Pedestrian trajectory prediction is pivotal and challenging in applications such as autonomous driving,social robotics,and intelligent surveillance systems.Pedestrian trajectory is governed not only by individual inte...Pedestrian trajectory prediction is pivotal and challenging in applications such as autonomous driving,social robotics,and intelligent surveillance systems.Pedestrian trajectory is governed not only by individual intent but also by interactions with surrounding agents.These interactions are critical to trajectory prediction accuracy.While prior studies have employed Convolutional Neural Networks(CNNs)and Graph Convolutional Networks(GCNs)to model such interactions,these methods fail to distinguish varying influence levels among neighboring pedestrians.To address this,we propose a novel model based on a bidirectional graph attention network and spatio-temporal graphs to capture dynamic interactions.Specifically,we construct temporal and spatial graphs encoding the sequential evolution and spatial proximity among pedestrians.These features are then fused and processed by the Bidirectional Graph Attention Network(Bi-GAT),which models the bidirectional interactions between the target pedestrian and its neighbors.The model computes node attention weights(i.e.,similarity scores)to differentially aggregate neighbor information,enabling fine-grained interaction representations.Extensive experiments conducted on two widely used pedestrian trajectory prediction benchmark datasets demonstrate that our approach outperforms existing state-of-theartmethods regarding Average Displacement Error(ADE)and Final Displacement Error(FDE),highlighting its strong prediction accuracy and generalization capability.展开更多
The increasingly severe state of coal burst disaster has emerged as a critical factor constraining coal mine safety production,and it has become a challenging task to enhance the accuracy of coal burst disaster predic...The increasingly severe state of coal burst disaster has emerged as a critical factor constraining coal mine safety production,and it has become a challenging task to enhance the accuracy of coal burst disaster prediction.To address the issue of insufficient exploration of the spatio-temporal characteristic of microseismic data and the challenging selection of the optimal time window size in spatio-temporal prediction,this paper integrates deep learning methods and theory to propose a novel coal burst spatio-temporal prediction method based on Bidirectional Long Short-Term Memory(Bi-LSTM)network.The method involves three main modules,including microseismic spatio-temporal characteristic indicators construction,temporal prediction model,and spatial prediction model.To validate the effectiveness of the proposed method,engineering application tests are conducted at a high-risk working face in the Ordos mining area of Inner Mongolia,focusing on 13 high-energy microseismic events with energy levels greater than 105 J.In terms of temporal prediction,the analysis indicates that the temporal prediction results consist of 10 strong predictions and 3 medium predictions,and there is no false alarm detected throughout the entire testing period.Moreover,compared to the traditional threshold-based coal burst temporal prediction method,the accuracy of the proposed method is increased by 38.5%.In terms of spatial prediction,the distribution of spatial prediction results for high-energy events comprises 6 strong hazard predictions,3 medium hazard predictions,and 4 weak hazard predictions.展开更多
The remaining useful life prediction of rolling bearing is vital in safety and reliability guarantee.In engineering scenarios,only a small amount of bearing performance degradation data can be obtained through acceler...The remaining useful life prediction of rolling bearing is vital in safety and reliability guarantee.In engineering scenarios,only a small amount of bearing performance degradation data can be obtained through accelerated life testing.In the absence of lifetime data,the hidden long-term correlation between performance degradation data is challenging to mine effectively,which is the main factor that restricts the prediction precision and engineering application of the residual life prediction method.To address this problem,a novel method based on the multi-layer perception neural network and bidirectional long short-term memory network is proposed.Firstly,a nonlinear health indicator(HI)calculation method based on kernel principal component analysis(KPCA)and exponential weighted moving average(EWMA)is designed.Then,using the raw vibration data and HI,a multi-layer perceptron(MLP)neural network is trained to further calculate the HI of the online bearing in real time.Furthermore,The bidirectional long short-term memory model(BiLSTM)optimized by particle swarm optimization(PSO)is used to mine the time series features of HI and predict the remaining service life.Performance verification experiments and comparative experiments are carried out on the XJTU-SY bearing open dataset.The research results indicate that this method has an excellent ability to predict future HI and remaining life.展开更多
Recognizing human interactions in RGB videos is a critical task in computer vision,with applications in video surveillance.Existing deep learning-based architectures have achieved strong results,but are computationall...Recognizing human interactions in RGB videos is a critical task in computer vision,with applications in video surveillance.Existing deep learning-based architectures have achieved strong results,but are computationally intensive,sensitive to video resolution changes and often fail in crowded scenes.We propose a novel hybrid system that is computationally efficient,robust to degraded video quality and able to filter out irrelevant individuals,making it suitable for real-life use.The system leverages multi-modal handcrafted features for interaction representation and a deep learning classifier for capturing complex dependencies.Using Mask R-CNN and YOLO11-Pose,we extract grayscale silhouettes and keypoint coordinates of interacting individuals,while filtering out irrelevant individuals using a proposed algorithm.From these,we extract silhouette-based features(local ternary pattern and histogram of optical flow)and keypoint-based features(distances,angles and velocities)that capture distinct spatial and temporal information.A Bidirectional Long Short-Term Memory network(BiLSTM)then classifies the interactions.Extensive experiments on the UT Interaction,SBU Kinect Interaction and the ISR-UOL 3D social activity datasets demonstrate that our system achieves competitive accuracy.They also validate the effectiveness of the chosen features and classifier,along with the proposed system’s computational efficiency and robustness to occlusion.展开更多
There are two technical challenges in predicting slope deformation.The first one is the random displacement,which could not be decomposed and predicted by numerically resolving the observed accumulated displacement an...There are two technical challenges in predicting slope deformation.The first one is the random displacement,which could not be decomposed and predicted by numerically resolving the observed accumulated displacement and time series of a landslide.The second one is the dynamic evolution of a landslide,which could not be feasibly simulated simply by traditional prediction models.In this paper,a dynamic model of displacement prediction is introduced for composite landslides based on a combination of empirical mode decomposition with soft screening stop criteria(SSSC-EMD)and deep bidirectional long short-term memory(DBi-LSTM)neural network.In the proposed model,the time series analysis and SSSC-EMD are used to decompose the observed accumulated displacements of a slope into three components,viz.trend displacement,periodic displacement,and random displacement.Then,by analyzing the evolution pattern of a landslide and its key factors triggering landslides,appropriate influencing factors are selected for each displacement component,and DBi-LSTM neural network to carry out multi-datadriven dynamic prediction for each displacement component.An accumulated displacement prediction has been obtained by a summation of each component.For accuracy verification and engineering practicability of the model,field observations from two known landslides in China,the Xintan landslide and the Bazimen landslide were collected for comparison and evaluation.The case study verified that the model proposed in this paper can better characterize the"stepwise"deformation characteristics of a slope.As compared with long short-term memory(LSTM)neural network,support vector machine(SVM),and autoregressive integrated moving average(ARIMA)model,DBi-LSTM neural network has higher accuracy in predicting the periodic displacement of slope deformation,with the mean absolute percentage error reduced by 3.063%,14.913%,and 13.960%respectively,and the root mean square error reduced by 1.951 mm,8.954 mm and 7.790 mm respectively.Conclusively,this model not only has high prediction accuracy but also is more stable,which can provide new insight for practical landslide prevention and control engineering.展开更多
