Recently,Industrial Control Systems(ICSs)have been changing from a closed environment to an open environment because of the expansion of digital transformation,smart factories,and Industrial Internet of Things(IIoT).S...Recently,Industrial Control Systems(ICSs)have been changing from a closed environment to an open environment because of the expansion of digital transformation,smart factories,and Industrial Internet of Things(IIoT).Since security accidents that occur in ICSs can cause national confusion and human casualties,research on detecting abnormalities by using normal operation data learning is being actively conducted.The single technique proposed by existing studies does not detect abnormalities well or provide satisfactory results.In this paper,we propose a GRU-based Buzzer Ensemble for AbnormalDetection(GBE-AD)model for detecting anomalies in industrial control systems to ensure rapid response and process availability.The newly proposed ensemble model of the buzzer method resolves False Negatives(FNs)by complementing the limited range that can be detected in a single model because of the internal models composing GBE-AD.Because the internal models remain suppressed for False Positives(FPs),GBE-AD provides better generalization.In addition,we generated mean prediction error data in GBE-AD and inferred abnormal processes using soft and hard clustering.We confirmed that the detection model’s Time-series Aware Precision(TaP)suppressed FPs at 97.67%.The final performance was 94.04%in an experiment using anHIL-basedAugmented ICS(HAI)Security Dataset(ver.21.03)among public datasets.展开更多
The rapid integration of Internet of Things(IoT)technologies is reshaping the global energy landscape by deploying smart meters that enable high-resolution consumption monitoring,two-way communication,and advanced met...The rapid integration of Internet of Things(IoT)technologies is reshaping the global energy landscape by deploying smart meters that enable high-resolution consumption monitoring,two-way communication,and advanced metering infrastructure services.However,this digital transformation also exposes power system to evolving threats,ranging from cyber intrusions and electricity theft to device malfunctions,and the unpredictable nature of these anomalies,coupled with the scarcity of labeled fault data,makes realtime detection exceptionally challenging.To address these difficulties,a real-time decision support framework is presented for smart meter anomality detection that leverages rolling time windows and two self-supervised contrastive learning modules.The first module synthesizes diverse negative samples to overcome the lack of labeled anomalies,while the second captures intrinsic temporal patterns for enhanced contextual discrimination.The end-to-end framework continuously updates its model with rolling updated meter data to deliver timely identification of emerging abnormal behaviors in evolving grids.Extensive evaluations on eight publicly available smart meter datasets over seven diverse abnormal patterns testing demonstrate the effectiveness of the proposed full framework,achieving average recall and F1 score of more than 0.85.展开更多
Nowadays,abnormal traffic detection for Software-Defined Networking(SDN)faces the challenges of large data volume and high dimensionality.Since traditional machine learning-based detection methods have the problem of ...Nowadays,abnormal traffic detection for Software-Defined Networking(SDN)faces the challenges of large data volume and high dimensionality.Since traditional machine learning-based detection methods have the problem of data redundancy,the Metaheuristic Algorithm(MA)is introduced to select features beforemachine learning to reduce the dimensionality of data.Since a Tyrannosaurus Optimization Algorithm(TROA)has the advantages of few parameters,simple implementation,and fast convergence,and it shows better results in feature selection,TROA can be applied to abnormal traffic detection for SDN.However,TROA suffers frominsufficient global search capability,is easily trapped in local optimums,and has poor search accuracy.Then,this paper tries to improve TROA,namely the Improved Tyrannosaurus Optimization Algorithm(ITROA).It proposes a metaheuristic-driven abnormal traffic detection model for SDN based on ITROA.Finally,the validity of the ITROA is verified by the benchmark function and the UCI dataset,and the feature selection optimization operation is performed on the InSDN dataset by ITROA and other MAs to obtain the optimized feature subset for SDN abnormal traffic detection.The experiment shows that the performance of the proposed ITROA outperforms compared MAs in terms of the metaheuristic-driven model for SDN,achieving an accuracy of 99.37%on binary classification and 96.73%on multiclassification.展开更多
Blockchain platform swith the unique characteristics of anonymity,decentralization,and transparency of their transactions,which are faced with abnormal activities such as money laundering,phishing scams,and fraudulent...Blockchain platform swith the unique characteristics of anonymity,decentralization,and transparency of their transactions,which are faced with abnormal activities such as money laundering,phishing scams,and fraudulent behavior,posing a serious threat to account asset security.For these potential security risks,this paper proposes a hybrid neural network detection method(HNND)that learns multiple types of account features and enhances fusion information among them to effectively detect abnormal transaction behaviors in the blockchain.In HNND,the Temporal Transaction Graph Attention Network(T2GAT)is first designed to learn biased aggregation representation of multi-attribute transactions among nodes,which can capture key temporal information from node neighborhood transactions.Then,the Graph Convolutional Network(GCN)is adopted which captures abstract structural features of the transaction network.Further,the Stacked Denoising Autoencode(SDA)is developed to achieve adaptive fusion of thses features from different modules.Moreover,the SDA enhances robustness and generalization ability of node representation,leading to higher binary classification accuracy in detecting abnormal behaviors of blockchain accounts.Evaluations on a real-world abnormal transaction dataset demonstrate great advantages of the proposed HNND method over other compared methods.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
