183 results about "Data repair" patented technology

The invention discloses an electric meter historical data restoration method based on a multi-dimensional incidence relation, and relates to the technical field of electric power data management and intelligent restoration, the accuracy and traceability of data restoration are realized by establishing a multi-source exception type library and an element time-varying degradation model, and the accuracy and traceability of data restoration are improved for multi-source data conflicts. The method comprises the following steps: pre-defining a plurality of abnormal mode combinations and corresponding repair strategies, positioning a fault source through mode matching of a real-time index and an abnormal type library, selecting reference data in combination with a historical average deviation rate of a target ammeter and an associated ammeter, avoiding deviation caused by single judgment, and for nonlinear degradation, firstly determining fault elements of a shunt and an ADC chip, a resistance degradation model and a conversion error model are constructed, a theoretical correction value is calculated, a result is optimized through a machine learning model trained by historical normal data, meanwhile, original readings, model parameters and calculation bases are stored in an associated mode, a traceable log is formed, and the repair accuracy is guaranteed.

The invention discloses a sewage treatment time series data missing value interpolation method based on a random configuration network. The method comprises the following steps: collecting water quality data in a sewage treatment process, and preprocessing the water quality data to obtain univariate time series data; performing seasonal trend decomposition on the single-variable time series data to obtain a seasonal component, a trend component and a residual component; respectively performing component interpolation on the season component, the trend component and the residual component; and performing addition reconstruction on the seasonal component, the trend component and the residual component after interpolation to obtain complete sewage treatment process monitoring data. According to the sewage treatment time sequence data missing value interpolation method based on the random configuration network, original data is decomposed into season, trend and residual components, interpolation is conducted on all the components through self-similarity decomposition and a random configuration network model, finally a complete sequence is reconstructed, the periodicity and trend characteristics of missing data can be effectively recovered, and the accuracy of the missing data is improved. And the restoration precision and the application reliability of the sewage treatment monitoring data are improved.

The invention relates to a data recovery method and device, computer equipment, a storage medium and a program product. The method comprises the steps that a current repairing task of a hybrid tree is determined, the hybrid tree is a hybrid storage structure of a log merging tree LSM tree and a B tree, the hybrid tree comprises multiple layers, the multiple layers comprise the first layer of the LSM tree and all layers of the B tree, ordered character string tables of the first layer of the LSM tree are merged and sunk into the B tree, and each node in the B tree stores at least one ordered character string table obtained through merging and sinking; under the condition that the current repair task indicates that the repaired damaged data storage object belongs to the B tree, obtaining a B tree index of the damaged data storage object from the current repair task; searching a damaged data storage object from the mixed tree according to the B tree index, wherein the damaged data storage object is an ordered character string table or a node; and under the condition that the damaged data storage object is found, repairing the damaged data storage object. By adopting the method, the data recovery efficiency can be improved.

The invention provides a sensor data recovery method and system based on mask perception space-time modeling, and belongs to the technical field of Internet of Things and intelligent sensing data processing. The method comprises the following steps: modeling a missing position into a trainable potential vector representation through an adaptive missing representation module, and avoiding noise introduced by traditional zero filling or mean filling; a missing perception space-time decoding module is adopted, and a dynamic weight distribution mechanism of mask constraint is combined in the decoding process, so that information is effectively prevented from being excessively smooth; designing a space-time dual-channel feature aggregation unit, and capturing spatial dependence and time sequence dependence between the sensors at the same time; and finally realizing accurate completion of a large-scale space-time sensor matrix through an output recovery unit. According to the method, the accuracy and robustness of sensor data restoration can be remarkably improved, the reliability of subsequent monitoring, prediction and anomaly detection is enhanced, and the method has wide engineering application value.

The invention discloses a BDDR backdoor detection and data recovery method and system oriented to a large model, and belongs to the field of backdoor defense. Comprising the following steps: constructing a knowledge distillation architecture under federal learning, including an edge server and a plurality of clients, and obtaining distillation data; inputting the distillation data into a randomly initialized model for training, recording the loss change of each batch of data, and screening out abnormal batches to form a backdoor data set; the edge server initializes two independent models, respectively uses a distillation data set and a backdoor data set for training, and guides learning of backdoor features; using probability distribution to calculate and correct a backdoor label, generating a clean data set by adding noise, finely adjusting a large model, detecting residual backdoor feature intensity, and adjusting probability distribution calculation parameters to further weaken backdoor features according to the residual backdoor feature intensity so as to obtain a final repaired data set; while the generalization ability of the large model is improved, backdoor attacks can be effectively identified and defended, the data privacy of the client is protected, and the model security is ensured.