Estimation of construction parameters is crucial for optimizing tunnel construction schedule.Due to the influence of routine activities and occasional risk events,these parameters are usually correlated and imbalanced...Estimation of construction parameters is crucial for optimizing tunnel construction schedule.Due to the influence of routine activities and occasional risk events,these parameters are usually correlated and imbalanced.To solve this issue,an improved bidirectional generative adversarial network(BiGAN)model with a joint discriminator structure and zero-centered gradient penalty(0-GP)is proposed.In this model,in order to improve the capability of original BiGAN in learning imbalanced parameters,the joint discriminator separately discriminates the routine activities and risk event durations to balance their influence weights.Then,the self-attention mechanism is embedded so that the discriminator can pay more attention to the imbalanced parameters.Finally,the 0-GP is adapted for the loss of the discrimi-nator to improve its convergence and stability.A case study of a tunnel in China shows that the improved BiGAN can obtain parameter estimates consistent with the classical Gauss mixture model,without the need of tedious and complex correlation analysis.The proposed joint discriminator can increase the ability of BiGAN in estimating imbalanced construction parameters,and the 0-GP can ensure the stability and convergence of the model.展开更多
Unsupervised methods based on density representation have shown their abilities in anomaly detection,but detection performance still needs to be improved.Specifically,approaches using normalizing flows can accurately ...Unsupervised methods based on density representation have shown their abilities in anomaly detection,but detection performance still needs to be improved.Specifically,approaches using normalizing flows can accurately evaluate sample distributions,mapping normal features to the normal distribution and anomalous features outside it.Consequently,this paper proposes a Normalizing Flow-based Bidirectional Mapping Residual Network(NF-BMR).It utilizes pre-trained Convolutional Neural Networks(CNN)and normalizing flows to construct discriminative source and target domain feature spaces.Additionally,to better learn feature information in both domain spaces,we propose the Bidirectional Mapping Residual Network(BMR),which maps sample features to these two spaces for anomaly detection.The two detection spaces effectively complement each other’s deficiencies and provide a comprehensive feature evaluation from two perspectives,which leads to the improvement of detection performance.Comparative experimental results on the MVTec AD and DAGM datasets against the Bidirectional Pre-trained Feature Mapping Network(B-PFM)and other state-of-the-art methods demonstrate that the proposed approach achieves superior performance.On the MVTec AD dataset,NF-BMR achieves an average AUROC of 98.7%for all 15 categories.Especially,it achieves 100%optimal detection performance in five categories.On the DAGM dataset,the average AUROC across ten categories is 98.7%,which is very close to supervised methods.展开更多
In this paper,the recurrent neural network structure of a bidirectional long shortterm memory network(Bi-LSTM)with special memory cells that store information is used to characterize the deep features of the variation...In this paper,the recurrent neural network structure of a bidirectional long shortterm memory network(Bi-LSTM)with special memory cells that store information is used to characterize the deep features of the variation pattern between logging and seismic data.A mapping relationship model between high-frequency logging data and low-frequency seismic data is established via nonlinear mapping.The seismic waveform is infinitely approximated using the logging curve in the low-frequency band to obtain a nonlinear mapping model of this scale,which then stepwise approach the logging curve in the high-frequency band.Finally,a seismic-inversion method of nonlinear mapping multilevel well–seismic matching based on the Bi-LSTM network is developed.The characteristic of this method is that by applying the multilevel well–seismic matching process,the seismic data are stepwise matched to the scale range that is consistent with the logging curve.Further,the matching operator at each level can be stably obtained to effectively overcome the problems that occur in the well–seismic matching process,such as the inconsistency in the scale of two types of data,accuracy in extracting the seismic wavelet of the well-side seismic traces,and multiplicity of solutions.Model test and practical application demonstrate that this method improves the vertical resolution of inversion results,and at the same time,the boundary and the lateral characteristics of the sand body are well maintained to improve the accuracy of thin-layer sand body prediction and achieve an improved practical application effect.展开更多
The aircraft braking system is critical to ensure the safe take-off and landing of the aircraft.However,the braking system is often exposed to high temperatures and strong vibration working environments,which makes th...The aircraft braking system is critical to ensure the safe take-off and landing of the aircraft.However,the braking system is often exposed to high temperatures and strong vibration working environments,which makes the sensor prone to failure.Sensor failure has the potential to compromise aircraft safety.In order to improve the safety of the aircraft braking system,a fault detection and fault-tolerant control(FDFTC)strategy for the aircraft brake pressure sensor is designed.Firstly,a model based on a bidirectional long short-term memory(Bi-LSTM)network is constructed to estimate the brake pressure.Then,the residual sequence is obtained by comparing the measured pressure with the estimated pressure.On this basis,the improved sequential probability ratio test(SPRT)method based on mathematical statistics is applied to analyze the residual sequence to detect the fault.Finally,simulation and hardware-in-the-loop(HIL)testing results indicate that the proposed FDFTC strategy can detect sensor faults in time and efficiently complete braking when faults occur.Hence,the proposed FDFTC strategy can effectively deal with the faults of the aircraft brake pressure sensor,which is of great significance to improve the reliability and safety of the aircraft.展开更多