This paper addresses the challenge of identifying abnormal states in Lithium-ion Battery(LiB)time series data.As the energy sector increasingly focuses on integrating distributed energy resources,Virtual Power Plants(...This paper addresses the challenge of identifying abnormal states in Lithium-ion Battery(LiB)time series data.As the energy sector increasingly focuses on integrating distributed energy resources,Virtual Power Plants(VPP)have become a vital new framework for energy management.LiBs are key in this context,owing to their high-efficiency energy storage capabilities essential for VPP operations.However,LiBs are prone to various abnormal states like overcharging,over-discharging,and internal short circuits,which impede power transmission efficiency.Traditional methods for detecting such abnormalities in LiB are too broad and lack precision for the dynamic and irregular nature of LiB data.In response,we introduce an innovative method:a Long Short-Term Memory(LSTM)autoencoder based on Dynamic Frequency Memory and Correlation Attention(DFMCA-LSTM-AE).This unsupervised,end-to-end approach is specifically designed for dynamically monitoring abnormal states in LiB data.The method starts with a Dynamic Frequency Fourier Transform module,which dynamically captures the frequency characteristics of time series data across three scales,incorporating a memory mechanism to reduce overgeneralization of abnormal frequencies.This is followed by integrating LSTM into both the encoder and decoder,enabling the model to effectively encode and decode the temporal relationships in the time series.Empirical tests on a real-world LiB dataset demonstrate that DFMCA-LSTM-AE outperforms existing models,achieving an average Area Under the Curve(AUC)of 90.73%and an F1 score of 83.83%.These results mark significant improvements over existing models,ranging from 2.4%–45.3%for AUC and 1.6%–28.9%for F1 score,showcasing the model’s enhanced accuracy and reliability in detecting abnormal states in LiB data.展开更多
Laboratory safety is a critical area of broad societal concern,particularly in the detection of abnormal actions.To enhance the efficiency and accuracy of detecting such actions,this paper introduces a novel method ca...Laboratory safety is a critical area of broad societal concern,particularly in the detection of abnormal actions.To enhance the efficiency and accuracy of detecting such actions,this paper introduces a novel method called TubeRAPT(Tubelet Transformer based onAdapter and Prefix TrainingModule).Thismethod primarily comprises three key components:the TubeR network,an adaptive clustering attention mechanism,and a prefix training module.These components work in synergy to address the challenge of knowledge preservation in models pretrained on large datasets while maintaining training efficiency.The TubeR network serves as the backbone for spatio-temporal feature extraction,while the adaptive clustering attention mechanism refines the focus on relevant information.The prefix training module facilitates efficient fine-tuning and knowledge transfer.Experimental results demonstrate the effectiveness of TubeRAPT,achieving a 68.44%mean Average Precision(mAP)on the CLA(Crazy LabActivity)small-scale dataset,marking a significant improvement of 1.53%over the previous TubeR method.This research not only showcases the potential applications of TubeRAPT in the field of abnormal action detection but also offers innovative ideas and technical support for the future development of laboratory safety monitoring technologies.The proposed method has implications for improving safety management systems in various laboratory environments,potentially reducing accidents and enhancing overall workplace safety.展开更多
Timely detection of abnormal electricity consumption behaviors plays a key role in saving energy.However,the detection of abnormal electricity consumption faces many problems.Imbalanced data are important challenges i...Timely detection of abnormal electricity consumption behaviors plays a key role in saving energy.However,the detection of abnormal electricity consumption faces many problems.Imbalanced data are important challenges in this field.When the normal data are much more than the abnormal data,the network can hardly recognize the features of the minority class data,which generates low detection efficiency.Therefore,in this paper,we employ adaptive synthetic sampling(ADASYN)to achieve effective expansion of the minority class data.In addition,we adopt gated recurrent units to complete the classification of electricity consumption data.We conduct detailed experiments to verify this proposed method.Experimental results show that this method is more effective than other methods.展开更多
Thermal power generation systems have stringent requirements for water and steam quality,i.e.,condensate water quality is one of the critical issues.In this paper,we designed a two-layer model based on an autoencoder ...Thermal power generation systems have stringent requirements for water and steam quality,i.e.,condensate water quality is one of the critical issues.In this paper,we designed a two-layer model based on an autoencoder and expert knowledge to achieve the early warning and causal analysis of condensate water quality abnormalities.An early warning model using an autoencoder model is built based on the historical data affecting the condensate water quality.Next,an analytical model of condensate water quality abnormalities was then developed by combining expert knowledge and trend test algorithms.Two different datasets were used to test the proposed model,respectively.The accuracy of the autoencoder model in the short-period test set is 88.83%,which shows that the early warning model can accurately analyze the condensate water quality data and achieve the purpose of early warning.For the long-time period test set,the model can correctly identify each abnormality and simultaneously indicates the cause of the abnormal condensate water quality.The proposed model can correctly identify abnormal working conditions and it is applicable to other thermal power plants.展开更多
With the evolution of video surveillance systems,the requirement of video storage grows rapidly;in addition,safe guards and forensic officers spend a great deal of time observing surveillance videos to find abnormal e...With the evolution of video surveillance systems,the requirement of video storage grows rapidly;in addition,safe guards and forensic officers spend a great deal of time observing surveillance videos to find abnormal events.As most of the scene in the surveillance video are redundant and contains no information needs attention,we propose a video condensation method to summarize the abnormal events in the video by rearranging the moving trajectory and sort them by the degree of anomaly.Our goal is to improve the condensation rate to reduce more storage size,and increase the accuracy in abnormal detection.As the trajectory feature is the key to both goals,in this paper,a new method for feature extraction of moving object trajectory is proposed,and we use the SOINN(Self-Organizing Incremental Neural Network)method to accomplish a high accuracy abnormal detection.In the results,our method is able to shirk the video size to 10%storage size of the original video,and achieves 95%accuracy of abnormal event detection,which shows our method is useful and applicable to the surveillance industry.展开更多
