,laser ablation in zero and negativedefocusing states manifested two distinct ignition modes termed Laser Ablation Direct Ignition(LADI)mode and Laser Ablation Re-Ignition(LARI)mode,correspondingly.LA ignition in alocal small cavity,created by depressing the flow field regulator,could facilitate the ignition modetransforming from LARI mode to Laser Ablation Transition Ignition(LATI)mode.On the otherhand,the elevation of the flow field regulator effectively inhibited the forward propagation of theinitial flame kernel and reduced the dissipation of LA plasma,further enhancing the LADI mode.Based on these characteristics,the LADI mode was subdivided into strong(LADI-S)and weak(LADI-W)modes.Facilitating the transition of ignition modes through alterations in the local flowfield could contribute to attaining a more effective and stable LA ignition.展开更多
Lightweight deep learning models are increasingly required in resource-constrained environments such as mobile devices and the Internet of Medical Things(IoMT).Multi-head convolution with channel attention can facilit...Lightweight deep learning models are increasingly required in resource-constrained environments such as mobile devices and the Internet of Medical Things(IoMT).Multi-head convolution with channel attention can facilitate learning activations relevant to different kernel sizes within a multi-head convolutional layer.Therefore,this study investigates the capability of novel lightweight models incorporating residual multi-head convolution with channel attention(ResMHCNN)blocks to classify medical images.We introduced three novel lightweight deep learning models(BT-Net,LCC-Net,and BC-Net)utilizing the ResMHCNN block as their backbone.These models were crossvalidated and tested on three publicly available medical image datasets:a brain tumor dataset from Figshare consisting of T1-weighted magnetic resonance imaging slices of meningioma,glioma,and pituitary tumors;the LC25000 dataset,which includes microscopic images of lung and colon cancers;and the BreaKHis dataset,containing benign and malignant breast microscopic images.The lightweight models achieved accuracies of 96.9%for 3-class brain tumor classification using BT-Net,and 99.7%for 5-class lung and colon cancer classification using LCC-Net.For 2-class breast cancer classification,BC-Net achieved an accuracy of 96.7%.The parameter counts for the proposed lightweight models—LCC-Net,BC-Net,and BT-Net—are 0.528,0.226,and 1.154 million,respectively.The presented lightweight models,featuring ResMHCNN blocks,may be effectively employed for accurate medical image classification.In the future,these models might be tested for viability in resource-constrained systems such as mobile devices and IoMT platforms.展开更多
Coal dust explosions are severe safety accidents in coal mine production,posing significant threats to life and property.Predicting the maximum explosion pressure(Pm)of coal dust using deep learning models can effecti...Coal dust explosions are severe safety accidents in coal mine production,posing significant threats to life and property.Predicting the maximum explosion pressure(Pm)of coal dust using deep learning models can effectively assess potential risks and provide a scientific basis for preventing coal dust explosions.In this study,a 20-L explosion sphere apparatus was used to test the maximum explosion pressure of coal dust under seven different particle sizes and ten mass concentrations(Cdust),resulting in a dataset of 70 experimental groups.Through Spearman correlation analysis and random forest feature selection methods,particle size(D_(10),D_(20),D_(50))and mass concentration(Cdust)were identified as critical feature parameters from the ten initial parameters of the coal dust samples.Based on this,a hybrid Long Short-Term Memory(LSTM)network model incorporating a Multi-Head Attention Mechanism and the Sparrow Search Algorithm(SSA)was proposed to predict the maximum explosion pressure of coal dust.The results demonstrate that the SSA-LSTM-Multi-Head Attention model excels in predicting the maximum explosion pressure of coal dust.The four evaluation metrics indicate that the model achieved a coefficient of determination(R^(2)),root mean square error(RMSE),mean absolute percentage error(MAPE),and mean absolute error(MAE)of 0.9841,0.0030,0.0074,and 0.0049,respectively,in the training set.In the testing set,these values were 0.9743,0.0087,0.0108,and 0.0069,respectively.Compared to artificial neural networks(ANN),random forest(RF),support vector machines(SVM),particle swarm optimized-SVM(PSO-SVM)neural networks,and the traditional single-model LSTM,the SSA-LSTM-Multi-Head Attention model demonstrated superior generalization capability and prediction accuracy.The findings of this study not only advance the application of deep learning in coal dust explosion prediction but also provide robust technical support for the prevention and risk assessment of coal dust explosions.展开更多
The self-attention mechanism of Transformers,which captures long-range contextual information,has demonstrated significant potential in image segmentation.However,their ability to learn local,contextual relationships ...The self-attention mechanism of Transformers,which captures long-range contextual information,has demonstrated significant potential in image segmentation.However,their ability to learn local,contextual relationships between pixels requires further improvement.Previous methods face challenges in efficiently managing multi-scale fea-tures of different granularities from the encoder backbone,leaving room for improvement in their global representation and feature extraction capabilities.To address these challenges,we propose a novel Decoder with Multi-Head Feature Receptors(DMHFR),which receives multi-scale features from the encoder backbone and organizes them into three feature groups with different granularities:coarse,fine-grained,and full set.These groups are subsequently processed by Multi-Head Feature Receptors(MHFRs)after feature capture and modeling operations.MHFRs include two Three-Head Feature Receptors(THFRs)and one Four-Head Feature Receptor(FHFR).Each group of features is passed through these MHFRs and then fed into axial transformers,which help the model capture long-range dependencies within the features.The three MHFRs produce three distinct feature outputs.The output from the FHFR serves as auxiliary auxiliary features in the prediction head,and the prediction output and their losses will eventually be aggregated.Experimental results show that the Transformer using DMHFR outperforms 15 state of the arts(SOTA)methods on five public datasets.Specifically,it achieved significant improvements in mean DICE scores over the classic Parallel Reverse Attention Network(PraNet)method,with gains of 4.1%,2.2%,1.4%,8.9%,and 16.3%on the CVC-ClinicDB,Kvasir-SEG,CVC-T,CVC-ColonDB,and ETIS-LaribPolypDB datasets,respectively.展开更多