Accurate and reliable power system data are fundamental for critical operations such as gridmonitoring,fault diagnosis,and load forecasting,underpinned by increasing intelligentization and digitalization.However,data ...Accurate and reliable power system data are fundamental for critical operations such as gridmonitoring,fault diagnosis,and load forecasting,underpinned by increasing intelligentization and digitalization.However,data loss and anomalies frequently compromise data integrity in practical settings,significantly impacting system operational efficiency and security.Most existing data recovery methods require complete datasets for training,leading to substantial data and computational demands and limited generalization.To address these limitations,this study proposes a missing data imputation model based on an improved Generative Adversarial Network(BAC-GAN).Within the BAC-GAN framework,the generator utilizes Bidirectional Long Short-Term Memory(BiLSTM)networks and Multi-Head Attention mechanisms to capture temporal dependencies and complex relationships within power system data.The discriminator employs a Convolutional Neural Network(CNN)architecture to integrate local features with global structures,effectivelymitigating the generation of implausible imputations.Experimental results on two public datasets demonstrate that the BAC-GAN model achieves superior data recovery accuracy compared to five state-of-the-art and classical benchmarkmethods,with an average improvement of 17.7%in reconstruction accuracy.The proposedmethod significantly enhances the accuracy of grid fault diagnosis and provides reliable data support for the stable operation of smart grids,showing great potential for practical applications in power systems.展开更多
Accurate prediction of remaining useful life serves as a reliable basis for maintenance strategies,effectively reducing both the frequency of failures and associated costs.As a core component of PHM,RUL prediction pla...Accurate prediction of remaining useful life serves as a reliable basis for maintenance strategies,effectively reducing both the frequency of failures and associated costs.As a core component of PHM,RUL prediction plays a crucial role in preventing equipment failures and optimizing maintenance decision-making.However,deep learning models often falter when processing raw,noisy temporal signals,fail to quantify prediction uncertainty,and face challenges in effectively capturing the nonlinear dynamics of equipment degradation.To address these issues,this study proposes a novel deep learning framework.First,a newbidirectional long short-termmemory network integrated with an attention mechanism is designed to enhance temporal feature extraction with improved noise robustness.Second,a probabilistic prediction framework based on kernel density estimation is constructed,incorporating residual connections and stochastic regularization to achieve precise RUL estimation.Finally,extensive experiments on the C-MAPSS dataset demonstrate that our method achieves competitive performance in terms of RMSE and Score metrics compared to state-of-the-artmodels.More importantly,the probabilistic output provides a quantifiablemeasure of prediction confidence,which is crucial for risk-informed maintenance planning,enabling managers to optimize maintenance strategies based on a quantifiable understanding of failure risk.展开更多
The battlefield environment is changing rapidly,and fast and accurate identification of the tactical intention of enemy targets is an important condition for gaining a decision-making advantage.The current Intention R...The battlefield environment is changing rapidly,and fast and accurate identification of the tactical intention of enemy targets is an important condition for gaining a decision-making advantage.The current Intention Recognition(IR)method for air targets has shortcomings in temporality,interpretability and back-and-forth dependency of intentions.To address these problems,this paper designs a novel air target intention recognition method named STABC-IR,which is based on Bidirectional Gated Recurrent Unit(Bi GRU)and Conditional Random Field(CRF)with Space-Time Attention mechanism(STA).First,the problem of intention recognition of air targets is described and analyzed in detail.Then,a temporal network based on Bi GRU is constructed to achieve the temporal requirement.Subsequently,STA is proposed to focus on the key parts of the features and timing information to meet certain interpretability requirements while strengthening the timing requirements.Finally,an intention transformation network based on CRF is proposed to solve the back-and-forth dependency and transformation problem by jointly modeling the tactical intention of the target at each moment.The experimental results show that the recognition accuracy of the jointly trained STABC-IR model can reach 95.7%,which is higher than other latest intention recognition methods.STABC-IR solves the problem of intention transformation for the first time and considers both temporality and interpretability,which is important for improving the tactical intention recognition capability and has reference value for the construction of command and control auxiliary decision-making system.展开更多
Studies on the stability of the equilibrium points of continuous bidirectional associative memory (BAM) neural network have yielded many useful results. A novel neural network model called standard neural network mode...Studies on the stability of the equilibrium points of continuous bidirectional associative memory (BAM) neural network have yielded many useful results. A novel neural network model called standard neural network model (SNNM) is ad- vanced. By using state affine transformation, the BAM neural networks were converted to SNNMs. Some sufficient conditions for the global asymptotic stability of continuous BAM neural networks were derived from studies on the SNNMs’ stability. These conditions were formulated as easily verifiable linear matrix inequalities (LMIs), whose conservativeness is relatively low. The approach proposed extends the known stability results, and can also be applied to other forms of recurrent neural networks (RNNs).展开更多
Several novel stability conditions for BAM neural networks with time-varying delays are studied.Based on Lyapunov-Krasovskii functional combined with linear matrix inequality approach,the delay-dependent linear matrix...Several novel stability conditions for BAM neural networks with time-varying delays are studied.Based on Lyapunov-Krasovskii functional combined with linear matrix inequality approach,the delay-dependent linear matrix inequality（LMI） conditions are established to guarantee robust asymptotic stability for given delayed BAM neural networks.These criteria can be easily verified by utilizing the recently developed algorithms for solving LMIs.A numerical example is provided to demonstrate the effectiveness and less conservatism of the main results.展开更多
Responding to the stochasticity and uncertainty in the power height of distributed photovoltaic power generation.This paper presents a distributed photovoltaic ultra-short-term power forecasting method based on Variat...Responding to the stochasticity and uncertainty in the power height of distributed photovoltaic power generation.This paper presents a distributed photovoltaic ultra-short-term power forecasting method based on Variational Mode Decomposition(VMD)and Channel Attention Mechanism.First,Pearson’s correlation coefficient was utilized to filter out the meteorological factors that had a high impact on historical power.Second,the distributed PV power data were decomposed into a relatively smooth power series with different fluctuation patterns using variational modal decomposition(VMD).Finally,the reconstructed distributed PV power as well as other features are input into the combined CNN-SENet-BiLSTM model.In this model,the convolutional neural network(CNN)and channel attention mechanism dynamically adjust the weights while capturing the spatial features of the input data to improve the discriminative ability of key features.The extracted data is then fed into the bidirectional long short-term memory network(BiLSTM)to capture the time-series features,and the final output is the prediction result.The verification is conducted using a dataset from a distributed photovoltaic power station in the Northwest region of China.The results show that compared with other prediction methods,the method proposed in this paper has a higher prediction accuracy,which helps to improve the proportion of distributed PV access to the grid,and can guarantee the safe and stable operation of the power grid.展开更多