To improve the detection accuracy and robustness of crowd anomaly detection,especially crowd emergency evacuation detection,the abnormal crowd behavior detection method is proposed.This method is based on the improved...To improve the detection accuracy and robustness of crowd anomaly detection,especially crowd emergency evacuation detection,the abnormal crowd behavior detection method is proposed.This method is based on the improved statistical global optical flow entropy which can better describe the degree of chaos of crowd.First,the optical flow field is extracted from the video sequences and a 2D optical flow histogram is gained.Then,the improved optical flow entropy,combining information theory with statistical physics is calculated from 2D optical flow histograms.Finally,the anomaly can be detected according to the abnormality judgment formula.The experimental results show that the detection accuracy achieved over 95%in three public video datasets,which indicates that the proposed algorithm outperforms other state-of-the-art algorithms.展开更多
Assessing machine's performance through comparing the same or similar machines is important to implement intelligent maintenance for swarm machine.In this paper,an outlier mining based abnormal machine detection a...Assessing machine's performance through comparing the same or similar machines is important to implement intelligent maintenance for swarm machine.In this paper,an outlier mining based abnormal machine detection algorithm is proposed for this purpose.Firstly,the outlier mining based on clustering is introduced and the definition of cluster-based global outlier factor(CBGOF) is presented.Then the modified swarm intelligence clustering(MSIC) algorithm is suggested and the outlier mining algorithm based on MSIC is proposed.The algorithm can not only cluster machines according to their performance but also detect possible abnormal machines.Finally,a comparison of mobile soccer robots' performance proves the algorithm is feasible and effective.展开更多
Video synopsis is an effective and innovative way to produce short video abstraction for huge video archives,while keeping the dynamic characteristic of activities in the original video.Abnormal activity,as the critic...Video synopsis is an effective and innovative way to produce short video abstraction for huge video archives,while keeping the dynamic characteristic of activities in the original video.Abnormal activity,as the critical event,is always the main concern in video surveillance context.However,in traditional video synopsis,all the normal and abnormal activities are condensed together equally,which can make the synopsis video confused and worthless.In addition,the traditional video synopsis methods always neglect redundancy in the content domain.To solve the above-mentioned issues,a novel video synopsis method is proposed based on abnormal activity detection and key observation selection.In the proposed algorithm,activities are classified into normal and abnormal ones based on the sparse reconstruction cost from an atomically learned activity dictionary.And key observation selection using the minimum description length principle is conducted for eliminating content redundancy in normal activity.Experiments conducted in publicly available datasets demonstrate that the proposed approach can effectively generate satisfying synopsis videos.展开更多
In this paper, we proposed a model-based abnormality detection scheme for a class of nonlinear parabolic distributed parameter systems (DPSs). The proposed methodology consists of the design of an observer and an abno...In this paper, we proposed a model-based abnormality detection scheme for a class of nonlinear parabolic distributed parameter systems (DPSs). The proposed methodology consists of the design of an observer and an abnormality detection filter (ADF) based on the backstepping technique and a limited number of in-domain measurements plus one boundary measurement. By taking the difference between the measured and estimated outputs from observer, a residual signal is generated for fault detection. For the detection purpose, the residual is evaluated in a lumped manner and we propose an explicit expression for the time-varying threshold. The convergence properties of the PDE observer and the residual are analyzed by Lyapunov stability theory. Eventually, the proposed abnormality detection scheme is demonstrated on a nonlinear DPS.展开更多
Power Line Communications-Artificial Intelligence of Things(PLC-AIo T)combines the low cost and high coverage of PLC with the learning ability of Artificial Intelligence(AI)to provide data collection and transmission ...Power Line Communications-Artificial Intelligence of Things(PLC-AIo T)combines the low cost and high coverage of PLC with the learning ability of Artificial Intelligence(AI)to provide data collection and transmission capabilities for PLC-AIo T devices in smart parks.With the development of smart parks,their emerging services require secure and accurate time synchronization of PLC-AIo T devices.However,the impact of attackers on the accuracy of time synchronization cannot be ignored.To solve the aforementioned problems,we propose a tampering attack-aware Deep Q-Network(DQN)-based time synchronization algorithm.First,we construct an abnormal clock source detection model.Then,the abnormal clock source is detected and excluded by comparing the time synchronization information between the device and the gateway.Finally,the proposed algorithm realizes the joint guarantee of high accuracy and low delay for PLC-AIo T in smart parks by intelligently selecting the multi-clock source cooperation strategy and timing weights.Simulation results show that the proposed algorithm has better time synchronization delay and accuracy performance.展开更多
In order to implement the real-time detection of abnormality of elder and devices in an empty nest home,multi-modal joint sensors are used to collect discrete action sequences of behavior,and the improved hierarchical...In order to implement the real-time detection of abnormality of elder and devices in an empty nest home,multi-modal joint sensors are used to collect discrete action sequences of behavior,and the improved hierarchical hidden Markov model is adopted to Abstract these discrete action sequences captured by multi-modal joint sensors into an occupant’s high-level behavior—event,then structure representation models of occupant normality are modeled from large amounts of spatio-temporal data. These models are used as classifiers of normality to detect an occupant’s abnormal behavior.In order to express context information needed by reasoning and detection,multi-media ontology （MMO） is designed to annotate and reason about the media information in the smart monitoring system.A pessimistic emotion model （PEM） is improved to analyze multi-interleaving events of multi-active devices in the home.Experiments demonstrate that the PEM can enhance the accuracy and reliability for detecting active devices when these devices are in blind regions or are occlusive. The above approach has good performance in detecting abnormalities involving occupants and devices in a real-time way.展开更多
