27 results about "Data level" patented technology

The invention discloses a segmented multi-source shoreline extraction and calibration method and system based on physical prior, and particularly relates to the technical field of remote sensing image processing and computer vision, a consistency evaluation and reliability measurement mechanism of a multi-source surface water product is integrated on a data level, and a high-quality training sample set is constructed by using water persistence characteristics; on the model level, a feature extraction framework fusing physical prior guidance and a geometric self-adaptive operator is researched and developed, a multi-dimensional physical constraint and noise tolerance loss function is constructed, and pixel-level accurate recognition and topological rigorous quality closed-loop calibration of a shoreline under a large-range remote sensing image are achieved.

A method for training a machine learning algorithm including uncertainties. The method includes the following steps: for each point in time of the plurality of points in time, determining in each case an influence, which the data detected at the corresponding point in time have on uncertainties instantaneously contained in the initial model, for each point in time of the plurality of points in time, determining a resolution of the corresponding detected data based on an established data level and on the respective influence, which the corresponding data have on uncertainties instantaneously contained in the initial model, for each point in time of the plurality of points in time, transferring the detected data to the control unit based on the corresponding determined resolution, and retraining of the initial model by the control unit based on the data transferred to the control unit.

The invention relates to the technical field of power system reliability evaluation, in particular to an offshore wind power submarine cable transmission power probability safety evaluation method and system. According to the method, multi-dimensional data management, dynamic prediction modeling and self-adaptive decision optimization are creatively fused, so that double breakthrough of safety assessment precision and robustness is realized. Abnormal data removal, blank value filling and normalization are carried out on the data in the data level, data noise is reduced, and key features are completely improved; in a prediction level, an Att-BiLSTM model is constructed, and wind power output prediction is realized through bidirectional time sequence feature extraction and attention weight dynamic distribution; in the aspect of probability safety assessment, a deep reinforcement learning framework based on a PPO algorithm is designed, the submarine cable power safety margin is dynamically described, full-chain technical support is provided for offshore wind power high-reliability grid connection and power transmission safety monitoring, and the method has wide industrial application prospects.

Real-time monitoring and analysis of data anomalies is described. An example system for detecting data anomalies includes a network interface configured to receive individual information from each of a plurality of individual devices, the individual devices being computing devices. The system also includes a processing unit configured to extract relevant data indicators from the individual information using predefined algorithms, compute a statistical value indicative of a collective data level of the plurality of individual devices by integrating the relevant data indicators, and dynamically adjust data monitoring parameters based on real-time data to enhance accuracy. The system also includes a memory unit configured to store the individual information, the relevant data indicators, and the computed statistical value. The system also includes a feedback module configured to provide personalized data management recommendations associated with the individual devices based on their collective data levels and predefined data relief protocols.

The application provides a sintering flue gas CO emission concentration prediction method, and relates to the field of sintering process pollutant treatment. It comprises the following steps: S1, data acquisition and preliminary rejection; S2, constructing a sample data set and preprocessing it; S3, characteristic value screening, based on the preprocessed sample data set, combining a nonlinear correlation analysis method and a model-driven recursive elimination strategy, constructing an optimal feature subset; S4, integrated modeling, adopting a Stacking integrated learning framework, integrating the advantages of XGBoost, random forest and gradient boosting tree algorithms, and constructing an effluent CO concentration prediction model. Through multi-layer screening of characteristic values, the application clearly determines the dominant action mechanism of each process link on CO generation from the data level, and significantly improves the prediction accuracy and robustness compared with existing models.

The application discloses a segmented multi-source shoreline extraction calibration method and system based on physical prior, and particularly relates to the technical field of remote sensing image processing and computer vision, and integrates a consistency evaluation and reliability measurement mechanism of multi-source surface water products at a data level, and constructs a high-quality training sample set by using water body persistence characteristics; at a model level, a feature extraction architecture is developed by fusing physical prior guidance and a geometric self-adaptive operator, a multi-dimensional physical constraint and a noise tolerance loss function are constructed, and pixel-level accurate identification of a shoreline under a large-range remote sensing image and quality closed-loop calibration with a rigorous topology are realized.

This invention discloses a method for inverting and cascading early warning of hidden dangers in reservoir groups based on spatiotemporal network coupling. The method includes: acquiring hydrological time-series monitoring data and geospatial attribute data of reservoir groups in the region, and constructing a multi-source feature matrix including water level data and elevation information; filtering water level data, constructing a cumulative deviation curve, and judging the hidden leakage trend; constructing a multi-dimensional resilience assessment model based on water level data, obtaining the comprehensive resilience score of each reservoir node, and mapping the comprehensive resilience score to the node's anti-disturbance ability. By fully exploring the value of existing water level and rainfall monitoring data, and using neighborhood dynamic benchmarks and residual accumulation models, natural evaporation and meteorological interference can be effectively filtered out at the data level, and hidden leakage signals inside the dam can be interpreted from conventional hydrological data, providing a low-cost and universally applicable safety monitoring path for small reservoir groups.