Traffic flow prediction is a crucial element of intelligent transportation systems.However,accu-rate traffic flow prediction is quite challenging because of its highly nonlinear,complex,and dynam-ic characteristics.To...Traffic flow prediction is a crucial element of intelligent transportation systems.However,accu-rate traffic flow prediction is quite challenging because of its highly nonlinear,complex,and dynam-ic characteristics.To address the difficulties in simultaneously capturing local and global dynamic spatiotemporal correlations in traffic flow,as well as the high time complexity of existing models,a multi-head flow attention-based local-global dynamic hypergraph convolution(MFA-LGDHC)pre-diction model is proposed.which consists of multi-head flow attention(MHFA)mechanism,graph convolution network(GCN),and local-global dynamic hypergraph convolution(LGHC).MHFA is utilized to extract the time dependency of traffic flow and reduce the time complexity of the model.GCN is employed to catch the spatial dependency of traffic flow.LGHC utilizes down-sampling con-volution and isometric convolution to capture the local and global spatial dependencies of traffic flow.And dynamic hypergraph convolution is used to model the dynamic higher-order relationships of the traffic road network.Experimental results indicate that the MFA-LGDHC model outperforms current popular baseline models and exhibits good prediction performance.展开更多
Precise and quantitative measurement of soot particle emission plays an essential role in accurately assessing the combustion performance of aero-engine combustors and infrared signature levels in aircraft exhausts.Am...Precise and quantitative measurement of soot particle emission plays an essential role in accurately assessing the combustion performance of aero-engine combustors and infrared signature levels in aircraft exhausts.Among various intrusive or non-intrusive approaches for soot diagnostics,Laser-Induced Incandescence(LⅡ)technique has been increasingly applied for soot concentration measurement in various combustion environments such as laminar flames and internal combustion engines due to its high spatial resolution and sensitivity.As for LⅡmeasurement in aero-engine combustors,however,it normally suffers from very limited optical accesses and often faces mandatory requirements of oblique imaging from a small backward angle.In this work,we demonstrate a Two-Color(2C)LⅡsystem that simultaneously captures LⅡsignal images at two distinct wavelengths using a Scheimpflug imaging configuration.A projective transformation algorithm and image overlapping procedures were employed to spatially correct the raw Scheimpflug LⅡimages.Performance validation of the developed 2C-Scheimpflug LⅡsystem was first conducted under specified conditions in a laminar C_(2)H_(4)/air McKenna flame.The obtained Soot Volume Fraction(SVF)level and its spatial distribution are in consistent with previous studies under identical flame conditions reported by other research groups.Finally,as a demonstration of engineering benchmark application,we applied the developed 2C-Scheimpflug LⅡsystem to measure SVF distribution in the cross-section plane perpendicular to the direction of flame propagation at the exhaust of a single-sector dual-swirl aero-engine model combustor.Transient soot production events were observed and characteristics of the SVF distribution were investigated.These experimental results suggest the feasibility of the 2C-Scheimpflug LⅡtechnique developed in this work for precise and quantitative measurements of soot concentration in practical environments.展开更多
This paper describes an experimental study investigating the effects of sinusoidal pulsed injection on the combustion mode transition in a dual-mode supersonic combustor.The results are obtained under inflow condition...This paper describes an experimental study investigating the effects of sinusoidal pulsed injection on the combustion mode transition in a dual-mode supersonic combustor.The results are obtained under inflow conditions of 2.9 MPa stagnation pressure,1900 K stagnation temperature,and Mach number of 3.0.It has been observed that,at the same equivalence ratio,the combustion mode and flow field structure undergo irreversible changes from a weak combustion state to a strong combustion state at a specific pulsed jet frequency compared to steady jet.For steady jet,the combustion mode is dual-mode.As the frequency of the unsteady jet changes,the combustion mode also changes:it becomes a transition mode at frequencies of 171 Hz and 260 Hz,and a ramjet mode at 216 Hz.Combustion instability under steady jet manifests as a transition in flame stabilization mode.In contrast,under pulsed jet,combustion instability appears either as a transition in flame stabilization mode or as flame blow-off and flashback.The flow field oscillation frequency in the non-reacting flow is 171 Hz,which may resonate with the 171 Hz pulsed jet frequency,making the combustion oscillations most pronounced at this frequency.When the jet frequency is increased to 216 Hz,the combustion intensity significantly increases,and the combustion mode transfers to the ramjet mode.However,further increasing the frequency to 260 Hz results in a decrease in combustion intensity,returning to the transition mode.The frequency of the flow field oscillations varies with the coupling of the pulsed injection frequency,shock wave,and flame,and if the system reaches an unstable state,that is,pre-combustion shock train moves far upstream of the isolator during the pulsed jet period,strong combustion state can be achieved,and this process is irreversible.展开更多