With the development of intelligent and netw orking technology in automobile,the malicious attacks against in-vehicle CAN netw orks are increasing day by day,and the problem of information safety in automobile is aggr...With the development of intelligent and netw orking technology in automobile,the malicious attacks against in-vehicle CAN netw orks are increasing day by day,and the problem of information safety in automobile is aggravated. In this regard,this paper analyzes the security loopholes and threats w hich the CAN bus faced,put forw ard a kind of anomaly detection algorithm for vehicle CAN bus. The method uses support vector machine algorithm to distinguish betw een normal message and abnormal message,so as to realize the CAN bus anomaly detection. Theoretical and experimental studies show that this method can effectively detect abnormal packets in the CAN bus w ith a detection rate of over 90%,w hich can effectively resist malicious attacks such as tampering and cheating on the vehicle CAN bus.展开更多
Background:The accurate identification of cardiac abnormalities is essential for proper diagnosis and effective treatment of cardiovascular diseases.Method:This work introduces an advanced methodology for detecting ca...Background:The accurate identification of cardiac abnormalities is essential for proper diagnosis and effective treatment of cardiovascular diseases.Method:This work introduces an advanced methodology for detecting cardiac abnormalities and estimating electrocardiographic age(ECG Age)using sophisticated signal processing and deep learning techniques.This study looks at six main heart conditions found in 12-lead electrocardiogram(ECG)data.It addresses important issues like class imbalances,missing lead scenarios,and model generalizations.A modified residual neural network(ResNet)architecture was developed to enhance the detection of cardiac abnormalities.Results:The proposed ResNet demonst rated superior performance when compared with two linear models and an alternative ResNet architectures,achieving an overall classification accuracy of 91.25%and an F1 score of 93.9%,surpassing baseline models.A comprehensive lead loss analysis was conducted,evaluating model performance across 4096 combinations of missing leads.The results revealed that pulse rate-based factors remained robust with up to 75%lead loss,while block-based factors experienced significant performance declines beyond the loss of four leads.Conclusion:This analysis highlighted the importance of addressing lead loss impacts to maintain a robust model.To optimize performance,targeted training approaches were developed for different conditions.Based on these insights,a grouping strategy was implemented to train specialized models for pulse rate-based and block-based conditions.This approach resulted in notable improvements,achieving an overall classification accuracy of 95.12%and an F1 score of 95.79%.展开更多
Aiming at the problem of abnormal data generated by a power transformer on-line monitoring system due to the influences of transformer operation state change,external environmental interference,communication interrupt...Aiming at the problem of abnormal data generated by a power transformer on-line monitoring system due to the influences of transformer operation state change,external environmental interference,communication interruption,and other factors,a method of anomaly recognition and differentiation for monitoring data was proposed.Firstly,the empirical wavelet transform(EWT)and the autoregressive integrated moving average(ARIMA)model were used for time series modelling of monitoring data to obtain the residual sequence reflecting the anomaly monitoring data value,and then the isolation forest algorithm was used to identify the abnormal information,and the monitoring sequence was segmented according to the recognition results.Secondly,the segmented sequence was symbolised by the improved multi-dimensional SAX vector representation method,and the assessment of the anomaly pattern was made by calculating the similarity score of the adjacent symbol vectors,and the monitoring sequence correlation was further used to verify the assessment.Finally,the case study result shows that the proposed method can reliably recognise abnormal data and accurately distinguish between invalid and valid anomaly patterns.展开更多
The deep learning method has made nurnerials achievements regarding anomaly detection in the field of time series.We introduce the speech production model in the field of artificial intelligence,changing the convoluti...The deep learning method has made nurnerials achievements regarding anomaly detection in the field of time series.We introduce the speech production model in the field of artificial intelligence,changing the convolution layer of the general convolution neural network to the residual element structure by adding identity mapping,and expanding the receptive domain of the model by using the dilated causal convolution.Based on the dilated causal convolution network and the method of log probability density function,the anomalous events are detected according to the anomaly scores.The validity of the method is verified by the simulation data,which is applied to the actual observed data on the observation staion of Pingliang geoeletric field in Gansu Province.The results show that one month before the Wenchuan M_S8.0,Lushan M_S7.0 and Minxian-Zhangxian M_S6.6 earthquakes,the daily cumulative error of log probability density of the predicted results in Pingliang Station suddenly decreases,which is consistent with the actual earthquake anomalies in a certain time range.After analyzing the combined factors including the spatial electromagnetic environment and the variation of micro fissures before the earthquake,we explain the possible causes of the anomalies in the geoelectric field of before the earthquake.The successful application of deep learning in observed data of the geoelectric field may behefit for improving the ultilization rate both the data and the efficiency of detection.Besides,it may provide technical support for more seismic research.展开更多
基金supported by Institute of Information&communications Technology Planning&Evaluation(IITP)grant funded by Korea government Ministry of Science,ICT(MSIT)(No.2019-0-01343,convergence security core talent training business).
