9results about How to "Improve anomaly detection accuracy" patented technology

A wind turbine gearbox anomaly detection method considering the similarity of multi-unit operating states, based on the piecewise linear time series similarity evaluation method, evaluates a single state variable; considering the similarity of multiple state variables, the wind turbine operating state spatiotemporal similarity quantification method evaluates the similarity of each wind turbine and the operating state of the unit to be detected; a state estimation LSTM model is constructed, different LSTM models are trained; the accuracy and adaptability of different LSTM model performances are verified; select several LSTM models that perform well in performance verification, construct a combined state estimation model CPEM through weighted combination; estimate the target variable of the unit to be detected, calculate the residual error between the estimated value and the true value, and based on the residual effective value comparison and residual information entropy comparison, identify the anomaly. This method detects the operating state of the gearbox based on the historical operating data of the wind turbine, without the need for maintenance personnel to conduct on-site detection, reducing the operating and maintenance costs of the wind turbine.

This invention relates to a method and related equipment for detecting abnormal object behavior. The method includes: acquiring a behavior log stream and mapping the logs to feature vectors including object identifiers; for any object, generating a behavior state sequence based on the feature vectors, constructing reachable links between state points based on the distance relationships between state points in the behavior state sequence, thereby constructing a topological structure of the object's behavior state space, calculating the topological closure of the object's historical state vector set under the topological structure, and using the topological closure as the normal behavior baseline region of the object; wherein, the behavior state space consists of multiple state points obtained by mapping state vectors from the behavior state sequence; acquiring the target state vector corresponding to the behavior of the object to be detected, determining whether the target state vector belongs to the normal behavior baseline region of the corresponding object, and if not, determining it as abnormal behavior. This method at least partially solves the problems of high storage overhead and poor anomaly detection effect in related technologies.

The application provides a kind of satellite telemetry multidimensional time series data-oriented anomaly detection method, device, medium and product, it is related to satellite telemetry multidimensional time series data anomaly detection field, method includes: the satellite telemetry multidimensional time series data to be detected is input into anomaly detection model, and the fusion error of each time is output;Anomaly detection model includes denoising sparse auto-encoder and graph attention network;Denoising sparse auto-encoder learns the low-dimensional feature of input satellite telemetry multidimensional time series data, reconstructs satellite telemetry multidimensional time series data, generates reconstruction error;Graph attention network extracts the correlation between different reconstructed satellite telemetry multidimensional time series data from two levels of causality and similarity, generates prediction error;Generate the fusion error of each time, judge whether the satellite telemetry multidimensional time series data to be detected exists anomaly;If yes, record abnormal time point.The application can be suitable for satellite telemetry multidimensional time series data of different dimensions, improve anomaly detection precision.

The invention provides an AI-based intelligent inspection method and system, and the method comprises the steps: collecting environment data, poultry behavior data and equipment operation data, forming a multi-dimensional fusion data set containing time-space coordinates, and constructing a digital twinborn model of a farm; updating the digital twinborn model and establishing a virtual-real mapping relation; inputting real-time data output by the digital twin model into a pre-trained multi-task learning AI model to obtain a comprehensive risk assessment report containing confidence and a time window; according to the comprehensive risk assessment report, an ant colony algorithm is combined with a real-time risk thermodynamic diagram to dynamically generate an optimal inspection path, and the inspection frequency and key areas are adjusted in real time; new data and processing results obtained in the inspection process are fed back to update parameters of the digital twinborn model and the multi-task learning AI model, and an inspection strategy is continuously optimized through a reinforcement learning algorithm. According to the method, the anomaly detection accuracy can be improved, the loss is greatly reduced, and the labor cost and poultry stress response are reduced.