The invention relates to the technical field of wind energy development and utilization, in particular to a multi-source meteorological data dynamic fusion wind power plant simulation method, which comprises the following steps of: accessing satellite remote sensing, a ground observation station, numerical forecasting and wind power plant historical operation data, and outputting standardized meteorological data through format analysis, standardized processing and cleaning; a credibility score is calculated based on multi-dimensional indexes such as data integrity and time consistency, and abnormal data restoration or standby data source switching is achieved; in combination with wind power plant DEM topographic features and a self-adaptive sliding window, the weight of a data source is adjusted in real time through reinforcement learning, and a dynamic weight value is generated; an improved Windpowerlib framework adapts to fusion data, fan parameters are dynamically configured, and an initial simulation result is generated; and collecting real-time data of the wind power plant, and correcting model parameters by using Kalman filtering. The wind power plant simulation precision and timeliness can be improved, the data utilization rate and the model adaptability are enhanced, and the deviation between the simulation result and the actual working condition is reduced.

The invention discloses a traffic flow state data time gathering method based on Gaussian mixture probability, and belongs to the technical field of calculation, reckoning or counting. The method comprises the following steps: acquiring traffic flow state data and vehicle trajectory data of each lane in a target road section; carrying out probability diagnosis on the traffic flow data by adopting a Gaussian mixture model, and identifying and deleting abnormal points; introducing a model based on Transform and graph neural network fusion, and realizing space-time repair of abnormal data through joint modeling of an attention mechanism and a space-time topology dependency relationship; and carrying out time window collection on the repaired data based on wave walking and stopping prediction, and calculating a traffic flow rate, an average speed and an occupancy rate. According to the method, the joint characteristics of the traffic flow state data and the trajectory data are fully utilized, meanwhile, the abnormal data repairing precision and the space-time consistency are improved, adaptive collection of the traffic operation dynamic characteristics is achieved by introducing walking and stopping wave prediction, and the robustness and reliability of traffic flow data processing are remarkably improved.

This invention discloses an artificial intelligence-based method for managing and controlling power grid water dispatch data, specifically relating to the field of power grid water dispatch data management technology. The method involves collecting multi-source heterogeneous water dispatch data and its metadata to construct a unified format dataset; generating a credibility score based on metadata and data features to form a credibility matrix; constructing a multi-dimensional quality assessment tensor and extracting the dominant factor subspace; inputting this tensor into a clustering perceptual neural network to identify abnormal data, outputting an abnormal dataset and abnormal labels; combining a large model enhanced by RAG retrieval with historical similar data to complete abnormal data repair; constructing a data quality improvement report using the difference vectors before and after repair and label information; and finally pushing the report to a data management platform and generating a traceable rectification work order. This method achieves automatic identification, intelligent repair, and feedback tracking of power grid water dispatch data, improving data credibility, governance efficiency, and security, and is suitable for data management needs in complex scenarios under new power systems.

The application provides a ground target threat evaluation method in an uncertain environment, and the target feature data is repaired by using an improved autoregressive model for the information missing problem of the ground target according to the acquired target situation information; the target threat data sample is generated by using the explainability of the Bayesian network for the problems existing in the current threat evaluation method, and the target threat evaluation model is established by using the nonlinear fitting ability of the BP neural network, so that the final ground target threat level evaluation result is output. The application can efficiently process data missing, solve the problems of threat evaluation sample shortage and model precision, and also integrates the advantages of the model, considers the performance and explainability, is more suitable for the time sequence characteristics and dynamic changes of the ground target data compared with the traditional method, improves the accuracy and robustness of data repair, and provides more complete and reliable basic data for subsequent evaluation.