Quantitative measurement of Soot Volume Fraction (SVF) is an essential prerequisite for controlling soot particle emissions from aero-engine combustors. As an in-situ and non-intrusive optical diagnostic technique, La...Quantitative measurement of Soot Volume Fraction (SVF) is an essential prerequisite for controlling soot particle emissions from aero-engine combustors. As an in-situ and non-intrusive optical diagnostic technique, Laser-Induced Incandescence (LII) has been increasingly applied for soot concentration quantification in various combustion environments such as laminar flame, vehicle exhaust, internal combustion chamber as well as aero-engine combustor. In this work, we experimentally measured the spatial and temporal distribution of SVF using two-color LII technique at the outlet of a single-sector dual-swirl aero-engine model combustor. The effect of inlet pressure and air preheat temperature on the SVF distribution was separately investigated within a pressure range of 241–425 kPa and a temperature range of 292–500 K. The results show that soot production increases with the inlet pressure but generally decreases with the air preheat temperature. Qualitative analysis was provided to explain the above results of parametric studies. The LII experiments were also conducted under 3 designed conditions to evaluate soot emission under practical operations. Particularly, weak soot emission was detected at the outlet under the idle condition. Our experimental results provide a valuable benchmark for evaluating soot emission in the exhaust plume of this aero-engine combustor during practical operations.展开更多
Brain tumor identification is a challenging task in neuro-oncology.The brain’s complex anatomy makes it a crucial part of the central nervous system.Accurate tumor classification is crucial for clinical diagnosis and...Brain tumor identification is a challenging task in neuro-oncology.The brain’s complex anatomy makes it a crucial part of the central nervous system.Accurate tumor classification is crucial for clinical diagnosis and treatment planning.This research presents a significant advancement in the multi-classification of brain tumors.This paper proposed a novel architecture that integrates Enhanced ResNeXt 101_32×8d,a Convolutional Neural Network(CNN)with a multi-head self-attention(MHSA)mechanism.This combination harnesses the strengths of the feature extraction,feature representation by CNN,and long-range dependencies by MHSA.Magnetic Resonance Imaging(MRI)datasets were employed to check the effectiveness of the proposed architecture.The first dataset(DS-1,Msoud)included four brain tumor classes,and the second dataset(DS-2)contained seven brain tumor classes.This methodology effectively distinguished various tumor classes,achieving high accuracies of 99.75% on DS-1 and 98.80% on DS-2.These impressive results indicate the superior performance and adaptability of our model for multiclass brain tumor classification.Evaluationmetrics such as accuracy,precision,recall,F1 score,and ROC(receiver operating characteristic)curve were utilized to comprehensively evaluate model validity.The performance results showed that the model is well-suited for clinical applications,with reduced errors and high accuracy.展开更多
To provide advanced diagnostic techniques for diagnosing the outlet temperature distribution and species concentrations of future advanced combustors,this study focuses on a dual-swirl single-dome rectangular combusto...To provide advanced diagnostic techniques for diagnosing the outlet temperature distribution and species concentrations of future advanced combustors,this study focuses on a dual-swirl single-dome rectangular combustor.Through the integration of multiple diagnostics,simultaneous measurement of outlet temperature distribution and species concentrations was achieved.The study validates the engineering applicability of these simultaneous measurements using tungsten-rhenium(W-Re)thermocouples and Coherent Anti-Stokes Raman Scattering(CARS),CARS and Tunable Diode Laser Absorption Spectroscopy(TDLAS),as well as Gas Analysis(GA)and Mass Spectrometry(MS).The results demonstrate that measurements by thermocouples and CARS exhibit good consistency and repeatability,with a relative deviation of less than 4%,fully meeting the requirements of engineering experiments.The spatial distribution reconstruction results of TDLAS can reflect the temperature distribution characteristics at the combustor outlet.Temperature comparison between TDLAS and CARS at single-point positions shows consistent results,with a relative deviation of less than 11%and 7%under both conditions,respectively.Simultaneous measurements by integrating GA and MS show high engineering applicability for the first time,meeting the requirements for measuring both inorganic species and free radicals at the combustor outlet.Under C_(1)condition,the relative deviations of four key species(Unburned Hydrocarbon(UHC),NO,O_(2),and CO_(2))remain within 2%,while that of NO_(2)is slightly higher at approximately 8%.Under C_(2)condition,the overall deviations increase for most species,with only O_(2)and CO_(2)maintaining relatively low deviations.The primary species of UHCs at the combustor outlet under both conditions are small molecular hydrocarbons(C_(3)-C_(8))and RO_(2)radicals,accounting for over 90%of total UHC.Specifically,RO_(2)species(R is C_(1)-C_(2)alkyl groups)are the predominant species,accounting for 74.3%and 82.1%of total RO_(2)under both conditions,respectively.These integrated diagnostic methods for temperature and species concentrations at the combustor outlet serve as a crucial reference for its engineering applications.展开更多
基金support from the National Natural Science Foundation of China(No.12002372)the Young Elite Scientists Sponsorship Program by China Association for Science and Technology(No.2022QNRC001)the Natural Science Foundation of Hunan Province,China(No.2021JJ40674)。
文摘The operational demands of a wide range significantly exacerbate combustion instability issues within ramjet combustor.To suppress combustion oscillations,an open-loop control system utilizing Linear Genetic Programming(LGP)has been developed for a full-scale annular ramjet combustor.The LGP is used to generate control laws that include multi-frequency forcing.These laws are then transformed into square waves to actuate the solenoid valve,which modulates the kerosene supply for open-loop control.The results show that the duty cycle has little effect on instability amplitude,whereas an increase in frequency leads to a remarked reduction in combustion amplitude.After five generations evolvements,the pressure amplitude is reduced by 40.6% under the optimal control law generated by LGP.Furthermore,the machine learning process is depicted using a proximity map of control law similarity,with the search pathway visualized by the steepest descent.All individuals go forward to the upper left corner of the map with the evolution process,terminating at the optimal individual of the fifth generation.
