34results about How to "Accurate Anomaly Detection" patented technology

The invention discloses an Internet of Things card anomaly detection method and system based on an improved K-means algorithm. The method comprises the following steps: S1, collecting and preprocessing feature data of an Internet of Things card; s2, performing clustering analysis on the preprocessed feature data by using an improved K-means algorithm; s3, calculating a distance threshold value of abnormal judgment according to a clustering result; and S4, judging whether the to-be-detected Internet of Things card is abnormal or not according to the threshold value. The system comprises a data preprocessing module, a clustering analysis module and an anomaly detection module. The invention aims to provide an efficient and intelligent anomaly detection method for the use behavior of the Internet of Things card through an improved K-means algorithm, and by automatically identifying the abnormal use behavior of the Internet of Things card in different application scenes, the accuracy of management and safety monitoring is improved, and malicious abuse and illegal behaviors are reduced.

The invention belongs to the field of time sequence anomaly detection, and discloses a time sequence anomaly detection method and system based on frequency enhanced convolution Transform, and the method comprises the steps: carrying out the sequence-level preprocessing of an input sequence, including normalization and block generation; encoding the input sequence by adopting a multi-head self-attention mechanism; extracting interrelationships in the sequence by a full convolution module having a plurality of stacked 1D convolution layers of different kernel sizes; s2 and S3 are repeated, and multi-layer coding is carried out on the sequence; converting the encoded embedding into a frequency signal by using a Fourier transform method; the KL divergence and the absolute error are respectively used for calculating the consistency loss of the time domain and the frequency domain, calculating the point-by-point anomaly score, and identifying whether the point is abnormal or not. Wide experiments on five common data sets show that the method is superior to an existing method in the aspect of abnormal recognition.

The invention relates to the technical field of power operation and maintenance data processing, in particular to a power operation and maintenance power distribution station intelligent monitoring platform based on artificial intelligence. The platform comprises a data acquisition unit used for acquiring current monitoring data of the power distribution station; the sample space construction unit is used for obtaining a data curve of each dimension and calculating a fluctuation state characteristic value of each dimension, and establishing a sample space based on the fluctuation state characteristic value; the data point suspected anomaly degree calculation unit is used for calculating the spatial distribution anomaly degree and the multi-dimensional similarity relevance of the data points, and the sum of the spatial distribution anomaly degree and the multi-dimensional similarity relevance of the data points is the suspected anomaly degree; the abnormal data point detection unit is used for correcting the initial neighbor quantity parameter of the data point by using the suspected abnormal degree; and detecting abnormal data points in the current monitoring data in the sample space by using the neighbor quantity parameters after data point correction. According to the invention, the power distribution station can be accurately monitored and abnormal data can be acquired.

The invention relates to the technical field of product quality management, in particular to a cable production quality data management system based on Internet of Things big data, and the system comprises an acquisition module which is used for acquiring a production quality sequence; the sample division module is used for screening target data points in the production quality sequence in sequence, and taking the target data points screened each time and adjacent data points thereof as a cluster until all the data points are divided to obtain a plurality of clusters; and the anomaly detection module is used for taking the clustering center of each clustering cluster as an initial clustering center of an ISODATA algorithm, clustering the production quality sequence to obtain a clustering result, and managing the cable production quality based on a result obtained by carrying out anomaly detection on the production quality sequence according to the clustering result. According to the invention, the accuracy of cable production quality management can be improved.

The invention discloses a network link security management method based on artificial intelligence, which relates to the technical field of network security management, and comprises the following steps: when a network receiving link receives a data file of a sending end, obtaining first parameter data of the sending end to generate a sending feature and a fingerprint mark, and performing security verification on the fingerprint mark; and after verification, the data file and the fingerprint mark are packaged to generate a to-be-transmitted file. And when the to-be-transmitted file passes through an intermediate node of the network transmission link, recording second parameter data of the intermediate node, gradually generating a dynamic path mark and inputting the dynamic path mark. And when the file is transmitted to a receiving end, obtaining a complete dynamic path mark, carrying out anomaly analysis on the path mark according to a preset anomaly analysis algorithm, and if an anomaly is detected, generating an anomaly signal and giving an alarm. By sending the fingerprint mark in combination with the dynamic path mark and real-time monitoring, the adaptability of the network environment is improved, and the abnormality in the transmission path is accurately detected.