Traditional deep learning methods pursue complex and single network architectures without considering the petrophysical relationship between different elastic parameters.The mathematical and statistical significance o...Traditional deep learning methods pursue complex and single network architectures without considering the petrophysical relationship between different elastic parameters.The mathematical and statistical significance of the inversion results may lead to model overfitting,especially when there are a limited number of well logs in a working area.Multitask learning provides an eff ective approach to addressing this issue.Simultaneously,learning multiple related tasks can improve a model’s generalization ability to a certain extent,thereby enhancing the performance of related tasks with an equal amount of labeled data.In this study,we propose an end-to-end multitask deep learning model that integrates a fully convolutional network and bidirectional gated recurrent unit for intelligent prestack inversion of“seismic data to elastic parameters.”The use of a Bayesian homoscedastic uncertainty-based loss function enables adaptive learning of the weight coeffi cients for diff erent elastic parameter inversion tasks,thereby reducing uncertainty during the inversion process.The proposed method combines the local feature perception of convolutional neural networks with the long-term memory of bidirectional gated recurrent networks.It maintains the rock physics constraint relationships among diff erent elastic parameters during the inversion process,demonstrating a high level of prediction accuracy.Numerical simulations and processing results of real seismic data validate the eff ectiveness and practicality of the proposed method.展开更多
Throughout the composite structure’s lifespan,it is subject to a range of environmental factors,including loads,vibrations,and conditions involving heat and humidity.These factors have the potential to compromise the...Throughout the composite structure’s lifespan,it is subject to a range of environmental factors,including loads,vibrations,and conditions involving heat and humidity.These factors have the potential to compromise the integrity of the structure.The estimation of the fatigue life of composite materials is imperative for ensuring the structural integrity of these materials.In this study,a methodology is proposed for predicting the fatigue life of composites that integrates ultrasonic guided waves and machine learning modeling.The method first screens the ultrasonic guided wave signal features that are significantly affected by fatigue damage.Subsequently,a covariance analysis is conducted to reduce the redundancy of the feature matrix.Furthermore,one-hot encoding is employed to incorporate boundary conditions as features,and the resulting data undergoes preprocessing to form a sample library.A composite fatigue life prediction model has been developed,employing the aforementioned sample library as the input source and utilizing remaining life as the output metric.The model synthesizes the strengths of convolutional neural networks(CNNs)and bidirectional long short-term memory networks(BiLSTMs)while leveraging Bayesian optimization(BO)to enhance the optimization of hyperparameters.The experimental results demonstrate that the proposed BO-CNN-BiLSTM model exhibits superior performance in terms of prediction accuracy and reliability in the damage regression task when compared to both the BiLSTM and CNN-BiLSTM models.展开更多
To address the limitations of existing abnormal traffic detection methods,such as insufficient temporal and spatial feature extraction,high false positive rate(FPR),poor generalization,and class imbalance,this study p...To address the limitations of existing abnormal traffic detection methods,such as insufficient temporal and spatial feature extraction,high false positive rate(FPR),poor generalization,and class imbalance,this study proposed an intelligent detection method that combines a Stacked Convolutional Network(SCN),Bidirectional Long Short-Term Memory(BiLSTM)network,and Equalization Loss v2(EQL v2).This method was divided into two components:a feature extraction model and a classification and detection model.First,SCN was constructed by combining a Convolutional Neural Network(CNN)with a Depthwise Separable Convolution(DSC)network to capture the abstract spatial features of traffic data.These features were then input into the BiLSTM to capture temporal dependencies.An attention mechanism was incorporated after SCN and BiLSTM to enhance the extraction of key spatiotemporal features.To address class imbalance,the classification detection model applied EQL v2 to adjust the weights of the minority classes,ensuring that they received equal focus during training.The experimental results indicated that the proposed method outperformed the existing methods in terms of accuracy,FPR,and F1-score and significantly improved the identification rate of minority classes.展开更多
Electronic Medical Records(EMR) with unstructured sentences and various conceptual expressions provide rich information for medical information extraction. However, common Named Entity Recognition(NER)in Natural Langu...Electronic Medical Records(EMR) with unstructured sentences and various conceptual expressions provide rich information for medical information extraction. However, common Named Entity Recognition(NER)in Natural Language Processing(NLP) are not well suitable for clinical NER in EMR. This study aims at applying neural networks to clinical concept extractions. We integrate Bidirectional Long Short-Term Memory Networks(Bi-LSTM) with a Conditional Random Fields(CRF) layer to detect three types of clinical named entities. Word representations fed into the neural networks are concatenated by character-based word embeddings and Continuous Bag of Words(CBOW) embeddings trained both on domain and non-domain corpus. We test our NER system on i2b2/VA open datasets and compare the performance with six related works, achieving the best result of NER with F1 value 0.853 7. We also point out a few specific problems in clinical concept extractions which will give some hints to deeper studies.展开更多
Metasurfaces,which are the two-dimensional counterparts of metamaterials,have demonstrated unprecedented capabilities to manipulate the wavefront of electromagnetic waves in a single flat device.Despite various advanc...Metasurfaces,which are the two-dimensional counterparts of metamaterials,have demonstrated unprecedented capabilities to manipulate the wavefront of electromagnetic waves in a single flat device.Despite various advances in this field,the unique functionalities achieved by metasurfaces have come at the cost of the structural complexity,resulting in a time-consuming parameter sweep for the conventional metasurface design.Although artificial neural networks provide a flexible platform for significantly improving the design process,the current metasurface designs are restricted to generating qualitative field distributions.In this study,we demonstrate that by combining a tandem neural network and an iterative algorithm,the previous restriction of the design of metasurfaces can be overcome with quantitative field distributions.As proof-of-principle examples,metalenses predicted via the designed network architecture that possess multiple focal points with identical/orthogonal polarisation states,as well as accurate intensity ratios(quantitative field distributions),were numerically calculated and experimentally demonstrated.The unique and robust approach for the metasurface design will enable the acceleration of the development of devices with high-accuracy functionalities,which can be applied in imaging,detecting,and sensing.展开更多
基金funded by the National Natural Science Foundation of China,grant number 624010funded by the Natural Science Foundation of Anhui Province,grant number 2408085QF202+1 种基金funded by the Anhui Future Technology Research Institute Industry Guidance Fund Project,grant number 2023cyyd04funded by the Project of Research of Anhui Polytechnic University,grant number Xjky2022150.