基金partially supported by the National Natural Science Foundation of China(62075137)the Guangdong Basic and Applied Basic Research Foundation(2023A1515140161)+3 种基金the Guangxi Natural Science Foundation of China(2021JJB 110003)the Dongguan Science and Technology of Social Development Program(20231800936312)the high-level talent program of Dongguan University of Technology(No.221110080)the Sanming Project of Medicine in Shenzhen(No.SZSM202103014).
文摘Many spore-forming Bacillus species can cause serious human diseases,because of accidental Bacillusspore infection.Thus,developing an identification strategy with both high sensitivity and specificity is greatly in demand.In this work,we proposed a novel approach named multi-head self-attention mechanism-guided neural network Raman platform to identify living Bacillus spores within a single-cell resolution.The multi-head self-attention mechanism-guided neural network Raman platform was created by combining single-cell Raman spectroscopy,convolutional neural network(CNN),and multi-head self-attention mechanism.To address the limited size of the original spectra dataset,Gaussian noise-based spectra augmentation was employed to increase the number of single-cell Raman spectra datasets for CNN training.Owing to the assistance of both spectra augmentation and multi-head self-attention mechanism,the obtained prediction accuracy of five Bacillus spore species was further improved from 92.29±0.82%to 99.43±0.15%.To figure out the spectra differences covered by the multi-head self-attention mechanism-guided CNN,the relative classification weight from typical Raman bands was visualized via multi-head self-attention mechanism curve.In the process of spectra augmentation from 0 to 1000,the distribution of relative classification weight varied from a discrete state to a more concentrated phase.More importantly,these highlighted four Raman bands(1017,1449,1576,and 1660 cm^(-1))were assigned large weights,showing that the spectra differences in the Raman bands produced the largest contribution to prediction accuracy.It can be foreseen that,our proposed sorting platform has great potential in accurately identifying Bacillus and its related genera species at a single-cell level.
基金supported by the Xiamen Science and Technology Subsidy Project(No.2023CXY0318).
文摘Abnormal network traffic, as a frequent security risk, requires a series of techniques to categorize and detect it. Existing network traffic anomaly detection still faces challenges: the inability to fully extract local and global features, as well as the lack of effective mechanisms to capture complex interactions between features;Additionally, when increasing the receptive field to obtain deeper feature representations, the reliance on increasing network depth leads to a significant increase in computational resource consumption, affecting the efficiency and performance of detection. Based on these issues, firstly, this paper proposes a network traffic anomaly detection model based on parallel dilated convolution and residual learning (Res-PDC). To better explore the interactive relationships between features, the traffic samples are converted into two-dimensional matrix. A module combining parallel dilated convolutions and residual learning (res-pdc) was designed to extract local and global features of traffic at different scales. By utilizing res-pdc modules with different dilation rates, we can effectively capture spatial features at different scales and explore feature dependencies spanning wider regions without increasing computational resources. Secondly, to focus and integrate the information in different feature subspaces, further enhance and extract the interactions among the features, multi-head attention is added to Res-PDC, resulting in the final model: multi-head attention enhanced parallel dilated convolution and residual learning (MHA-Res-PDC) for network traffic anomaly detection. Finally, comparisons with other machine learning and deep learning algorithms are conducted on the NSL-KDD and CIC-IDS-2018 datasets. The experimental results demonstrate that the proposed method in this paper can effectively improve the detection performance.
文摘1.Challenges Thermoacoustic instability in combustors arises from the interaction between sound waves and unsteady heat release,commonly found in systems like gas turbines and aeroengines.This instability leads to undesirable consequences such as structural damage and performance deterioration.The challenge lies in predicting and mitigating these instabilities due to the complex interplay of various physical phenomena like acoustic propagation,turbulent flow,and combustion chemistry,which are summarized in detail in Aimee S.Morgans and Dong Yang's published article.
基金support of the National Science and Technology Major Project of China(No.J2019-V-0005-0096)the National Key Research and Development Program of China(No.2020YFA0405700).
文摘The spatiotemporal distribution of soot concentration in aero-engine combustor is important for assessing its combustion performance.Here,we report experimental measurements of soot concentration in terms of Soot Volume Fraction(SVF)and its spatiotemporal distribution in a single-sector dual-swirl aero-engine combustor using Two-Color Laser-Induced Incandescence(2C-LII).It is shown that soot predominantly forms in the symmetrical vortices of the primary combustion zone,exhibiting a V-type distribution with higher concentration in the lower half of the zone than the upper half,with a small amount distributed in the secondary recirculation zone.Soot emissions at the combustor outlet are relatively low under three typical operating conditions by LII experiments,which is aligned with Smoke Number(SN)from gas analysis.The effect of inlet air temperature on SVF distribution and dynamics in the primary combustion zone is studied,which suggests that the SVF level in the primary combustion zone monotonically increases with the temperature.Meanwhile,the SVF distribution becomes more symmetrical as the inlet temperature increases,although the overall SVF level in the lower half of the zone is still higher.We also investigate the influence of the inlet air pressure on the SVF distribution at the combustor outlet.The soot concentration at the combustor outlet increases with inlet pressure,mainly distributed irregularly across both sides and the center.On both sides,the distribution is continuous,while the center exhibits dot-like and linear patterns.Numerical simulations correlated SVF distribution with the flow field in the primary combustion zone,qualitatively explaining the observed SVF distribution behavior.These results under various conditions can provide valuable insights for improving the performance of this specific combustor and designing high-temperature-rise combustors in the future.