The invention discloses an intelligent amount and fee auditing method based on an artificial intelligence large model, and the method comprises the steps: S1, carrying out the automatic collection and arrangement of needed data through an intelligent corpus collection technology and a natural language processing technology, and forming a structured and unstructured multi-modal data set; s2, after the multi-modal data set is preprocessed, key features are extracted, and fusion processing is conducted on the multi-modal data set with the key features extracted; s3, through a graph neural network, based on the multi-modal data set, establishing a relation model for the power consumption behavior and the charging relation between the users, and using the relation model to automatically detect and identify the amount and fee abnormity problem; and S4, establishing an auditing rule model, and training the graph neural network output vector and the auditing rule model based on a reinforcement learning method to improve the accuracy of the amount and cost anomaly detection. According to the invention, the problems of insufficient scene adaptability, low data utilization rate and lagging auditing rule and technology update of the existing amount and cost auditing mode are solved.

The invention relates to the technical field of transformer area line loss supervision, in particular to a transformer area line loss intelligent supervision system and method, and the system comprises a high-frequency data collection module, a data preliminary screening module, a line loss recognition and positioning module, a control center, and a load balancing module. According to the invention, the intelligent electric meters and the sensors are installed at each power consumption point, each distribution transformer and each branch line, so that full-coverage power data acquisition is realized, and the data acquisition frequency is improved to a millisecond level; the system can obtain the operation data of the power system in real time; meanwhile, an edge computing device is integrated in the high-frequency acquisition module, data can be processed and analyzed in real time, real-time monitoring of the power system is achieved, and the line loss problem is found and processed in time; denoising and smoothing processing are carried out on data by adopting wavelet transform, main features are extracted by utilizing principal component analysis (PCA), a Gaussian mixture model is trained according to historical data, modeling is carried out on feature vectors, probability density is calculated, and the depth and precision of data processing are improved.

The invention provides a sensor real-time anomaly detection method and system based on LSTM-AE-Bayes, and relates to the technical field of machine learning and deep learning. Current monitoring time sequence data is obtained by obtaining previous normal time sequence data of a sensor to train an auto-encoder and a prediction model, and the current monitoring time sequence data is input into the auto-encoder for reconstruction; the method comprises the following steps: reconstructing data, predicting current data, calculating a reconstruction error value between the current data and the reconstruction data and a prediction error value between the current data and the prediction data, generating a correction factor by combining a Bayesian anomaly probability formula, correcting the formula, calculating an anomaly probability at the current moment, and finally setting an anomaly threshold. And when the abnormal probability exceeds the threshold value, judging that the sensor is in an abnormal state. According to the method, the abnormal state of the sensor is efficiently detected in real time by combining the LSTM prediction model, the self-encoder reconstruction error value and the correction of the Bayesian abnormal probability formula, and the accuracy and reliability of abnormal detection are improved.

The invention discloses a service system user behavior relativity abnormity detection method and system and related equipment, and the method comprises the steps: firstly obtaining user behavior log data from a plurality of service systems, carrying out the preprocessing, and extracting key features corresponding to different service systems; then, corresponding business knowledge is extracted, a knowledge graph is constructed according to the business knowledge, and entities and relationships in user behavior logs of different business systems are mapped into the knowledge graph; then, based on the constructed corresponding user behavior knowledge graph, learning features in the user behavior knowledge graph in a learning mode; then, on the basis of feature values extracted through learning, deep anomaly detection model training is completed in an unsupervised learning mode; and carrying out anomaly detection on user behavior logs collected in real time based on the trained depth anomaly detection model so as to intelligently judge user behaviors corresponding to the logs. According to the scheme, complex user behavior data in a multi-service system can be effectively processed, and efficient and accurate anomaly detection is realized.