文摘Recently,Industrial Control Systems(ICSs)have been changing from a closed environment to an open environment because of the expansion of digital transformation,smart factories,and Industrial Internet of Things(IIoT).Since security accidents that occur in ICSs can cause national confusion and human casualties,research on detecting abnormalities by using normal operation data learning is being actively conducted.The single technique proposed by existing studies does not detect abnormalities well or provide satisfactory results.In this paper,we propose a GRU-based Buzzer Ensemble for AbnormalDetection(GBE-AD)model for detecting anomalies in industrial control systems to ensure rapid response and process availability.The newly proposed ensemble model of the buzzer method resolves False Negatives(FNs)by complementing the limited range that can be detected in a single model because of the internal models composing GBE-AD.Because the internal models remain suppressed for False Positives(FPs),GBE-AD provides better generalization.In addition,we generated mean prediction error data in GBE-AD and inferred abnormal processes using soft and hard clustering.We confirmed that the detection model’s Time-series Aware Precision(TaP)suppressed FPs at 97.67%.The final performance was 94.04%in an experiment using anHIL-basedAugmented ICS(HAI)Security Dataset(ver.21.03)among public datasets.
文摘The rapid integration of Internet of Things(IoT)technologies is reshaping the global energy landscape by deploying smart meters that enable high-resolution consumption monitoring,two-way communication,and advanced metering infrastructure services.However,this digital transformation also exposes power system to evolving threats,ranging from cyber intrusions and electricity theft to device malfunctions,and the unpredictable nature of these anomalies,coupled with the scarcity of labeled fault data,makes realtime detection exceptionally challenging.To address these difficulties,a real-time decision support framework is presented for smart meter anomality detection that leverages rolling time windows and two self-supervised contrastive learning modules.The first module synthesizes diverse negative samples to overcome the lack of labeled anomalies,while the second captures intrinsic temporal patterns for enhanced contextual discrimination.The end-to-end framework continuously updates its model with rolling updated meter data to deliver timely identification of emerging abnormal behaviors in evolving grids.Extensive evaluations on eight publicly available smart meter datasets over seven diverse abnormal patterns testing demonstrate the effectiveness of the proposed full framework,achieving average recall and F1 score of more than 0.85.
基金supported by the National Natural Science Foundation of China under Grant 61602162the Hubei Provincial Science and Technology Plan Project under Grant 2023BCB041.
文摘Nowadays,abnormal traffic detection for Software-Defined Networking(SDN)faces the challenges of large data volume and high dimensionality.Since traditional machine learning-based detection methods have the problem of data redundancy,the Metaheuristic Algorithm(MA)is introduced to select features beforemachine learning to reduce the dimensionality of data.Since a Tyrannosaurus Optimization Algorithm(TROA)has the advantages of few parameters,simple implementation,and fast convergence,and it shows better results in feature selection,TROA can be applied to abnormal traffic detection for SDN.However,TROA suffers frominsufficient global search capability,is easily trapped in local optimums,and has poor search accuracy.Then,this paper tries to improve TROA,namely the Improved Tyrannosaurus Optimization Algorithm(ITROA).It proposes a metaheuristic-driven abnormal traffic detection model for SDN based on ITROA.Finally,the validity of the ITROA is verified by the benchmark function and the UCI dataset,and the feature selection optimization operation is performed on the InSDN dataset by ITROA and other MAs to obtain the optimized feature subset for SDN abnormal traffic detection.The experiment shows that the performance of the proposed ITROA outperforms compared MAs in terms of the metaheuristic-driven model for SDN,achieving an accuracy of 99.37%on binary classification and 96.73%on multiclassification.
文摘Blockchain platform swith the unique characteristics of anonymity,decentralization,and transparency of their transactions,which are faced with abnormal activities such as money laundering,phishing scams,and fraudulent behavior,posing a serious threat to account asset security.For these potential security risks,this paper proposes a hybrid neural network detection method(HNND)that learns multiple types of account features and enhances fusion information among them to effectively detect abnormal transaction behaviors in the blockchain.In HNND,the Temporal Transaction Graph Attention Network(T2GAT)is first designed to learn biased aggregation representation of multi-attribute transactions among nodes,which can capture key temporal information from node neighborhood transactions.Then,the Graph Convolutional Network(GCN)is adopted which captures abstract structural features of the transaction network.Further,the Stacked Denoising Autoencode(SDA)is developed to achieve adaptive fusion of thses features from different modules.Moreover,the SDA enhances robustness and generalization ability of node representation,leading to higher binary classification accuracy in detecting abnormal behaviors of blockchain accounts.Evaluations on a real-world abnormal transaction dataset demonstrate great advantages of the proposed HNND method over other compared methods.