文摘Pedestrian trajectory prediction is pivotal and challenging in applications such as autonomous driving,social robotics,and intelligent surveillance systems.Pedestrian trajectory is governed not only by individual intent but also by interactions with surrounding agents.These interactions are critical to trajectory prediction accuracy.While prior studies have employed Convolutional Neural Networks(CNNs)and Graph Convolutional Networks(GCNs)to model such interactions,these methods fail to distinguish varying influence levels among neighboring pedestrians.To address this,we propose a novel model based on a bidirectional graph attention network and spatio-temporal graphs to capture dynamic interactions.Specifically,we construct temporal and spatial graphs encoding the sequential evolution and spatial proximity among pedestrians.These features are then fused and processed by the Bidirectional Graph Attention Network(Bi-GAT),which models the bidirectional interactions between the target pedestrian and its neighbors.The model computes node attention weights(i.e.,similarity scores)to differentially aggregate neighbor information,enabling fine-grained interaction representations.Extensive experiments conducted on two widely used pedestrian trajectory prediction benchmark datasets demonstrate that our approach outperforms existing state-of-theartmethods regarding Average Displacement Error(ADE)and Final Displacement Error(FDE),highlighting its strong prediction accuracy and generalization capability.
基金supported by the National Research and Development Program(2022YFC3004603)the Jiangsu Province International Collaboration Program-Key National Industrial Technology Research and Development Cooperation Projects(BZ2023050)+1 种基金the Natural Science Foundation of Jiangsu Province(BK20221109)the National Natural Science Foundation of China(52274098).
文摘The increasingly severe state of coal burst disaster has emerged as a critical factor constraining coal mine safety production,and it has become a challenging task to enhance the accuracy of coal burst disaster prediction.To address the issue of insufficient exploration of the spatio-temporal characteristic of microseismic data and the challenging selection of the optimal time window size in spatio-temporal prediction,this paper integrates deep learning methods and theory to propose a novel coal burst spatio-temporal prediction method based on Bidirectional Long Short-Term Memory(Bi-LSTM)network.The method involves three main modules,including microseismic spatio-temporal characteristic indicators construction,temporal prediction model,and spatial prediction model.To validate the effectiveness of the proposed method,engineering application tests are conducted at a high-risk working face in the Ordos mining area of Inner Mongolia,focusing on 13 high-energy microseismic events with energy levels greater than 105 J.In terms of temporal prediction,the analysis indicates that the temporal prediction results consist of 10 strong predictions and 3 medium predictions,and there is no false alarm detected throughout the entire testing period.Moreover,compared to the traditional threshold-based coal burst temporal prediction method,the accuracy of the proposed method is increased by 38.5%.In terms of spatial prediction,the distribution of spatial prediction results for high-energy events comprises 6 strong hazard predictions,3 medium hazard predictions,and 4 weak hazard predictions.
基金supported by the National Key Research and Development Project(Grant Number 2023YFB3709601)the National Natural Science Foundation of China(Grant Numbers 62373215,62373219,62073193)+2 种基金the Key Research and Development Plan of Shandong Province(Grant Numbers 2021CXGC010204,2022CXGC020902)the Fundamental Research Funds of Shandong University(Grant Number 2021JCG008)the Natural Science Foundation of Shandong Province(Grant Number ZR2023MF100).
文摘The remaining useful life prediction of rolling bearing is vital in safety and reliability guarantee.In engineering scenarios,only a small amount of bearing performance degradation data can be obtained through accelerated life testing.In the absence of lifetime data,the hidden long-term correlation between performance degradation data is challenging to mine effectively,which is the main factor that restricts the prediction precision and engineering application of the residual life prediction method.To address this problem,a novel method based on the multi-layer perception neural network and bidirectional long short-term memory network is proposed.Firstly,a nonlinear health indicator(HI)calculation method based on kernel principal component analysis(KPCA)and exponential weighted moving average(EWMA)is designed.Then,using the raw vibration data and HI,a multi-layer perceptron(MLP)neural network is trained to further calculate the HI of the online bearing in real time.Furthermore,The bidirectional long short-term memory model(BiLSTM)optimized by particle swarm optimization(PSO)is used to mine the time series features of HI and predict the remaining service life.Performance verification experiments and comparative experiments are carried out on the XJTU-SY bearing open dataset.The research results indicate that this method has an excellent ability to predict future HI and remaining life.
基金supported and funded by Princess Nourah bint Abdulrahman University Researchers Supporting Project number(PNURSP2025R410),Princess Nourah bint Abdulrahman University,Riyadh,Saudi Arabia.
文摘Recognizing human interactions in RGB videos is a critical task in computer vision,with applications in video surveillance.Existing deep learning-based architectures have achieved strong results,but are computationally intensive,sensitive to video resolution changes and often fail in crowded scenes.We propose a novel hybrid system that is computationally efficient,robust to degraded video quality and able to filter out irrelevant individuals,making it suitable for real-life use.The system leverages multi-modal handcrafted features for interaction representation and a deep learning classifier for capturing complex dependencies.Using Mask R-CNN and YOLO11-Pose,we extract grayscale silhouettes and keypoint coordinates of interacting individuals,while filtering out irrelevant individuals using a proposed algorithm.From these,we extract silhouette-based features(local ternary pattern and histogram of optical flow)and keypoint-based features(distances,angles and velocities)that capture distinct spatial and temporal information.A Bidirectional Long Short-Term Memory network(BiLSTM)then classifies the interactions.Extensive experiments on the UT Interaction,SBU Kinect Interaction and the ISR-UOL 3D social activity datasets demonstrate that our system achieves competitive accuracy.They also validate the effectiveness of the chosen features and classifier,along with the proposed system’s computational efficiency and robustness to occlusion.