The application discloses an agricultural process big data statistical analysis and alarm method, collects meteorological, soil, insect, irrigation state and equipment working condition data in agricultural production, carries out cleaning, time alignment and unified modeling on multi-source data, forms a monitoring sequence, carries out standardization and rank statistical processing on the monitoring sequence, identifies candidate abnormal points, further combines local significance test and interval distribution change test, confirms abnormal intervals, and then comprehensively scores mutation strength, significance strength and distribution change strength, generates early warning grades, early warning time and related index information, and outputs alarm results. The application realizes statistical analysis and graded early warning of abnormal changes in agricultural processes.

The application provides a kind of vehicle-mounted full work health monitoring system, method and computer program product. Including: physiological state perception module fuses the fusion feature vector of each channel physiological signal of occupant, obtains basic physiological index by time series multi-task regression model, calculates fatigue discrimination result, autonomic nervous balance index, combined with basic physiological index, obtains emotional stress level by single task classification model;Abnormal feature acquisition module when physiological state index, physiological signal deviates from preset benchmark, respectively quantitatively obtained first, second abnormal feature;Genetic risk coefficient calculation module calculates the PRS value of occupant;Disease risk probability calculation module combines disease diagnosis rule with abnormal feature, genetic risk coefficient, input probability type classification and survival analysis model obtains two kinds of probability and is fused into disease risk probability.The application accurately perceives the physiological state of occupant, establishes individualized physiological baseline to improve the precision of abnormal detection, and realizes the accurate prediction of disease risk in two dimensions.

The application discloses a kind of data imbalance weak supervision video anomaly detection method and system, including extracting the video feature of the video segment level of the video to be measured;Video segment level video feature is obtained by the segment level anomaly score of the confrontation training module;Video segment level video feature is obtained by the anomaly score of the segment level of the focusing training module;The anomaly score after fusion is obtained by fusing segment level anomaly score and anomaly score;Again compared with threshold value, all frames in the video segment greater than threshold value are regarded as anomaly, and all frames in the video segment less than threshold value are regarded as normal, to realize data imbalance weak supervision video anomaly detection and the time positioning of anomaly.The anomaly score of segment level obtained by fusing confrontation training module and focusing training module makes it easier to learn the difference between normal and abnormal video segments, and the detection of abnormal events is closer to reality, thereby improving the accuracy of video anomaly detection.

This invention discloses a method for abnormal traffic detection and root cause localization in communication networks based on graph neural networks. The method includes: unifying the network hierarchy, port location, and forwarding adjacency relationships of communication network devices through hierarchical port indexing, and merging communication session features, forwarding adjacency features, and communication connection state features to form a port session matrix, expanding the anomaly analysis object from single traffic statistics to a session state expression under port connection relationships; generating an anomaly propagation phase matrix by comparing upstream and downstream sequence port mutation locations, enabling the identification of the propagation sequence of abnormal traffic between ports; eliminating overlapping phase segments caused by routing path switching using a scheduling phase matrix, reducing false alarms caused by normal scheduling disturbances; and performing root cause localization by combining a physical precedence matrix with a candidate propagation matrix, distinguishing between physical root causes of communication connections and root causes of traffic behavior, thereby improving the accuracy of anomaly detection and the precision of root cause localization.

The application provides an anomaly detection method, device and storage medium for image data, and relates to the technical field of image processing. The method comprises the following steps: performing feature extraction on image data, fusing spectral and spatial features to obtain a joint feature matrix, and inputting the joint feature matrix into an anomaly detection model; the model uses an alternating direction multiplier algorithm to solve a low-rank sparse decomposition problem, and a target function comprises a data fidelity term, a regularization term and a band weight term; the regularization term comprises a low-rank constraint and a sparse constraint, and the band weight term acts on the low-rank constraint in a weighted form; in the solving process, an iteration method is used to update a background low-rank tensor, an anomaly sparse tensor and a Lagrange multiplier, as well as a sparse constraint weight, a band weight term and a penalty parameter of the algorithm; the iteration is repeated until a preset termination condition is reached, an anomaly score map is calculated based on the anomaly sparse tensor, and an anomaly target is determined by comparison. The application can solve the problem that it is difficult to accurately identify an anomaly target in a complex scene, and improve detection accuracy and efficiency.