The invention discloses a mechanical anomaly detection method and system based on learnable sparse regular dictionary learning, and the method comprises the steps: collecting vibration signals during the normal operation of a machine, and further constructing a training data set; constructing a dictionary learning model containing learnable sparse regular terms, and representing an input signal as a convolution form of dictionary and sparse coding; deducing a model iterative calculation formula by using a semi-secondary splitting algorithm; a corresponding network module is designed for each variable and weight in the formula, the combination of the modules represents one-time iterative calculation, and a plurality of iterative calculation processes are superposed to form a learnable sparse regular network; training the network by using the training data set, extracting normal sample features, and updating network parameters by using a back propagation technology; inputting a test sample into the trained network, calculating an error between the reconstruction data and the input sample, and judging whether the machine is in a normal state or not according to an error value; the reconstruction output of the network is further analyzed, and post-interpretable description of the network is completed.

The invention discloses a data acquisition method based on big data, and relates to the technical field of big data, and the data acquisition method comprises the following steps: obtaining to-be-acquired data; obtaining target collection data; marking target acquisition abnormal data; evaluating the influence degree of the target acquisition abnormal data; and judging whether the target acquisition abnormal data meets a preset data acquisition standard or not. Based on the anomaly recognition algorithm, anomaly detection and marking can be quickly performed on the target data, accurate positioning of potential problem data is realized, the data anomaly processing efficiency can be improved, the abnormal data is compared with the existing database through the evaluation algorithm, the influence degree of the abnormal data can be accurately evaluated, and the accuracy of data anomaly processing is improved. The method shortens the manual judgment time, judges whether the data meets the collection standard or not according to the evaluation result, can effectively eliminate irrelevant or low-value data, guarantees the high correlation and application value of the collected data, and reduces the processing cost of invalid data.

The invention relates to the technical field of anomaly detection, in particular to a medical data anomaly detection method and device, equipment and a medium, and the method comprises the steps that to-be-detected medical data is acquired, and the to-be-detected medical data comprises target text data and target image data; performing feature extraction on the target text data by using a pre-trained first feature extractor to obtain target text vector features; performing feature extraction on the target image data by using a pre-trained second feature extractor to obtain target image features; calculating the similarity between the target text vector feature and the target image feature; if the similarity is smaller than or equal to a preset threshold value, the to-be-detected medical data are abnormal, and if the similarity is larger than the preset threshold value, the to-be-detected medical data are normal. According to the technical scheme, anomaly detection can be accurately carried out on the multi-modal medical data.

The present invention relates to an IT operation and maintenance service management method based on big data, comprising the following steps: S1: collecting data from logs, monitoring tools, and user behaviors, and performing preprocessing; S2: performing anomaly detection on the preprocessed data set, and performing secondary data cleaning; S3: predicting the trend of system performance indicators based on the data after secondary data cleaning; S4: constructing a static user portrait based on user behavior data using hierarchical clustering and cluster analysis, and dynamically updating the user portrait to adapt to the latest behavior data based on path dynamic analysis; S5: optimizing resource allocation and task scheduling using a reinforcement learning algorithm based on the trend of system performance indicators and user portraits; S6: integrating external security intelligence and using an adaptive strategy adjustment mechanism. The present invention not only effectively improves system operation efficiency and reliability, but also improves effective management in an environment of IT operation and maintenance complexity and dynamic changes.

The present invention relates to the field of station monitoring technology, specifically to a smart station monitoring visualization system and method based on multi-source heterogeneous data fusion, the system includes a data integration module, an anomaly detection module, a trend prediction module, a dynamic adjustment module, an intelligent alarm module and a learning optimization module. In the present invention, by using time series analysis and data cleaning, the accuracy and availability of data are improved, thereby ensuring the quality of monitoring data, applying long short-term memory networks to make anomaly detection more accurate, able to identify small data changes, thereby early warning of potential safety risks, through in-depth analysis of trends and probability calculations, can output more accurate future state prediction information, enhance the basis for decision support, dynamic adjustment strategies allow real-time adjustment of monitoring parameters, make resource allocation more flexible and efficient, and the response process adapts to different emergency situations through multi-level trigger conditions, thereby improving the efficiency of emergency management.