This invention discloses a method, device, system, and storage medium for longitudinal protection of distribution networks, including: 1) extraction of key response data strongly correlated with line protection and repair of abnormal data; 2) a longitudinal protection method based on Hellinger distance; and 3) a distribution network longitudinal protection method based on abnormal data repair and Hellinger distance. This application uses cubic trigonometric Hermite interpolation with shape parameters to identify and repair fault current data, effectively solving the problem of protection maloperation or failure to operate due to data loss or distortion, and improving the data reliability and adaptability of protection under complex operating conditions.

The application discloses a data repairing method, comprising: segmenting an RGBW image according to point cloud data of a workpiece to be measured, the RGBW image and a preset segmentation algorithm, determining a plurality of target regions and a data missing region; fitting a zero plane according to the point cloud data and a least square method, and converting the point cloud data to the zero plane to obtain converted point cloud data; determining point cloud data located around each point in the data missing region in the converted point cloud data, and determining an effective value of each point; repairing the data missing region according to a positional relationship between the data missing region and the plurality of target regions, an angle value and the effective value of each point in the data missing region, and obtaining repaired point cloud data of the workpiece to be measured. Thus, corresponding repairing modes are collected for data missing regions located in different target regions, so that the repaired point cloud data of the workpiece to be measured can better restore the actual height of the workpiece to be measured.

This invention relates to the field of data monitoring technology, and more particularly to an enterprise data monitoring system based on big data analytics, used to solve the problems of incomplete, inaccurate, and duplicated data from multiple channels. It includes a data acquisition module to determine the data missing rate and data consistency; a data diagnosis module to determine data categories to differentiate data analysis methods; a data analysis module to calculate data anomaly characteristics, classify data anomaly trends to determine data repair methods, or mark usable data; a data repair module, for data with strong anomaly trends, dividing the data into several anomalous text data frames, predicting the anomalous text data frame groups, and verifying the predicted anomalous text data frame groups; for data with weak anomaly trends, using a trained convolutional neural network for repair, determining successfully replaced data; and a data processing module to determine data availability and mark usable data.

This invention discloses a real-time data anomaly monitoring and repair method based on machine learning, comprising the following steps: S1, constructing data fragments and generating a consistent constraint set; S2, performing self-consistent representation construction; S3, performing anomaly detection and generating anomaly label set; S4, performing masking processing to generate masked data fragments, inputting them into an improved UniTS model, introducing a structural perturbation suppression mechanism, and generating a repair candidate set; S5, performing structural consistency evaluation and screening target repair data that meets the minimum structural perturbation condition; S6, performing data repair and generating repaired data fragments; S7, feeding back the updated consistent constraint set to the improved UniTS model. This invention achieves highly sensitive real-time identification of anomalous data items, effectively reducing the anomaly false judgment rate and repair distortion rate, and improving the data security, stability, continuity, and reliability in high-concurrency data stream environments.

Embodiments of the present application provide a data storage method and device, a control center receives a data storage request; the control center determines a first storage node storing a master copy from each node array of an optimal node array set, and determines N second storage nodes storing a copy from each node array having the first storage node; wherein each storage node in the same node array is located in different racks, and the optimal node array set is a grouping result of array pair statistical values being minimum in T times of random grouping of each storage node; the array pair statistical value is used to count the number of array pairs in which the number of same storage nodes in each grouping is greater than a set threshold; and the control center stores data corresponding to the data storage request to the first storage node and N second storage nodes respectively. The above method can reduce the data loss rate while ensuring the data repair speed.

The invention relates to a data restoration method and device for a data warehouse, equipment, a storage medium and a program product. Comprising the steps of receiving structured data input by a user; comprising to-be-repaired field information and screening conditions; generating an SQL statement according to the structured data and a preset SQL template; carrying out pre-verification on the SQL statement; after the pre-verification is passed, determining a to-be-repaired data magnitude corresponding to the SQL statement; if the magnitude of the to-be-repaired data is smaller than or equal to the preset magnitude, calling a first number repairing strategy, and executing ACID transaction operation according to the SQL statement so as to realize data repair for the data warehouse; and if the magnitude of the to-be-repaired data is greater than the preset magnitude, calling a second repair strategy, and performing data repair on the data warehouse by adopting an operation of overwriting data according to the SQL statement. By adopting the method, the number correction logic is automatically generated, misoperation is avoided, the labor cost can be saved, the repeated data correction probability can be reduced, and the situation that supervision early warning is triggered due to frequent rereporting is avoided.