A big data problem clue mining method based on mutual exclusivity rules. It relates to the technical field of big data supervision, and particularly relates to the technical field of a big data problem clue mining method based on mutual exclusivity rules. The present application takes the mutual exclusivity contradiction between data records as the core breakthrough point and analysis dimension, accurately captures at the data level, and extracts the key features that violate logical consistency. The method comprises the following steps: obtaining a multi-source heterogeneous data set and pre-processing to obtain an initial feature set and an entity subgraph; extracting mutual exclusion features according to the initial feature set; generating evidence features based on the mutual exclusion features; constructing a double-channel model to calculate the mutual exclusion risk: using a decision tree improvement algorithm as a rule-enhanced feature channel; constructing a graph relationship perception channel: through a dynamic weighting method, the abnormal probability of the feature space and the abnormal probability based on the graph structure are aggregated into a risk score.

The invention belongs to the technical field of industrial process monitoring and fault diagnosis, and provides a heavy medium coal separation process fault detection method based on a dynamic and static mixed graph, which can be wholly called a space-time synchronous attention network (HG-STAN) based on the dynamic and static mixed graph. The method comprises the steps that key variable data in the dense medium coal separation process are collected and preprocessed; constructing a static graph based on process prior knowledge, learning dynamic association from data by using a self-attention mechanism, and fusing to form a dynamic and static mixed graph; constructing space neighborhood information and historical time information of synchronous aggregation nodes of a space-time synchronization graph attention auto-encoder (STGAAE); training the model by using a combined loss function including reconstruction loss and graph structure sparsity regularization; establishing a multi-level monitoring system of data level statistics and graph level statistics; and each statistic control limit is determined through kernel density estimation, and online fault detection is realized. According to the method, the accuracy and sensitivity of fault detection in the dense medium coal separation process are remarkably improved.

The invention discloses a water quality monitoring system and method for environmental protection, relates to the technical field of water quality monitoring, and combines water quality data and environmental data to form multi-dimensional data fusion analysis and avoid the limitation of single parameter analysis, so that the system has global water quality change judgment capability. The obtained water quality data and environment data are combined and fitted into the original data set W, and optimization is carried out on the data level, so that the data have higher consistency and integrity, the influence of missing data on subsequent analysis is reduced, and the data stability is improved. By monitoring the complexity of the algae community structure, the eutrophication degree of the water body can be accurately evaluated, and an accurate regulation and control basis is provided for water quality treatment; a high-precision sensor and a remote sensing technology are adopted, automatic, remote and continuous collection of water quality data is achieved, manual intervention is reduced, the monitoring cost is reduced, the efficiency and the real-time performance of data acquisition are improved, and a foundation is laid for intelligent water affair and automatic environment management.

This invention provides a remote sensing data management method, apparatus, device, storage medium, and product. The method includes: upon receiving a retrieval request from a user terminal for retrieving remote sensing data, acquiring the requested data present in the retrieval request; calculating the target image boundaries of the requested remote sensing image based on the request data; determining a storage index based on the target image boundaries and the data level of the request data; the storage index indicating the storage location of the data corresponding to the retrieval request in the target storage level of a preset remote sensing image management system; retrieving the remote sensing image from the remote sensing image management system according to the storage index; and outputting it to the user terminal. In other words, this invention improves the efficiency of remote sensing data management, thereby enhancing the near real-time application of remote sensing images.

The invention discloses an oversampling and software defect prediction method based on space mapping and probability constraint. The method comprises the following steps: firstly, constructing a minority class probability score model by using dual-kernel density estimation, and realizing layered operation on a preprocessed defect sample; secondly, calculating the number of synthetic samples based on layered weight, performing three-dimensional space mapping and adaptive spherical domain nonlinear sampling on the seed samples, and reconstructing high-dimensional synthetic samples through inverse mapping; and finally, introducing manifold consistency and likelihood ratio dual probability constraints to perform quality verification on the samples, and triggering a radius feedback adjustment mechanism for unqualified samples until a balanced data set is constructed. And training the classifier by using the balanced data set to realize software defect prediction of an unknown sample. By selecting the high-quality seed samples and controlling the synthetic samples, the class imbalance problem is further processed from the data level, and the overall prediction performance of the model is remarkably improved.