The embodiment of the invention provides an air conditioning system and a control method thereof, relates to the technical field of air conditioners, and aims to improve the fault detection precision of the air conditioning system. The air conditioning system includes: at least one processor; and at least one computer memory operably connected to the at least one processor and storing instructions, and when the instructions are executed by the at least one processor, the at least one processor performs the following operations: obtaining real-time operation data of the air conditioning system; on the basis of the multiple pieces of normal operation data and the real-time operation data under the same operation category where the air conditioning system is currently located, a local outlier factor of the real-time operation data is determined, and the local outlier factor is used for representing the outlier degree of the real-time operation data in the multiple pieces of normal operation data; and based on the local outlier factor, fault detection of the refrigerant charging amount is conducted on the air conditioner system, a fault detection result is obtained, and the fault detection result is used for representing whether the air conditioner system is normally charged with the refrigerant or not.

The application discloses a kind of warehouse safety monitoring method and system based on visual detection, it is related to warehouse safety monitoring technical field.A kind of warehouse safety monitoring system based on visual detection, including have: warehouse goods monitoring module and warehouse safety judging module.The application can detect the state of warehouse stacking in real time by stereoscopic modeling analysis and continuous monitoring to warehouse stacking, and improve the sensitivity to small changes through feature enhancement processing, to realize the dynamic monitoring and timely warning to warehouse safety, effectively prevent the loss caused by stacking instability or other safety problems;Feature enhancement processing based on video stream identifies anomaly, can also classify the safety state, provide different risk levels and corresponding safety operation suggestions, ensure that necessary intervention measures are taken in time in warehouse environment.

The invention discloses an intelligent operation leakage analysis system and method applied to a cryptographic device, and relates to the technical field of encryption device.Dynamic thresholds of an abnormal signal and a characteristic signal are obtained through calculation based on a dynamic threshold factor model, and a fluctuation score is obtained through calculation according to the difference value of the abnormal signal and the characteristic signal; a fluctuation score curve graph is drawn, change parameters of each abnormal signal and each characteristic signal are calculated, and an influence factor model is generated in combination with fluctuation scores; the method comprises the following steps: acquiring a real-time side channel signal, calculating to obtain a real-time dynamic threshold based on a dynamic threshold factor model, calculating to obtain a key time point through an influence factor model according to the real-time dynamic threshold, a real-time fluctuation score and a real-time change parameter, and reminding a worker to carry out safety assessment at the key time point. And the timeliness and pertinence of a safety protection strategy are improved.

The invention provides a hydrogen transmission pipe network flow anomaly detection method and system based on two-dimensional characteristics, and the method comprises the steps: carrying out the adaptive slicing in a time dimension and a channel dimension, and obtaining data slices with the same size; a time feature extractor, a space feature extractor and a trunk feature extractor are used for capturing feature association in the slices and between the slices; a reconstruction network formed by multiple groups of hidden layers with different feature scales is utilized to perform reconstruction under multiple views, and anomaly scoring is performed on a reconstruction result so as to realize accurate anomaly detection; according to the method, the abnormal flow data in the industrial hydrogen conveying pipe network can be effectively identified, and the industrial safety management level is improved.

The present invention relates to a method for detecting anomalies in time series data based on a multi-head attention model, and belongs to the technical field of anomaly detection in computer databases. The present invention builds a network framework for detecting anomalies in time series data based on a multi-head attention model, which can process multi-source time series data, has robustness and good generalization ability, and can achieve faster network training and more accurate abnormal sample detection. A frequency domain analysis module is introduced to make full use of the frequency domain information of time series data. By performing frequency domain analysis on the input sequence and the output sequence, the model can obtain effective sequence information. By measuring the distance between the two frequency domain analysis results, the network can be supervised to generate a more accurate output sequence, thereby obtaining a more accurate abnormal judgment threshold and achieving more accurate anomaly detection. The present invention has obvious advantages in the feature extraction capability of time series data and the detection accuracy of abnormal samples.