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supported by“Regional Innovation Strategy(RIS)”through the National Research Foundation of Korea(NRF)funded by the Ministry of Education(MOE)(2021RIS-002)the Technology Development Program(RS-2023-00278623)funded by the Ministry of SMEs and Startups(MSS,Korea).
文摘This paper addresses the challenge of identifying abnormal states in Lithium-ion Battery(LiB)time series data.As the energy sector increasingly focuses on integrating distributed energy resources,Virtual Power Plants(VPP)have become a vital new framework for energy management.LiBs are key in this context,owing to their high-efficiency energy storage capabilities essential for VPP operations.However,LiBs are prone to various abnormal states like overcharging,over-discharging,and internal short circuits,which impede power transmission efficiency.Traditional methods for detecting such abnormalities in LiB are too broad and lack precision for the dynamic and irregular nature of LiB data.In response,we introduce an innovative method:a Long Short-Term Memory(LSTM)autoencoder based on Dynamic Frequency Memory and Correlation Attention(DFMCA-LSTM-AE).This unsupervised,end-to-end approach is specifically designed for dynamically monitoring abnormal states in LiB data.The method starts with a Dynamic Frequency Fourier Transform module,which dynamically captures the frequency characteristics of time series data across three scales,incorporating a memory mechanism to reduce overgeneralization of abnormal frequencies.This is followed by integrating LSTM into both the encoder and decoder,enabling the model to effectively encode and decode the temporal relationships in the time series.Empirical tests on a real-world LiB dataset demonstrate that DFMCA-LSTM-AE outperforms existing models,achieving an average Area Under the Curve(AUC)of 90.73%and an F1 score of 83.83%.These results mark significant improvements over existing models,ranging from 2.4%–45.3%for AUC and 1.6%–28.9%for F1 score,showcasing the model’s enhanced accuracy and reliability in detecting abnormal states in LiB data.
基金supported by the Philosophy and Social Sciences Planning Project of Guangdong Province of China(GD23XGL099)the Guangdong General Universities Young Innovative Talents Project(2023KQNCX247)the Research Project of Shanwei Institute of Technology(SWKT22-019).
文摘Laboratory safety is a critical area of broad societal concern,particularly in the detection of abnormal actions.To enhance the efficiency and accuracy of detecting such actions,this paper introduces a novel method called TubeRAPT(Tubelet Transformer based onAdapter and Prefix TrainingModule).Thismethod primarily comprises three key components:the TubeR network,an adaptive clustering attention mechanism,and a prefix training module.These components work in synergy to address the challenge of knowledge preservation in models pretrained on large datasets while maintaining training efficiency.The TubeR network serves as the backbone for spatio-temporal feature extraction,while the adaptive clustering attention mechanism refines the focus on relevant information.The prefix training module facilitates efficient fine-tuning and knowledge transfer.Experimental results demonstrate the effectiveness of TubeRAPT,achieving a 68.44%mean Average Precision(mAP)on the CLA(Crazy LabActivity)small-scale dataset,marking a significant improvement of 1.53%over the previous TubeR method.This research not only showcases the potential applications of TubeRAPT in the field of abnormal action detection but also offers innovative ideas and technical support for the future development of laboratory safety monitoring technologies.The proposed method has implications for improving safety management systems in various laboratory environments,potentially reducing accidents and enhancing overall workplace safety.
基金supported by the National Natural Science Foundation of China(Nos.62076150,62133008,61903226,and 62173216)the Taishan Scholar Project of Shandong Province(No.TSQN201812092)+1 种基金the Key Research and Development Program of Shandong Province(Nos.2021CXGC011205 and 2021TSGC1053)the Youth Innovation Technology Project of Higher School in Shandong Province(No.2019KJN005).
文摘Timely detection of abnormal electricity consumption behaviors plays a key role in saving energy.However,the detection of abnormal electricity consumption faces many problems.Imbalanced data are important challenges in this field.When the normal data are much more than the abnormal data,the network can hardly recognize the features of the minority class data,which generates low detection efficiency.Therefore,in this paper,we employ adaptive synthetic sampling(ADASYN)to achieve effective expansion of the minority class data.In addition,we adopt gated recurrent units to complete the classification of electricity consumption data.We conduct detailed experiments to verify this proposed method.Experimental results show that this method is more effective than other methods.
基金supported by the Jingneng Shiyan Thermal Power Co.,Ltd.(TPRI/TR-CA-006-2023)Huaihe Energy Power Group Co.,Ltd.(TPRI/TR-CA-040-2023)Xi'an Thermal Power Research Institute Co.,Ltd.(TPRI/TR-CA-110-2021A/H1).
文摘Thermal power generation systems have stringent requirements for water and steam quality,i.e.,condensate water quality is one of the critical issues.In this paper,we designed a two-layer model based on an autoencoder and expert knowledge to achieve the early warning and causal analysis of condensate water quality abnormalities.An early warning model using an autoencoder model is built based on the historical data affecting the condensate water quality.Next,an analytical model of condensate water quality abnormalities was then developed by combining expert knowledge and trend test algorithms.Two different datasets were used to test the proposed model,respectively.The accuracy of the autoencoder model in the short-period test set is 88.83%,which shows that the early warning model can accurately analyze the condensate water quality data and achieve the purpose of early warning.For the long-time period test set,the model can correctly identify each abnormality and simultaneously indicates the cause of the abnormal condensate water quality.The proposed model can correctly identify abnormal working conditions and it is applicable to other thermal power plants.