The utility model discloses a water turbine unit frequency modulation AGC collaborative control device based on multi -sensor fusion, including multi -sensor acquisition module, data fusion processing module, frequency modulation AGC collaborative control decision module, execution drive module and state monitoring and feedback module, the signal output of sensor acquisition module is connected with the signal input of data fusion processing module through the cable. The utility model discloses, through setting each way sensor of pertinence selection type distribution point, guaranteeing the original collection data precision reliable of key parameters such as main shaft speed, pressure, the problem of traditional sensor data precision deficiency has been solved. Meanwhile through the time and space alignment and abnormal data repair mechanism when data fusion processing, in combination with the differentiated processing mode of dynamic static data, let unit state data be more close to actual operation situation, solved the problem that multiple source data is not consistent.

This invention proposes a day-ahead spot clearing price-load data repair method based on generative adversarial networks. A training dataset is established; K-means clustering is used to cluster the user electricity load dataset, classifying it into typical user groups; a deep convolutional generative adversarial network (DCGAN) model is established; the DCGAN model is trained using historical day-ahead spot market clearing price data and historical user electricity load data after classifying typical user groups, retaining the generator structure for each typical user group; a binary mask matrix is ​​introduced to represent the missing locations of newly collected user electricity load data; then, the newly collected day-ahead spot market clearing price dataset, typical user electricity load dataset, and target user electricity load data with missing data are input into the generator corresponding to the typical user group, outputting new generated data; the missing parts of the target user electricity load data are filled in using the new generated data samples.

The invention discloses a data remote management system and method of a glycosylation monitor, and relates to the field of data management.The data remote management system comprises a data management module used for monitoring the data state of blood glucose data in the transmission process, and the data management module comprises a data collection unit, a data transmission unit, a data tracking unit and a suspension state data recognition unit; the constraint mapping module is used for performing constraint mapping on the blood glucose data, and the constraint mapping module comprises a constraint space modeling unit, a constraint drift detection unit and a data restoration and adjustment unit; and the constraint consensus module is used for performing consistency processing on the blood glucose data, and the constraint consensus module comprises a state synchronization unit, a convergence control unit and a consensus adjustment unit. By introducing a data tracking unit, a suspended state data identification unit and a dynamic adjustment mechanism, intermediate state data can be tracked through an independent state chain, and suspended state identification and marking are performed on the data, so that error analysis caused by the intermediate state data is effectively prevented.

The invention relates to the technical field of postoperative neck deformation monitoring, in particular to a multi-modal fusion postoperative neck deformation monitoring and grading alarm system which comprises a multi-modal data acquisition module, a data processing module, a grading alarm module, a display module and a power management module. The system acquires data through multi-mode sensors such as an ultrasonic detection array and a pressure sensing matrix, and combines the functions of real-time correction, abnormal data restoration and dynamic adjustment, so that the monitoring precision and the alarm accuracy are remarkably improved. The grading alarm module generates an alarm signal by adopting a support vector machine algorithm, and the display module realizes multi-view synchronous display. According to the application, the safety and efficiency of postoperative neck deformation monitoring can be improved, and reliable technical support is provided for clinical nursing.

The invention provides a hazardous chemical substance transportation accident liquid leakage detection method and system. When image monitoring data of a hazardous chemical substance transportation accident is repaired, an optical image data repairing scheme of multi-scale feature association and physical motion constraint is provided, vibration motion estimation and visual displacement estimation are carried out on a hazardous chemical substance transportation vehicle, joint optimization is carried out, and a final motion track of the hazardous chemical substance transportation vehicle is obtained. Carrying out image segmentation on the blurred frame optical image to obtain different regions, and carrying out blurred kernel generation on the different regions; and according to the blurring kernels of different areas, layered restoration of optical flow guidance is carried out on the blurred frame optical image, and high-precision optical image data restoration is realized.