The invention discloses an industrial product anomaly detection method based on image reconstruction, and the method comprises the following steps: S1, obtaining a to-be-detected image, and loading an image reconstruction model which comprises a feature memory library, a gray contour feature extraction module, an attention module and a jump connection module; s2, inputting a to-be-detected image into the image reconstruction model to obtain a detection reconstruction image; s3, defining an anomaly evaluation function; and S4, comparing the to-be-detected image with the detection reconstruction image through an abnormal evaluation function, and obtaining a detection result. According to the method, gray contour features and normal memory features are extracted and fused, so that high-frequency and low-frequency information of the image in the reconstruction process is more comprehensive and complete; according to the method, an attention module is added on the basis of jump connection, information of an abnormal area is screened, a reconstructed image is more complete, a new anomaly evaluation function is provided, the abnormal area is better positioned, and more accurate anomaly detection is achieved.

The invention relates to an equipment anomaly detection method and device, computer equipment and a storage medium. The method comprises the following steps: acquiring historical traffic data of to-be-tested equipment, and extracting traffic feature data of the to-be-tested equipment based on the historical traffic data; determining a dynamic flow model according to the flow characteristic data; wherein the dynamic flow model comprises a plurality of flow modes, and each flow mode corresponds to one test model; according to the dynamic flow model, dynamically adjusting a test model corresponding to the flow data of the to-be-tested equipment in the test time period; and testing the traffic data of the to-be-tested device in the test time period according to the plurality of test models to obtain an abnormal test report of the to-be-tested device. By adopting the method, the equipment in a complex operation scene can be accurately detected, and the equipment anomaly detection accuracy is improved.

The invention discloses a hearth water wall anomaly detection and identification method based on deep convolutional neural network learning, and relates to the field of hearth water wall anomaly detection. In order to solve the technical problems of inaccurate identification and inaccurate classification of detection, identification and classification of the water cooling wall of the cooling pipe in the prior art, the technical scheme provided by the invention is as follows: the method comprises the following steps of: obtaining original pipeline data shot at each pipeline position in a hearth water cooling wall system of a thermal power plant; preprocessing the original pipeline data, and dividing the original pipeline data into a training data sample and a test data sample; respectively carrying out hybrid enhancement processing on the training data sample and the test data sample; and inputting the training data sample subjected to hybrid enhancement processing into a deep convolutional neural network for model training to obtain a trained model. The method can be applied to abnormal detection and identification of the hearth water wall.

The application discloses a data anomaly detection method based on intelligent industry, a storage medium and a terminal, belongs to the technical field of anomaly detection, and various industrial data are processed by using an improved KF algorithm and then input into a first neural network model for anomaly prediction. The improved KF algorithm specifically learns the change trend of the covariance matrix by using a second neural network model, and then updates the covariance matrix. The improved KF algorithm can dynamically adjust the process excitation noise covariance matrix Q, thereby reducing the influence of noise on the KF algorithm, and then ensuring the anomaly prediction accuracy and reliability of the subsequent neural network model. Meanwhile, the improved KF algorithm can process the industrial data, eliminate redundant data, uniformly process each information source, and thus ensure the measurement accuracy. Meanwhile, the improved KF algorithm can also determine the time sequence characteristics of the industrial data stream, and thus the method can accurately detect the data stream without time sequence characteristics.

The present invention discloses a forest fire image detection method based on a generative adversarial network, and the steps are as follows: step 1: obtain forest fire monitoring images to form a data set; step 2: extract the image feature matrix of each image in the data set; step 3: construct the dictionary and classifier parameters in the discriminator and generator through deep dictionary learning; step 4: use the image feature matrix as the input of the discriminator, obtain the dictionary and sparse matrix of each layer in the discriminator; obtain the dictionary and generated image of each layer in the generator; step 5: according to the objective function, repeat step 4 until the dictionary, sparse matrix and generated image are stable; step 6: use the stabilized generative adversarial network to detect forest fire. The present invention constructs the dictionary and classifier parameters in the discriminator and generator of the generative adversarial network through deep dictionary learning, and the constructed generative adversarial network can more effectively identify anomalies, making anomaly detection more accurate.