文摘With the evolution of video surveillance systems,the requirement of video storage grows rapidly;in addition,safe guards and forensic officers spend a great deal of time observing surveillance videos to find abnormal events.As most of the scene in the surveillance video are redundant and contains no information needs attention,we propose a video condensation method to summarize the abnormal events in the video by rearranging the moving trajectory and sort them by the degree of anomaly.Our goal is to improve the condensation rate to reduce more storage size,and increase the accuracy in abnormal detection.As the trajectory feature is the key to both goals,in this paper,a new method for feature extraction of moving object trajectory is proposed,and we use the SOINN(Self-Organizing Incremental Neural Network)method to accomplish a high accuracy abnormal detection.In the results,our method is able to shirk the video size to 10%storage size of the original video,and achieves 95%accuracy of abnormal event detection,which shows our method is useful and applicable to the surveillance industry.
基金National Natural Science Foundation of China(61701029)。
文摘To improve the detection accuracy and robustness of crowd anomaly detection,especially crowd emergency evacuation detection,the abnormal crowd behavior detection method is proposed.This method is based on the improved statistical global optical flow entropy which can better describe the degree of chaos of crowd.First,the optical flow field is extracted from the video sequences and a 2D optical flow histogram is gained.Then,the improved optical flow entropy,combining information theory with statistical physics is calculated from 2D optical flow histograms.Finally,the anomaly can be detected according to the abnormality judgment formula.The experimental results show that the detection accuracy achieved over 95%in three public video datasets,which indicates that the proposed algorithm outperforms other state-of-the-art algorithms.
基金the National Natural Science Foundation of China (No. 50705054)
文摘Assessing machine's performance through comparing the same or similar machines is important to implement intelligent maintenance for swarm machine.In this paper,an outlier mining based abnormal machine detection algorithm is proposed for this purpose.Firstly,the outlier mining based on clustering is introduced and the definition of cluster-based global outlier factor(CBGOF) is presented.Then the modified swarm intelligence clustering(MSIC) algorithm is suggested and the outlier mining algorithm based on MSIC is proposed.The algorithm can not only cluster machines according to their performance but also detect possible abnormal machines.Finally,a comparison of mobile soccer robots' performance proves the algorithm is feasible and effective.
基金Supported by the National Natural Science Foundation of China(No.61402023)Beijing Technology and Business' University Youth Fund(No.QNJJ2014-23)Beijing Natural Science Foundation(No.4162019)
文摘Video synopsis is an effective and innovative way to produce short video abstraction for huge video archives,while keeping the dynamic characteristic of activities in the original video.Abnormal activity,as the critical event,is always the main concern in video surveillance context.However,in traditional video synopsis,all the normal and abnormal activities are condensed together equally,which can make the synopsis video confused and worthless.In addition,the traditional video synopsis methods always neglect redundancy in the content domain.To solve the above-mentioned issues,a novel video synopsis method is proposed based on abnormal activity detection and key observation selection.In the proposed algorithm,activities are classified into normal and abnormal ones based on the sparse reconstruction cost from an atomically learned activity dictionary.And key observation selection using the minimum description length principle is conducted for eliminating content redundancy in normal activity.Experiments conducted in publicly available datasets demonstrate that the proposed approach can effectively generate satisfying synopsis videos.
文摘In this paper, we proposed a model-based abnormality detection scheme for a class of nonlinear parabolic distributed parameter systems (DPSs). The proposed methodology consists of the design of an observer and an abnormality detection filter (ADF) based on the backstepping technique and a limited number of in-domain measurements plus one boundary measurement. By taking the difference between the measured and estimated outputs from observer, a residual signal is generated for fault detection. For the detection purpose, the residual is evaluated in a lumped manner and we propose an explicit expression for the time-varying threshold. The convergence properties of the PDE observer and the residual are analyzed by Lyapunov stability theory. Eventually, the proposed abnormality detection scheme is demonstrated on a nonlinear DPS.
基金supported by the Science and Technology Project of the State Grid Corporation of China under Grant Number 5400202199541A-0-5-ZN。
文摘Power Line Communications-Artificial Intelligence of Things(PLC-AIo T)combines the low cost and high coverage of PLC with the learning ability of Artificial Intelligence(AI)to provide data collection and transmission capabilities for PLC-AIo T devices in smart parks.With the development of smart parks,their emerging services require secure and accurate time synchronization of PLC-AIo T devices.However,the impact of attackers on the accuracy of time synchronization cannot be ignored.To solve the aforementioned problems,we propose a tampering attack-aware Deep Q-Network(DQN)-based time synchronization algorithm.First,we construct an abnormal clock source detection model.Then,the abnormal clock source is detected and excluded by comparing the time synchronization information between the device and the gateway.Finally,the proposed algorithm realizes the joint guarantee of high accuracy and low delay for PLC-AIo T in smart parks by intelligently selecting the multi-clock source cooperation strategy and timing weights.Simulation results show that the proposed algorithm has better time synchronization delay and accuracy performance.