基金granted by Qin Xin Talents Cultivation Program(No.QXTCP C202115),Beijing Information Science&Technology Universitythe Beijing Advanced Innovation Center for Future Blockchain and Privacy Computing Fund(No.GJJ-23),National Social Science Foundation,China(No.21BTQ079).
文摘Sarcasm detection is a complex and challenging task,particularly in the context of Chinese social media,where it exhibits strong contextual dependencies and cultural specificity.To address the limitations of existing methods in capturing the implicit semantics and contextual associations in sarcastic expressions,this paper proposes an event-aware model for Chinese sarcasm detection,leveraging a multi-head attention(MHA)mechanism and contrastive learning(CL)strategies.The proposed model employs a dual-path Bidirectional Encoder Representations from Transformers(BERT)encoder to process comment text and event context separately and integrates an MHA mechanism to facilitate deep interactions between the two,thereby capturing multidimensional semantic associations.Additionally,a CL strategy is introduced to enhance feature representation capabilities,further improving the model’s performance in handling class imbalance and complex contextual scenarios.The model achieves state-of-the-art performance on the Chinese sarcasm dataset,with significant improvements in accuracy(79.55%),F1-score(84.22%),and an area under the curve(AUC,84.35%).
基金supported by the Key Research and Development Program of Heilongjiang Province(No.2022ZX01A35).
文摘As the group-buying model shows significant progress in attracting new users,enhancing user engagement,and increasing platform profitability,providing personalized recommendations for group-buying users has emerged as a new challenge in the field of recommendation systems.This paper introduces a group-buying recommendation model based on multi-head attention mechanisms and multi-task learning,termed the Multi-head Attention Mechanisms and Multi-task Learning Group-Buying Recommendation(MAMGBR)model,specifically designed to optimize group-buying recommendations on e-commerce platforms.The core dataset of this study comes from the Chinese maternal and infant e-commerce platform“Beibei,”encompassing approximately 430,000 successful groupbuying actions and over 120,000 users.Themodel focuses on twomain tasks:recommending items for group organizers(Task Ⅰ)and recommending participants for a given group-buying event(Task Ⅱ).In model evaluation,MAMGBR achieves an MRR@10 of 0.7696 for Task I,marking a 20.23%improvement over baseline models.Furthermore,in Task II,where complex interaction patterns prevail,MAMGBR utilizes auxiliary loss functions to effectively model the multifaceted roles of users,items,and participants,leading to a 24.08%increase in MRR@100 under a 1:99 sample ratio.Experimental results show that compared to benchmark models,such as NGCF and EATNN,MAMGBR’s integration ofmulti-head attentionmechanisms,expert networks,and gating mechanisms enables more accurate modeling of user preferences and social associations within group-buying scenarios,significantly enhancing recommendation accuracy and platform group-buying success rates.
基金Supported by the Shanghai Rising Star Program (Grant No. 21QB1403900)the Shanghai Municipal Commission of Science and Technology (Grant No. 22170712600)。
文摘Knowing the optimal operating parameters of Stirling engines is important for efficient combustion through adaptability to changed pressures and oxygen atmospheres. In this study, the optimum operating conditions for efficient combustion in a singular Stirling engine combustor at different oxygen atmospheres were investigated and determined. Numerical simulations were performed to investigate the effects of ejection ratio and pressure on combustion performance. In an oxygen/carbon dioxide atmosphere, the results show that increasing the ejection ratio substantially alters the flame distribution in the Stirling engine combustor, increasing heat transfer and external combustion efficiency. In contrast, increasing the ejection ratio reduces the average and maximum temperatures of the Stirling engine combustor. Increased pressure affects the flame distribution in the Stirling engine combustor and impedes the flow and convective heat transfer in the combustor, reducing the overall external combustion efficiency at pressures above 6.5 MPa. In an air/carbon dioxide atmosphere, an increased ejection ratio reduces the average and maximum temperatures in the Stirling engine combustor. However, the overall flame distribution does not change substantially. The external combustion efficiency tends to increase and then decrease because of two opposing factors: the increase in the convective heat transfer coefficient and the decrease in the temperature difference. Increasing pressure inhibits forced convection heat transfer in the Stirling engine combustor, reducing external combustion efficiency, which drops from 78% to 65% when pressure increases from 0.2 MPa to 0.5 MPa.
基金supported in part by Major Program of the National Natural Science Foundation of China under Grant 62127803.
文摘Safety maintenance of power equipment is of great importance in power grids,in which image-processing-based defect recognition is supposed to classify abnormal conditions during daily inspection.However,owing to the blurred features of defect images,the current defect recognition algorithm has poor fine-grained recognition ability.Visual attention can achieve fine-grained recognition with its abil-ity to model long-range dependencies while introducing extra computational complexity,especially for multi-head attention in vision transformer structures.Under these circumstances,this paper proposes a self-reduction multi-head attention module that can reduce computational complexity and be easily combined with a Convolutional Neural Network(CNN).In this manner,local and global fea-tures can be calculated simultaneously in our proposed structure,aiming to improve the defect recognition performance.Specifically,the proposed self-reduction multi-head attention can reduce redundant parameters,thereby solving the problem of limited computational resources.Experimental results were obtained based on the defect dataset collected from the substation.The results demonstrated the efficiency and superiority of the proposed method over other advanced algorithms.