The invention provides a networked radar anomaly detection method and system based on a graph neural network, and belongs to the technical field of Internet of Things security, and the method comprises the steps: completing a target track route detected by a radar according to the recording of a track detected by the radar for a target, and determining a track result observed by each radar in the networked radar; calculating the attention value of the track observed by each radar node, adaptively determining an adjacent matrix between the nodes, and taking the adjacent matrix as one of the characteristics of the networking radar; predicting the node condition at the next moment through a graph neural network according to the observation result of each node in the networking radar and the adjacent matrix between the radar nodes; and comparing the node track information predicted by each radar node at the next moment with a track result observed by the radar at the next moment so as to judge the radar node in an abnormal state in the networking radar. According to the method, the sensor in the abnormal state in the Internet of Things can be rapidly detected in real time, the robustness is relatively high, and the accuracy is relatively high.

The invention discloses a charging service platform data analysis method and system based on the Internet of Things, and relates to the technical field of data analysis, and the charging service platform data analysis method comprises the following steps: extracting key operation feature data of a charging service platform; identifying abnormal operation data in the current charging service platform; key factors influencing the operation fault of the charging pile equipment are identified; and dynamically adjusting the operation parameters of the charging pile equipment, and optimizing the operation of the charging service platform in combination with the real-time monitoring data. According to the method, key operation characteristic data are analyzed through a fault diagnosis algorithm, abnormal data in the platform are quickly identified, the accuracy and response speed of fault detection are remarkably improved, and service interruption caused by faults is reduced.

The invention belongs to but is not limited to the technical field of electrocardiogram anomaly detection, and discloses an ST segment anomaly detection method based on inversion Transform and fusion attention, which adopts a median filtering method to carry out de-noising processing on an input data sample, and carries out data amplification on abnormal ST data; and classifying the electrocardiosignals by using a convolutional neural network model which fuses an inversion Transform and a channel prior attention mechanism. According to the method, the features can be effectively extracted and fused by utilizing the inversion Transform and the time sequence channel convolution, and the model can keep related time sequence information while accurately capturing the spatial correlation of each dimension in the electrocardiosignal in combination with a channel prior attention mechanism, so that the accuracy of ST segment anomaly detection is improved.

The present application belongs to the technical field of intelligent video processing, and particularly relates to a future frame anomaly detection method based on meta learning and space-time relationship. The present application method proposes a meta learning module, and a model-based meta learning method enables the model to learn general features from multiple tasks, so that the system can learn more discriminative and generalizable feature representations from data. In different monitoring scenes and abnormal behaviors, the model can also accurately perform anomaly detection without overfitting or underfitting. The present application method introduces the meta learning module into the autoencoder, uses the learning characteristics of the meta learning module to extract, save and update the features extracted by the autoencoder, automatically acquires the feature importance of the input video frame through learning, and assigns important weights to the features that are more worthy of attention, which helps to improve the utilization rate of key features in the input stage of the future frame prediction network.

The invention provides a rapid detection method and system based on big data, and the method comprises the steps: d1, carrying out the first data acquisition, collecting the primary multi-source data of a current image, and collecting the primary multi-source data of a similar image from a big data platform based on the primary multi-source data of the current image; d2, performing first round of data processing, preprocessing primary multi-source data in the current image and the similar image, sorting the preprocessed data into a first data set and a second data set, performing integration calculation on the first data set and the second data set to generate an abnormal coefficient YCZ, and performing second round of data processing on the abnormal coefficient YCZ; and comparing the abnormal coefficient YCZ with a preset first threshold Y to generate a first comparison result, and judging whether the first image is abnormal or not according to the first comparison result. According to the method, the abnormal features in the image can be captured more comprehensively, the data processing efficiency is improved, and the accuracy and the real-time performance of abnormal recognition are enhanced.