基金The National Natural Science Foundation of China(No.60773110)the Youth Education Fund of Hunan Province(No.07B014)
文摘In order to implement the real-time detection of abnormality of elder and devices in an empty nest home,multi-modal joint sensors are used to collect discrete action sequences of behavior,and the improved hierarchical hidden Markov model is adopted to Abstract these discrete action sequences captured by multi-modal joint sensors into an occupant’s high-level behavior—event,then structure representation models of occupant normality are modeled from large amounts of spatio-temporal data. These models are used as classifiers of normality to detect an occupant’s abnormal behavior.In order to express context information needed by reasoning and detection,multi-media ontology （MMO） is designed to annotate and reason about the media information in the smart monitoring system.A pessimistic emotion model （PEM） is improved to analyze multi-interleaving events of multi-active devices in the home.Experiments demonstrate that the PEM can enhance the accuracy and reliability for detecting active devices when these devices are in blind regions or are occlusive. The above approach has good performance in detecting abnormalities involving occupants and devices in a real-time way.
文摘With the development of intelligent and netw orking technology in automobile,the malicious attacks against in-vehicle CAN netw orks are increasing day by day,and the problem of information safety in automobile is aggravated. In this regard,this paper analyzes the security loopholes and threats w hich the CAN bus faced,put forw ard a kind of anomaly detection algorithm for vehicle CAN bus. The method uses support vector machine algorithm to distinguish betw een normal message and abnormal message,so as to realize the CAN bus anomaly detection. Theoretical and experimental studies show that this method can effectively detect abnormal packets in the CAN bus w ith a detection rate of over 90%,w hich can effectively resist malicious attacks such as tampering and cheating on the vehicle CAN bus.
文摘Background:The accurate identification of cardiac abnormalities is essential for proper diagnosis and effective treatment of cardiovascular diseases.Method:This work introduces an advanced methodology for detecting cardiac abnormalities and estimating electrocardiographic age(ECG Age)using sophisticated signal processing and deep learning techniques.This study looks at six main heart conditions found in 12-lead electrocardiogram(ECG)data.It addresses important issues like class imbalances,missing lead scenarios,and model generalizations.A modified residual neural network(ResNet)architecture was developed to enhance the detection of cardiac abnormalities.Results:The proposed ResNet demonst rated superior performance when compared with two linear models and an alternative ResNet architectures,achieving an overall classification accuracy of 91.25%and an F1 score of 93.9%,surpassing baseline models.A comprehensive lead loss analysis was conducted,evaluating model performance across 4096 combinations of missing leads.The results revealed that pulse rate-based factors remained robust with up to 75%lead loss,while block-based factors experienced significant performance declines beyond the loss of four leads.Conclusion:This analysis highlighted the importance of addressing lead loss impacts to maintain a robust model.To optimize performance,targeted training approaches were developed for different conditions.Based on these insights,a grouping strategy was implemented to train specialized models for pulse rate-based and block-based conditions.This approach resulted in notable improvements,achieving an overall classification accuracy of 95.12%and an F1 score of 95.79%.
基金supported by State Grid Hebei Electric Power Co.,Ltd.(kj2020-040).
文摘Aiming at the problem of abnormal data generated by a power transformer on-line monitoring system due to the influences of transformer operation state change,external environmental interference,communication interruption,and other factors,a method of anomaly recognition and differentiation for monitoring data was proposed.Firstly,the empirical wavelet transform(EWT)and the autoregressive integrated moving average(ARIMA)model were used for time series modelling of monitoring data to obtain the residual sequence reflecting the anomaly monitoring data value,and then the isolation forest algorithm was used to identify the abnormal information,and the monitoring sequence was segmented according to the recognition results.Secondly,the segmented sequence was symbolised by the improved multi-dimensional SAX vector representation method,and the assessment of the anomaly pattern was made by calculating the similarity score of the adjacent symbol vectors,and the monitoring sequence correlation was further used to verify the assessment.Finally,the case study result shows that the proposed method can reliably recognise abnormal data and accurately distinguish between invalid and valid anomaly patterns.
基金sponsored by the Special Project of China Earthquake Administration(ZX1903006)Earthquake Science Spark Program of China Earthquake Administration(XH16037)Science and Technology Program of Gansu Province(17JR5RA338)。
文摘The deep learning method has made nurnerials achievements regarding anomaly detection in the field of time series.We introduce the speech production model in the field of artificial intelligence,changing the convolution layer of the general convolution neural network to the residual element structure by adding identity mapping,and expanding the receptive domain of the model by using the dilated causal convolution.Based on the dilated causal convolution network and the method of log probability density function,the anomalous events are detected according to the anomaly scores.The validity of the method is verified by the simulation data,which is applied to the actual observed data on the observation staion of Pingliang geoeletric field in Gansu Province.The results show that one month before the Wenchuan M_S8.0,Lushan M_S7.0 and Minxian-Zhangxian M_S6.6 earthquakes,the daily cumulative error of log probability density of the predicted results in Pingliang Station suddenly decreases,which is consistent with the actual earthquake anomalies in a certain time range.After analyzing the combined factors including the spatial electromagnetic environment and the variation of micro fissures before the earthquake,we explain the possible causes of the anomalies in the geoelectric field of before the earthquake.The successful application of deep learning in observed data of the geoelectric field may behefit for improving the ultilization rate both the data and the efficiency of detection.Besides,it may provide technical support for more seismic research.