The application discloses an enterprise data sharing optimization method based on a local area network and relates to the technical field of data transmission; cross-enterprise distribution information and enterprise internal distribution information are acquired, and cross-enterprise storage networks and enterprise internal storage networks are constructed, respectively; enterprise internal storage and cross-enterprise storage are performed on enterprise storage data; the storage state of the enterprise storage data is acquired; data repair is performed on the enterprise storage data by using the cross-enterprise storage networks; the sharing request of a user is acquired, and the request type is judged; for enterprise internal data sharing, enterprise internal sharing receipts are generated for the user; enterprise internal data sharing is performed by using the enterprise internal sharing receipts; for cross-enterprise data sharing, cross-enterprise sharing receipts are generated for the user; and cross-enterprise data sharing is performed by using the cross-enterprise sharing receipts; and the technical scheme has the advantages that the efficiency of cross-enterprise data sharing and the security of internal data sharing are improved.

The invention provides a method for establishing a marine image data restoration large model based on spatial-temporal characteristics, and belongs to the technical field of marine image data restoration large models.The method comprises the steps that sampling is conducted through a sliding window, a sample set containing spatial-temporal information is established, normalization processing is conducted on samples, training data are formed through a random missing method, and the training data are obtained; a data restoration network model based on a Swin Transformer encoder and a CNN decoder is constructed to realize spatio-temporal feature extraction and data reconstruction, a missing data mean square error loss function is calculated by using a mask mechanism, and model training is performed by using an Adam optimizer and a cosine annealing hot restart method. The model effect is verified through the verification set and the test set to prevent over-fitting, and the optimal hyper-parameter is determined to form the pre-training ocean data restoration model, so that the technical problem of low data restoration precision caused by insufficient ocean data space-time integration feature extraction is solved.

The application relates to a special variable user load feature extraction method, device, equipment and medium based on multi-dimensional time sequence data. The method comprises the following steps: preprocessing an original data set collected to obtain a reference data set, obtaining a to-be-repaired data point in the reference data set, respectively obtaining a spatial repair value and a time repair value of the to-be-repaired data point, fusing the spatial repair value and the time repair value based on a fusion weight to obtain a data repair value, updating a data value corresponding to the to-be-repaired data to the data repair value to obtain a repaired data set, performing time sequence alignment on a load repaired data set and an environment repaired data set to obtain a multi-dimensional time sequence data set, extracting a time sequence feature set of the multi-dimensional time sequence data set, screening the time sequence feature set to obtain a target feature set, and constructing a special variable user load space-time feature sample library based on the target feature set. The method can improve data processing precision.

This application discloses a method and apparatus for intelligent driving data trajectory repair based on segmented point cloud registration, relating to the field of autonomous driving technology. The method includes the following steps: dividing the data segment to be repaired into multiple smaller data segments; registering each frame of point cloud data in a smaller data segment with the first frame of point cloud data in that smaller data segment to obtain a data segment repaired within the segment; and registering the point cloud data of adjacent data segments repaired within the segment on the time axis to obtain point cloud data repaired between segments. Through partial-to-global joint trajectory correction, data repair can be achieved without relying on additional hardware upgrades or external auxiliary data. This significantly improves global trajectory consistency and repair success rate, efficiently repairing trajectory deviations in existing data, avoiding data resource waste, and adapting to the repair needs of newly added data, thus greatly reducing data processing costs.

This invention relates to the field of data monitoring technology and discloses a method for real-time monitoring of the operating status data of on-site equipment in the production industry. The method includes: acquiring production-level master data; constructing a regional list table and a collection point list table; establishing a dynamic mapping mechanism to filter the effective collection point set; completing adaptive frequency conversion acquisition and data quality verification; generating standard data messages and sending them to a message queue; caching the effective information from the regional list table and the collection point list table to edge-side memory; differentiating anomaly types to complete differentiated data repair and full standardization processing; and encapsulating the processed data messages; performing dual-dimensional anomaly identification combining fixed absolute safety thresholds and adaptive dynamic thresholds based on operating conditions; implementing hierarchical control and root cause convergence for identified alarms; completing the closed-loop binding of the entire process from alarm triggering to handling and archiving; providing feedback to optimize the core configuration and processing rules of the first three steps; and updating the collection point list table to build an equipment fault knowledge base.