文摘There are two technical challenges in predicting slope deformation.The first one is the random displacement,which could not be decomposed and predicted by numerically resolving the observed accumulated displacement and time series of a landslide.The second one is the dynamic evolution of a landslide,which could not be feasibly simulated simply by traditional prediction models.In this paper,a dynamic model of displacement prediction is introduced for composite landslides based on a combination of empirical mode decomposition with soft screening stop criteria(SSSC-EMD)and deep bidirectional long short-term memory(DBi-LSTM)neural network.In the proposed model,the time series analysis and SSSC-EMD are used to decompose the observed accumulated displacements of a slope into three components,viz.trend displacement,periodic displacement,and random displacement.Then,by analyzing the evolution pattern of a landslide and its key factors triggering landslides,appropriate influencing factors are selected for each displacement component,and DBi-LSTM neural network to carry out multi-datadriven dynamic prediction for each displacement component.An accumulated displacement prediction has been obtained by a summation of each component.For accuracy verification and engineering practicability of the model,field observations from two known landslides in China,the Xintan landslide and the Bazimen landslide were collected for comparison and evaluation.The case study verified that the model proposed in this paper can better characterize the"stepwise"deformation characteristics of a slope.As compared with long short-term memory(LSTM)neural network,support vector machine(SVM),and autoregressive integrated moving average(ARIMA)model,DBi-LSTM neural network has higher accuracy in predicting the periodic displacement of slope deformation,with the mean absolute percentage error reduced by 3.063%,14.913%,and 13.960%respectively,and the root mean square error reduced by 1.951 mm,8.954 mm and 7.790 mm respectively.Conclusively,this model not only has high prediction accuracy but also is more stable,which can provide new insight for practical landslide prevention and control engineering.
基金supported by National Natural Science Foundation of China(Grant Nos.52279137,52009090).
文摘Estimation of construction parameters is crucial for optimizing tunnel construction schedule.Due to the influence of routine activities and occasional risk events,these parameters are usually correlated and imbalanced.To solve this issue,an improved bidirectional generative adversarial network(BiGAN)model with a joint discriminator structure and zero-centered gradient penalty(0-GP)is proposed.In this model,in order to improve the capability of original BiGAN in learning imbalanced parameters,the joint discriminator separately discriminates the routine activities and risk event durations to balance their influence weights.Then,the self-attention mechanism is embedded so that the discriminator can pay more attention to the imbalanced parameters.Finally,the 0-GP is adapted for the loss of the discrimi-nator to improve its convergence and stability.A case study of a tunnel in China shows that the improved BiGAN can obtain parameter estimates consistent with the classical Gauss mixture model,without the need of tedious and complex correlation analysis.The proposed joint discriminator can increase the ability of BiGAN in estimating imbalanced construction parameters,and the 0-GP can ensure the stability and convergence of the model.
基金This work was supported in part by the National Key R&D Program of China 2021YFE0110500in part by the National Natural Science Foundation of China under Grant 62062021in part by the Guiyang Scientific Plan Project[2023]48-11.
文摘Unsupervised methods based on density representation have shown their abilities in anomaly detection,but detection performance still needs to be improved.Specifically,approaches using normalizing flows can accurately evaluate sample distributions,mapping normal features to the normal distribution and anomalous features outside it.Consequently,this paper proposes a Normalizing Flow-based Bidirectional Mapping Residual Network(NF-BMR).It utilizes pre-trained Convolutional Neural Networks(CNN)and normalizing flows to construct discriminative source and target domain feature spaces.Additionally,to better learn feature information in both domain spaces,we propose the Bidirectional Mapping Residual Network(BMR),which maps sample features to these two spaces for anomaly detection.The two detection spaces effectively complement each other’s deficiencies and provide a comprehensive feature evaluation from two perspectives,which leads to the improvement of detection performance.Comparative experimental results on the MVTec AD and DAGM datasets against the Bidirectional Pre-trained Feature Mapping Network(B-PFM)and other state-of-the-art methods demonstrate that the proposed approach achieves superior performance.On the MVTec AD dataset,NF-BMR achieves an average AUROC of 98.7%for all 15 categories.Especially,it achieves 100%optimal detection performance in five categories.On the DAGM dataset,the average AUROC across ten categories is 98.7%,which is very close to supervised methods.
基金supported by the National Major Science and Technology Special Project(No.2016ZX05026-002).
文摘In this paper,the recurrent neural network structure of a bidirectional long shortterm memory network(Bi-LSTM)with special memory cells that store information is used to characterize the deep features of the variation pattern between logging and seismic data.A mapping relationship model between high-frequency logging data and low-frequency seismic data is established via nonlinear mapping.The seismic waveform is infinitely approximated using the logging curve in the low-frequency band to obtain a nonlinear mapping model of this scale,which then stepwise approach the logging curve in the high-frequency band.Finally,a seismic-inversion method of nonlinear mapping multilevel well–seismic matching based on the Bi-LSTM network is developed.The characteristic of this method is that by applying the multilevel well–seismic matching process,the seismic data are stepwise matched to the scale range that is consistent with the logging curve.Further,the matching operator at each level can be stably obtained to effectively overcome the problems that occur in the well–seismic matching process,such as the inconsistency in the scale of two types of data,accuracy in extracting the seismic wavelet of the well-side seismic traces,and multiplicity of solutions.Model test and practical application demonstrate that this method improves the vertical resolution of inversion results,and at the same time,the boundary and the lateral characteristics of the sand body are well maintained to improve the accuracy of thin-layer sand body prediction and achieve an improved practical application effect.
基金Supported by National Natural Science Foundation of China(Grant No.52205045)National Key Research and Development Program of China(Grant No.2021YFB2011300)+2 种基金Aeronautical Science Foundation of China(Grant No.2022Z029051001)Zhejiang Provincial Natural Science Foundation of China(Grant No.LZ24E050006)Research Fund of State Key Laboratory of Mechanics and Control for Aerospace Structures(Nanjing University of Aeronautics and Astronautics)(Grant No.MCAS-E-0224G01).
文摘The aircraft braking system is critical to ensure the safe take-off and landing of the aircraft.However,the braking system is often exposed to high temperatures and strong vibration working environments,which makes the sensor prone to failure.Sensor failure has the potential to compromise aircraft safety.In order to improve the safety of the aircraft braking system,a fault detection and fault-tolerant control(FDFTC)strategy for the aircraft brake pressure sensor is designed.Firstly,a model based on a bidirectional long short-term memory(Bi-LSTM)network is constructed to estimate the brake pressure.Then,the residual sequence is obtained by comparing the measured pressure with the estimated pressure.On this basis,the improved sequential probability ratio test(SPRT)method based on mathematical statistics is applied to analyze the residual sequence to detect the fault.Finally,simulation and hardware-in-the-loop(HIL)testing results indicate that the proposed FDFTC strategy can detect sensor faults in time and efficiently complete braking when faults occur.Hence,the proposed FDFTC strategy can effectively deal with the faults of the aircraft brake pressure sensor,which is of great significance to improve the reliability and safety of the aircraft.