基金supported by the National Natural Science Foundation of China(Nos.12272408 and 11925207)the Natural Science Foundation for Distinguished Young Scholars of Hunan Province,China(No.2024J12057)。
文摘A numerical and experimental study was conducted to investigate the Laser Ablation(LA)ignition mode in an ethylene-fueled supersonic combustor with a cavity flameholder.Theexperiments were operated under a Mach number 2.92 supersonic inflow,with stagnation pressureof 2.4 MPa and stagnation temperature of 1600 K.Reynolds-averaged Navier-Stokes simulationswere conducted to characterize the mixing process and flow field structure.This study identifiedfour distinct LA ignition modes.Under the specified condition,laser ablation in zero and negativedefocusing states manifested two distinct ignition modes termed Laser Ablation Direct Ignition(LADI)mode and Laser Ablation Re-Ignition(LARI)mode,correspondingly.LA ignition in alocal small cavity,created by depressing the flow field regulator,could facilitate the ignition modetransforming from LARI mode to Laser Ablation Transition Ignition(LATI)mode.On the otherhand,the elevation of the flow field regulator effectively inhibited the forward propagation of theinitial flame kernel and reduced the dissipation of LA plasma,further enhancing the LADI mode.Based on these characteristics,the LADI mode was subdivided into strong(LADI-S)and weak(LADI-W)modes.Facilitating the transition of ignition modes through alterations in the local flowfield could contribute to attaining a more effective and stable LA ignition.
基金supported by the Institute of Information&Communications Technology Planning&Evaluation(IITP)-Innovative Human Resource Development for Local Intellectualization program grant funded by the Korea government(MSIT)(IITP-2025-RS-2023-00259678)by INHA UNIVERSITY Research Grant.
文摘Lightweight deep learning models are increasingly required in resource-constrained environments such as mobile devices and the Internet of Medical Things(IoMT).Multi-head convolution with channel attention can facilitate learning activations relevant to different kernel sizes within a multi-head convolutional layer.Therefore,this study investigates the capability of novel lightweight models incorporating residual multi-head convolution with channel attention(ResMHCNN)blocks to classify medical images.We introduced three novel lightweight deep learning models(BT-Net,LCC-Net,and BC-Net)utilizing the ResMHCNN block as their backbone.These models were crossvalidated and tested on three publicly available medical image datasets:a brain tumor dataset from Figshare consisting of T1-weighted magnetic resonance imaging slices of meningioma,glioma,and pituitary tumors;the LC25000 dataset,which includes microscopic images of lung and colon cancers;and the BreaKHis dataset,containing benign and malignant breast microscopic images.The lightweight models achieved accuracies of 96.9%for 3-class brain tumor classification using BT-Net,and 99.7%for 5-class lung and colon cancer classification using LCC-Net.For 2-class breast cancer classification,BC-Net achieved an accuracy of 96.7%.The parameter counts for the proposed lightweight models—LCC-Net,BC-Net,and BT-Net—are 0.528,0.226,and 1.154 million,respectively.The presented lightweight models,featuring ResMHCNN blocks,may be effectively employed for accurate medical image classification.In the future,these models might be tested for viability in resource-constrained systems such as mobile devices and IoMT platforms.
基金funded by the Research on Intelligent Mining Geological Model and Ventilation Model for Extremely Thin Coal Seam in Heilongjiang Province,China(2021ZXJ02A03)the Demonstration of Intelligent Mining for Comprehensive Mining Face in Extremely Thin Coal Seam in Heilongjiang Province,China(2021ZXJ02A04)the Natural Science Foundation of Heilongjiang Province,China(LH2024E112).
文摘Coal dust explosions are severe safety accidents in coal mine production,posing significant threats to life and property.Predicting the maximum explosion pressure(Pm)of coal dust using deep learning models can effectively assess potential risks and provide a scientific basis for preventing coal dust explosions.In this study,a 20-L explosion sphere apparatus was used to test the maximum explosion pressure of coal dust under seven different particle sizes and ten mass concentrations(Cdust),resulting in a dataset of 70 experimental groups.Through Spearman correlation analysis and random forest feature selection methods,particle size(D_(10),D_(20),D_(50))and mass concentration(Cdust)were identified as critical feature parameters from the ten initial parameters of the coal dust samples.Based on this,a hybrid Long Short-Term Memory(LSTM)network model incorporating a Multi-Head Attention Mechanism and the Sparrow Search Algorithm(SSA)was proposed to predict the maximum explosion pressure of coal dust.The results demonstrate that the SSA-LSTM-Multi-Head Attention model excels in predicting the maximum explosion pressure of coal dust.The four evaluation metrics indicate that the model achieved a coefficient of determination(R^(2)),root mean square error(RMSE),mean absolute percentage error(MAPE),and mean absolute error(MAE)of 0.9841,0.0030,0.0074,and 0.0049,respectively,in the training set.In the testing set,these values were 0.9743,0.0087,0.0108,and 0.0069,respectively.Compared to artificial neural networks(ANN),random forest(RF),support vector machines(SVM),particle swarm optimized-SVM(PSO-SVM)neural networks,and the traditional single-model LSTM,the SSA-LSTM-Multi-Head Attention model demonstrated superior generalization capability and prediction accuracy.The findings of this study not only advance the application of deep learning in coal dust explosion prediction but also provide robust technical support for the prevention and risk assessment of coal dust explosions.