The invention belongs to the technical field of encrypted wallet damaged data recovery systems, and particularly relates to a zero-knowledge proof encrypted wallet damaged data recovery system which comprises a distributed key fragment storage module, a zero-knowledge proof generation and verification module, a data recovery engine and a privacy protection interaction module. The distributed secret key fragment storage module is used for splitting a secret key for encrypting a wallet by a user into a plurality of fragments and dispersedly storing the fragments in different nodes, a threshold value k is set, secret key reconstruction can be carried out only when not less than k secret key fragments are collected, and the zero-knowledge proof generation and verification module is used for generating and verifying a data recovery request when the user initiates the data recovery request. A user generates a zero-knowledge proof to prove the ownership of the user to the encrypted wallet, the system verifies the zero-knowledge proof, and after the verification is passed, a subsequent recovery process can be entered.

The application provides a kind of spatio-temporal intelligent marine satellite internet of things sensing information screening method and device, belong to data processing technical field, this method is based on convolution bidirectional long short-term memory neural network realizes marine environment time series data repair, based on block time series transformer carries out marine environment data space-time prediction and based on mean shift density clustering algorithm carries out marine hotspot area mining, the marine environment data includes sea surface temperature, salinity, wave height and other sensing data, based on BP neural network carries out marine environment risk assessment and multi-agent reinforcement learning carries out information screening feedback control, meet the complex marine environment detection scene efficient information processing demand, realize low redundancy, high-precision data acquisition transmission adaptive feedback control effect.The present application can be directly applied to marine internet of things in marine observation detection facility networking system, can provide strong support in military application and civil application, has wide and important application prospect and value.

This invention relates to the field of distributed storage technology, specifically to a distributed audio and video data storage and recovery method and system. The method includes the following steps: based on the audio and video data stream, decomposing the image and analyzing audio changes, extracting and standardizing features, adjusting encoding and grouping compression based on the features, inserting redundant information, allocating data based on node status, establishing mapping information, retrieving and repairing data, and obtaining the audio and video data repair result. This invention establishes a content- and change-oriented compressed redundancy structure through multi-dimensional feature extraction and sequence standardization, refines data block attributes, dynamically adjusts node allocation and resource mapping, constructs a multi-level index and caching optimization mechanism, enhances the flexibility and completeness of data retrieval, improves the targeting and collaborative repair capabilities of the recovery process, ensures seamless connection and complete restoration of audio and video content, and adapts to dynamic fluctuations in nodes and changes in storage resources in a distributed environment.

The application discloses a power consumption data verification and repair method and device, equipment and medium, which are applied to a power data processing system, and relate to the field of smart power grids. The method comprises the following steps: receiving user power consumption data of a target user collected by a preset collection terminal, and predicting the power consumption data of the target user by using a target power consumption prediction model; determining a current target power consumption dynamic threshold based on obtained target predicted power consumption data and historical power consumption data of the target user, and determining whether a target difference between the user power consumption data and the target predicted power consumption data is greater than the target power consumption dynamic threshold; if the target difference is greater than the target power consumption dynamic threshold, performing abnormal analysis on the user power consumption data to determine an abnormal type of the user power consumption data, and performing corresponding abnormal alarm based on the abnormal type; and performing data repair on the user power consumption data based on the abnormal type to obtain repaired user power consumption data. Therefore, automatic abnormal identification and repair of power data can be realized.

The invention relates to a hemodynamic data processing method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: acquiring synchronous measurement data of various hemodynamic parameters of a target object; performing data anomaly detection on each piece of synchronous measurement data in each pulse period to obtain an anomaly detection result corresponding to each pulse period; if the abnormal data volume of at least one piece of synchronous measurement data in the pulse period does not meet the data recovery condition, data corresponding to the pulse period in the synchronous measurement data is abandoned, and if the abnormal data volume of the synchronous measurement data in the pulse period meets the data recovery condition, data recovery processing is conducted on the abnormal data in the pulse period; and obtaining a synchronous measurement result of the hemodynamic parameters of the target object according to the updated synchronous measurement data. By adopting the method, a high-quality data basis can be provided for evaluation of the automatic cerebral blood flow regulation function.