基金supported by the National Natural Science Foundation of China(No.51977113)the Science and Technology Project of State Grid Zhejiang Electric Power Co.,Ltd.(No.5211JX240001).
文摘Accurate and reliable power system data are fundamental for critical operations such as gridmonitoring,fault diagnosis,and load forecasting,underpinned by increasing intelligentization and digitalization.However,data loss and anomalies frequently compromise data integrity in practical settings,significantly impacting system operational efficiency and security.Most existing data recovery methods require complete datasets for training,leading to substantial data and computational demands and limited generalization.To address these limitations,this study proposes a missing data imputation model based on an improved Generative Adversarial Network(BAC-GAN).Within the BAC-GAN framework,the generator utilizes Bidirectional Long Short-Term Memory(BiLSTM)networks and Multi-Head Attention mechanisms to capture temporal dependencies and complex relationships within power system data.The discriminator employs a Convolutional Neural Network(CNN)architecture to integrate local features with global structures,effectivelymitigating the generation of implausible imputations.Experimental results on two public datasets demonstrate that the BAC-GAN model achieves superior data recovery accuracy compared to five state-of-the-art and classical benchmarkmethods,with an average improvement of 17.7%in reconstruction accuracy.The proposedmethod significantly enhances the accuracy of grid fault diagnosis and provides reliable data support for the stable operation of smart grids,showing great potential for practical applications in power systems.
基金funded by scientific research projects under Grant JY2024B011.
文摘Accurate prediction of remaining useful life serves as a reliable basis for maintenance strategies,effectively reducing both the frequency of failures and associated costs.As a core component of PHM,RUL prediction plays a crucial role in preventing equipment failures and optimizing maintenance decision-making.However,deep learning models often falter when processing raw,noisy temporal signals,fail to quantify prediction uncertainty,and face challenges in effectively capturing the nonlinear dynamics of equipment degradation.To address these issues,this study proposes a novel deep learning framework.First,a newbidirectional long short-termmemory network integrated with an attention mechanism is designed to enhance temporal feature extraction with improved noise robustness.Second,a probabilistic prediction framework based on kernel density estimation is constructed,incorporating residual connections and stochastic regularization to achieve precise RUL estimation.Finally,extensive experiments on the C-MAPSS dataset demonstrate that our method achieves competitive performance in terms of RMSE and Score metrics compared to state-of-the-artmodels.More importantly,the probabilistic output provides a quantifiablemeasure of prediction confidence,which is crucial for risk-informed maintenance planning,enabling managers to optimize maintenance strategies based on a quantifiable understanding of failure risk.
基金supported by the National Natural Science Foundation of China(Nos.62106283 and 72001214)。
文摘The battlefield environment is changing rapidly,and fast and accurate identification of the tactical intention of enemy targets is an important condition for gaining a decision-making advantage.The current Intention Recognition(IR)method for air targets has shortcomings in temporality,interpretability and back-and-forth dependency of intentions.To address these problems,this paper designs a novel air target intention recognition method named STABC-IR,which is based on Bidirectional Gated Recurrent Unit(Bi GRU)and Conditional Random Field(CRF)with Space-Time Attention mechanism(STA).First,the problem of intention recognition of air targets is described and analyzed in detail.Then,a temporal network based on Bi GRU is constructed to achieve the temporal requirement.Subsequently,STA is proposed to focus on the key parts of the features and timing information to meet certain interpretability requirements while strengthening the timing requirements.Finally,an intention transformation network based on CRF is proposed to solve the back-and-forth dependency and transformation problem by jointly modeling the tactical intention of the target at each moment.The experimental results show that the recognition accuracy of the jointly trained STABC-IR model can reach 95.7%,which is higher than other latest intention recognition methods.STABC-IR solves the problem of intention transformation for the first time and considers both temporality and interpretability,which is important for improving the tactical intention recognition capability and has reference value for the construction of command and control auxiliary decision-making system.
基金Project (No. 60074008) supported by the National Natural Science Foundation of China
文摘Studies on the stability of the equilibrium points of continuous bidirectional associative memory (BAM) neural network have yielded many useful results. A novel neural network model called standard neural network model (SNNM) is ad- vanced. By using state affine transformation, the BAM neural networks were converted to SNNMs. Some sufficient conditions for the global asymptotic stability of continuous BAM neural networks were derived from studies on the SNNMs’ stability. These conditions were formulated as easily verifiable linear matrix inequalities (LMIs), whose conservativeness is relatively low. The approach proposed extends the known stability results, and can also be applied to other forms of recurrent neural networks (RNNs).
基金Supported by the National Natural Science Foundation of China (6067402760875039)+1 种基金Specialized Research Fund for the Doctoral Program of Higher Education (20050446001)Scientific Research Foundation of Qufu Normal University
文摘Several novel stability conditions for BAM neural networks with time-varying delays are studied.Based on Lyapunov-Krasovskii functional combined with linear matrix inequality approach,the delay-dependent linear matrix inequality（LMI） conditions are established to guarantee robust asymptotic stability for given delayed BAM neural networks.These criteria can be easily verified by utilizing the recently developed algorithms for solving LMIs.A numerical example is provided to demonstrate the effectiveness and less conservatism of the main results.
基金supported by the Inner Mongolia Power Company 2024 Staff Innovation Studio Innovation Project“Research on Cluster Output Prediction and Group Control Technology for County-Wide Distributed Photovoltaic Construction”.
文摘Responding to the stochasticity and uncertainty in the power height of distributed photovoltaic power generation.This paper presents a distributed photovoltaic ultra-short-term power forecasting method based on Variational Mode Decomposition(VMD)and Channel Attention Mechanism.First,Pearson’s correlation coefficient was utilized to filter out the meteorological factors that had a high impact on historical power.Second,the distributed PV power data were decomposed into a relatively smooth power series with different fluctuation patterns using variational modal decomposition(VMD).Finally,the reconstructed distributed PV power as well as other features are input into the combined CNN-SENet-BiLSTM model.In this model,the convolutional neural network(CNN)and channel attention mechanism dynamically adjust the weights while capturing the spatial features of the input data to improve the discriminative ability of key features.The extracted data is then fed into the bidirectional long short-term memory network(BiLSTM)to capture the time-series features,and the final output is the prediction result.The verification is conducted using a dataset from a distributed photovoltaic power station in the Northwest region of China.The results show that compared with other prediction methods,the method proposed in this paper has a higher prediction accuracy,which helps to improve the proportion of distributed PV access to the grid,and can guarantee the safe and stable operation of the power grid.