基金supported by Xiamen Medical and Health Guidance Project in 2021(No.3502Z20214ZD1070)supported by a grant from Guangxi Key Laboratory of Machine Vision and Intelligent Control,China(No.2023B02).
文摘The self-attention mechanism of Transformers,which captures long-range contextual information,has demonstrated significant potential in image segmentation.However,their ability to learn local,contextual relationships between pixels requires further improvement.Previous methods face challenges in efficiently managing multi-scale fea-tures of different granularities from the encoder backbone,leaving room for improvement in their global representation and feature extraction capabilities.To address these challenges,we propose a novel Decoder with Multi-Head Feature Receptors(DMHFR),which receives multi-scale features from the encoder backbone and organizes them into three feature groups with different granularities:coarse,fine-grained,and full set.These groups are subsequently processed by Multi-Head Feature Receptors(MHFRs)after feature capture and modeling operations.MHFRs include two Three-Head Feature Receptors(THFRs)and one Four-Head Feature Receptor(FHFR).Each group of features is passed through these MHFRs and then fed into axial transformers,which help the model capture long-range dependencies within the features.The three MHFRs produce three distinct feature outputs.The output from the FHFR serves as auxiliary auxiliary features in the prediction head,and the prediction output and their losses will eventually be aggregated.Experimental results show that the Transformer using DMHFR outperforms 15 state of the arts(SOTA)methods on five public datasets.Specifically,it achieved significant improvements in mean DICE scores over the classic Parallel Reverse Attention Network(PraNet)method,with gains of 4.1%,2.2%,1.4%,8.9%,and 16.3%on the CVC-ClinicDB,Kvasir-SEG,CVC-T,CVC-ColonDB,and ETIS-LaribPolypDB datasets,respectively.
基金Supported by the Key R&D Program of Gansu Province(No.23YFGA0063)the Key Talent Project of Gansu Province(No.2024RCXM57,2024RCXM22)the Major Science and Technology Special Program of Gansu Province(No.25ZYJA037).
文摘Traffic flow prediction is a crucial element of intelligent transportation systems.However,accu-rate traffic flow prediction is quite challenging because of its highly nonlinear,complex,and dynam-ic characteristics.To address the difficulties in simultaneously capturing local and global dynamic spatiotemporal correlations in traffic flow,as well as the high time complexity of existing models,a multi-head flow attention-based local-global dynamic hypergraph convolution(MFA-LGDHC)pre-diction model is proposed.which consists of multi-head flow attention(MHFA)mechanism,graph convolution network(GCN),and local-global dynamic hypergraph convolution(LGHC).MHFA is utilized to extract the time dependency of traffic flow and reduce the time complexity of the model.GCN is employed to catch the spatial dependency of traffic flow.LGHC utilizes down-sampling con-volution and isometric convolution to capture the local and global spatial dependencies of traffic flow.And dynamic hypergraph convolution is used to model the dynamic higher-order relationships of the traffic road network.Experimental results indicate that the MFA-LGDHC model outperforms current popular baseline models and exhibits good prediction performance.
基金supported by the Equipment Test and Evaluation Technology Research Project,China(No.2100070017)the Natural Science Foundation of Gansu Province,China(No.24JRRA415)。
文摘Precise and quantitative measurement of soot particle emission plays an essential role in accurately assessing the combustion performance of aero-engine combustors and infrared signature levels in aircraft exhausts.Among various intrusive or non-intrusive approaches for soot diagnostics,Laser-Induced Incandescence(LⅡ)technique has been increasingly applied for soot concentration measurement in various combustion environments such as laminar flames and internal combustion engines due to its high spatial resolution and sensitivity.As for LⅡmeasurement in aero-engine combustors,however,it normally suffers from very limited optical accesses and often faces mandatory requirements of oblique imaging from a small backward angle.In this work,we demonstrate a Two-Color(2C)LⅡsystem that simultaneously captures LⅡsignal images at two distinct wavelengths using a Scheimpflug imaging configuration.A projective transformation algorithm and image overlapping procedures were employed to spatially correct the raw Scheimpflug LⅡimages.Performance validation of the developed 2C-Scheimpflug LⅡsystem was first conducted under specified conditions in a laminar C_(2)H_(4)/air McKenna flame.The obtained Soot Volume Fraction(SVF)level and its spatial distribution are in consistent with previous studies under identical flame conditions reported by other research groups.Finally,as a demonstration of engineering benchmark application,we applied the developed 2C-Scheimpflug LⅡsystem to measure SVF distribution in the cross-section plane perpendicular to the direction of flame propagation at the exhaust of a single-sector dual-swirl aero-engine model combustor.Transient soot production events were observed and characteristics of the SVF distribution were investigated.These experimental results suggest the feasibility of the 2C-Scheimpflug LⅡtechnique developed in this work for precise and quantitative measurements of soot concentration in practical environments.
基金supported by the Program of Key Laboratory of Cross-Domain Flight Interdisciplinary Technology,China(No.2023-ZY0205)。
文摘This paper describes an experimental study investigating the effects of sinusoidal pulsed injection on the combustion mode transition in a dual-mode supersonic combustor.The results are obtained under inflow conditions of 2.9 MPa stagnation pressure,1900 K stagnation temperature,and Mach number of 3.0.It has been observed that,at the same equivalence ratio,the combustion mode and flow field structure undergo irreversible changes from a weak combustion state to a strong combustion state at a specific pulsed jet frequency compared to steady jet.For steady jet,the combustion mode is dual-mode.As the frequency of the unsteady jet changes,the combustion mode also changes:it becomes a transition mode at frequencies of 171 Hz and 260 Hz,and a ramjet mode at 216 Hz.Combustion instability under steady jet manifests as a transition in flame stabilization mode.In contrast,under pulsed jet,combustion instability appears either as a transition in flame stabilization mode or as flame blow-off and flashback.The flow field oscillation frequency in the non-reacting flow is 171 Hz,which may resonate with the 171 Hz pulsed jet frequency,making the combustion oscillations most pronounced at this frequency.When the jet frequency is increased to 216 Hz,the combustion intensity significantly increases,and the combustion mode transfers to the ramjet mode.However,further increasing the frequency to 260 Hz results in a decrease in combustion intensity,returning to the transition mode.The frequency of the flow field oscillations varies with the coupling of the pulsed injection frequency,shock wave,and flame,and if the system reaches an unstable state,that is,pre-combustion shock train moves far upstream of the isolator during the pulsed jet period,strong combustion state can be achieved,and this process is irreversible.
基金supported by the National Key Research and Development Program of China(No.2020YFA0405700).
文摘Quantitative measurement of Soot Volume Fraction (SVF) is an essential prerequisite for controlling soot particle emissions from aero-engine combustors. As an in-situ and non-intrusive optical diagnostic technique, Laser-Induced Incandescence (LII) has been increasingly applied for soot concentration quantification in various combustion environments such as laminar flame, vehicle exhaust, internal combustion chamber as well as aero-engine combustor. In this work, we experimentally measured the spatial and temporal distribution of SVF using two-color LII technique at the outlet of a single-sector dual-swirl aero-engine model combustor. The effect of inlet pressure and air preheat temperature on the SVF distribution was separately investigated within a pressure range of 241–425 kPa and a temperature range of 292–500 K. The results show that soot production increases with the inlet pressure but generally decreases with the air preheat temperature. Qualitative analysis was provided to explain the above results of parametric studies. The LII experiments were also conducted under 3 designed conditions to evaluate soot emission under practical operations. Particularly, weak soot emission was detected at the outlet under the idle condition. Our experimental results provide a valuable benchmark for evaluating soot emission in the exhaust plume of this aero-engine combustor during practical operations.
文摘Brain tumor identification is a challenging task in neuro-oncology.The brain’s complex anatomy makes it a crucial part of the central nervous system.Accurate tumor classification is crucial for clinical diagnosis and treatment planning.This research presents a significant advancement in the multi-classification of brain tumors.This paper proposed a novel architecture that integrates Enhanced ResNeXt 101_32×8d,a Convolutional Neural Network(CNN)with a multi-head self-attention(MHSA)mechanism.This combination harnesses the strengths of the feature extraction,feature representation by CNN,and long-range dependencies by MHSA.Magnetic Resonance Imaging(MRI)datasets were employed to check the effectiveness of the proposed architecture.The first dataset(DS-1,Msoud)included four brain tumor classes,and the second dataset(DS-2)contained seven brain tumor classes.This methodology effectively distinguished various tumor classes,achieving high accuracies of 99.75% on DS-1 and 98.80% on DS-2.These impressive results indicate the superior performance and adaptability of our model for multiclass brain tumor classification.Evaluationmetrics such as accuracy,precision,recall,F1 score,and ROC(receiver operating characteristic)curve were utilized to comprehensively evaluate model validity.The performance results showed that the model is well-suited for clinical applications,with reduced errors and high accuracy.
基金support of the National Major Science and Technology Projects of China(No.J2019-V-0005-0096)the National Key Research and Development Program of China(No.2020YFA0405700).
文摘To provide advanced diagnostic techniques for diagnosing the outlet temperature distribution and species concentrations of future advanced combustors,this study focuses on a dual-swirl single-dome rectangular combustor.Through the integration of multiple diagnostics,simultaneous measurement of outlet temperature distribution and species concentrations was achieved.The study validates the engineering applicability of these simultaneous measurements using tungsten-rhenium(W-Re)thermocouples and Coherent Anti-Stokes Raman Scattering(CARS),CARS and Tunable Diode Laser Absorption Spectroscopy(TDLAS),as well as Gas Analysis(GA)and Mass Spectrometry(MS).The results demonstrate that measurements by thermocouples and CARS exhibit good consistency and repeatability,with a relative deviation of less than 4%,fully meeting the requirements of engineering experiments.The spatial distribution reconstruction results of TDLAS can reflect the temperature distribution characteristics at the combustor outlet.Temperature comparison between TDLAS and CARS at single-point positions shows consistent results,with a relative deviation of less than 11%and 7%under both conditions,respectively.Simultaneous measurements by integrating GA and MS show high engineering applicability for the first time,meeting the requirements for measuring both inorganic species and free radicals at the combustor outlet.Under C_(1)condition,the relative deviations of four key species(Unburned Hydrocarbon(UHC),NO,O_(2),and CO_(2))remain within 2%,while that of NO_(2)is slightly higher at approximately 8%.Under C_(2)condition,the overall deviations increase for most species,with only O_(2)and CO_(2)maintaining relatively low deviations.The primary species of UHCs at the combustor outlet under both conditions are small molecular hydrocarbons(C_(3)-C_(8))and RO_(2)radicals,accounting for over 90%of total UHC.Specifically,RO_(2)species(R is C_(1)-C_(2)alkyl groups)are the predominant species,accounting for 74.3%and 82.1%of total RO_(2)under both conditions,respectively.These integrated diagnostic methods for temperature and species concentrations at the combustor outlet serve as a crucial reference for its engineering applications.