基金supported by National Key R&D Program of China(2018YFA0702501)National Natural Science Foundation of China (41974140)+1 种基金Science and Technology Management Department,China National Petroleum Corporation(2022DQ0604-01)China National Petroleum Corporation-China University of Petroleum (Beijing) Strategy。
文摘Traditional deep learning methods pursue complex and single network architectures without considering the petrophysical relationship between different elastic parameters.The mathematical and statistical significance of the inversion results may lead to model overfitting,especially when there are a limited number of well logs in a working area.Multitask learning provides an eff ective approach to addressing this issue.Simultaneously,learning multiple related tasks can improve a model’s generalization ability to a certain extent,thereby enhancing the performance of related tasks with an equal amount of labeled data.In this study,we propose an end-to-end multitask deep learning model that integrates a fully convolutional network and bidirectional gated recurrent unit for intelligent prestack inversion of“seismic data to elastic parameters.”The use of a Bayesian homoscedastic uncertainty-based loss function enables adaptive learning of the weight coeffi cients for diff erent elastic parameter inversion tasks,thereby reducing uncertainty during the inversion process.The proposed method combines the local feature perception of convolutional neural networks with the long-term memory of bidirectional gated recurrent networks.It maintains the rock physics constraint relationships among diff erent elastic parameters during the inversion process,demonstrating a high level of prediction accuracy.Numerical simulations and processing results of real seismic data validate the eff ectiveness and practicality of the proposed method.
基金funded by the Key Technologies R&D Program of CNBM(2023SJYL01)Postgraduate Research&Practice Innovation Program of Jiangsu Province(SJCX24_1356).
文摘Throughout the composite structure’s lifespan,it is subject to a range of environmental factors,including loads,vibrations,and conditions involving heat and humidity.These factors have the potential to compromise the integrity of the structure.The estimation of the fatigue life of composite materials is imperative for ensuring the structural integrity of these materials.In this study,a methodology is proposed for predicting the fatigue life of composites that integrates ultrasonic guided waves and machine learning modeling.The method first screens the ultrasonic guided wave signal features that are significantly affected by fatigue damage.Subsequently,a covariance analysis is conducted to reduce the redundancy of the feature matrix.Furthermore,one-hot encoding is employed to incorporate boundary conditions as features,and the resulting data undergoes preprocessing to form a sample library.A composite fatigue life prediction model has been developed,employing the aforementioned sample library as the input source and utilizing remaining life as the output metric.The model synthesizes the strengths of convolutional neural networks(CNNs)and bidirectional long short-term memory networks(BiLSTMs)while leveraging Bayesian optimization(BO)to enhance the optimization of hyperparameters.The experimental results demonstrate that the proposed BO-CNN-BiLSTM model exhibits superior performance in terms of prediction accuracy and reliability in the damage regression task when compared to both the BiLSTM and CNN-BiLSTM models.
基金supported by the National Natural Science Foundation of China(Grant No.62102449).
文摘To address the limitations of existing abnormal traffic detection methods,such as insufficient temporal and spatial feature extraction,high false positive rate(FPR),poor generalization,and class imbalance,this study proposed an intelligent detection method that combines a Stacked Convolutional Network(SCN),Bidirectional Long Short-Term Memory(BiLSTM)network,and Equalization Loss v2(EQL v2).This method was divided into two components:a feature extraction model and a classification and detection model.First,SCN was constructed by combining a Convolutional Neural Network(CNN)with a Depthwise Separable Convolution(DSC)network to capture the abstract spatial features of traffic data.These features were then input into the BiLSTM to capture temporal dependencies.An attention mechanism was incorporated after SCN and BiLSTM to enhance the extraction of key spatiotemporal features.To address class imbalance,the classification detection model applied EQL v2 to adjust the weights of the minority classes,ensuring that they received equal focus during training.The experimental results indicated that the proposed method outperformed the existing methods in terms of accuracy,FPR,and F1-score and significantly improved the identification rate of minority classes.
基金the National Social Science Foundation of China(No.17BYY047)
文摘Electronic Medical Records(EMR) with unstructured sentences and various conceptual expressions provide rich information for medical information extraction. However, common Named Entity Recognition(NER)in Natural Language Processing(NLP) are not well suitable for clinical NER in EMR. This study aims at applying neural networks to clinical concept extractions. We integrate Bidirectional Long Short-Term Memory Networks(Bi-LSTM) with a Conditional Random Fields(CRF) layer to detect three types of clinical named entities. Word representations fed into the neural networks are concatenated by character-based word embeddings and Continuous Bag of Words(CBOW) embeddings trained both on domain and non-domain corpus. We test our NER system on i2b2/VA open datasets and compare the performance with six related works, achieving the best result of NER with F1 value 0.853 7. We also point out a few specific problems in clinical concept extractions which will give some hints to deeper studies.
基金the National Key Research and Development Program of China(2017YFA0701005)National Natural Science Foundation of China(62271320,61871268)+1 种基金“Shuguang”Program of Shanghai Education Commission(19SG44)the 111 Project(D18014).
文摘Metasurfaces,which are the two-dimensional counterparts of metamaterials,have demonstrated unprecedented capabilities to manipulate the wavefront of electromagnetic waves in a single flat device.Despite various advances in this field,the unique functionalities achieved by metasurfaces have come at the cost of the structural complexity,resulting in a time-consuming parameter sweep for the conventional metasurface design.Although artificial neural networks provide a flexible platform for significantly improving the design process,the current metasurface designs are restricted to generating qualitative field distributions.In this study,we demonstrate that by combining a tandem neural network and an iterative algorithm,the previous restriction of the design of metasurfaces can be overcome with quantitative field distributions.As proof-of-principle examples,metalenses predicted via the designed network architecture that possess multiple focal points with identical/orthogonal polarisation states,as well as accurate intensity ratios(quantitative field distributions),were numerically calculated and experimentally demonstrated.The unique and robust approach for the metasurface design will enable the acceleration of the development of devices with high-accuracy functionalities,which can be applied in imaging,detecting,and sensing.