The invention relates to a data analysis method and device based on an auditing platform, equipment and a storage medium, and the method comprises the steps: determining a current data level where a data analysis control is located and a target data level where candidate data associated with the data analysis control is located in response to a click operation for the data analysis control in a data analysis page of the auditing platform; based on the current data hierarchy and the target data hierarchy, screening out a target associated field matched with the current data hierarchy and the target data hierarchy from a preset hierarchy association rule; according to page display content of the data analysis page, filling the target associated field to obtain a target associated field value; and querying candidate data associated with the data analysis control from a database according to the target data hierarchy and the target associated field value, and outputting abnormal target data in the candidate data according to a business rule corresponding to the data analysis control. The tedious operation of querying the business systems one by one is avoided, and the data analysis efficiency is improved.

The application discloses a broken surface attribution determination method and device, equipment and a storage medium, and relates to the technical field of geographic information science, comprising: obtaining a to-be-processed broken surface and a plurality of peripheral polygons within a first preset radius of the to-be-processed broken surface from the polygon data set of a target area; establishing an association relationship list based on the to-be-processed broken surface and each peripheral polygon; obtaining a comprehensive score of each peripheral polygon according to the association relationship list, and determining a candidate merging object based on the comprehensive score; verifying the candidate merging object based on the ownership information and data level of the to-be-processed broken surface, and obtaining a verification result; when the verification result is verified, it is determined that the to-be-processed broken surface belongs to the candidate merging object, so that the broken surface merging of the target area is performed according to the to-be-processed broken surface and the candidate merging object. The problem of insufficient attribution determination accuracy caused by single determination decision dimension and lack of compliance constraints in the prior art is solved.

This invention discloses a method, apparatus, and electronic equipment for processing aerodynamic load data corresponding to tasks, belonging to the field of aerospace measurement technology. The task processing method establishes a unified data profile model for aerodynamic load data at the data level through three-layer decoupling of physical parameters, operating conditions, and tasks. At the algorithm level, it achieves standardized management of multi-source algorithms and flexible adaptation across parameters and operating conditions through a plug-in collaborative engine and algorithm template mechanism. At the task scheduling level, it achieves collaborative optimization between acquisition and computation through real-time requirement hierarchies and task profile paradigms. Compared with existing technologies, this invention can maintain the consistency and physical rationality of the analysis process when facing different combinations of physical parameters and complex operating conditions, and can explicitly manage multi-level task requirements under limited computing power, thereby improving the overall reliability and engineering application value of flexible intelligent skin aerodynamic load measurement and digital twin systems.

This application discloses a hierarchical display system for power plant operation data, belonging to the field of power plant data management technology. The system includes: a data processing module configured to acquire power plant operation data from a power plant data source and perform quality assessment processing on the power plant operation data to generate quality labels characterizing the reliability of the power plant operation data; a hierarchical aggregation module configured to perform hierarchical aggregation processing on the power plant operation data based on the quality labels to form aggregated data with different display levels; and a display control module configured to determine the target data level of aggregated data from the aggregated data for loading and rendering according to the quality labels and current display requirements; different loading and rendering strategies are adopted for different display levels. This application is applicable to large-scale power plant operation data scenarios and can solve the problems of data loading delay and limited display performance.

A user load group typical power consumption mode extraction method, system, device and medium, the method comprises: acquiring meteorological and power consumption data of a selected area; extracting dimension reduction features of load data in time domain and frequency domain at data level; at mechanism level, obtaining the correlation between different meteorological factors and load by factor analysis, combining redundant meteorological information, then summarizing the statistical characteristics of load from four aspects of load size characteristics, load characteristic time point, load demand response potential and industry characteristics, and calculating the importance score of meteorological characteristics and statistical characteristics, selecting the characteristics according to the importance score to obtain key mechanism characteristics; clustering the dimension reduction features of load data in time domain and frequency domain again to obtain a primary classification result; then based on the key mechanism characteristics, adopting hierarchical clustering algorithm to perform hierarchical clustering on the primary classification result to obtain the user load group typical power consumption mode. The present application can better reveal the power consumption demand of users.

The application provides a data query method for cascade screening of new energy industry organization institutions, comprising: obtaining a permission mark of a current login user; comparing the permission mark of the user with a permission / level dictionary to obtain a user level of the user; determining a permission mark of data to be queried; comparing the permission mark of the data to be queried with the permission / level dictionary to obtain a data level of the data to be queried; comparing the user level of the user with the data level of the data to be queried to determine whether the user has the permission to query the data; when the user has the permission to query the data, displaying the next level data corresponding to the user level through a drop-down box; and returning the data to be queried or expanding the cascade data level by level according to the selection operation of the user. The application saves all permission levels and related logic of permission interface calling in a dictionary, and the dictionary is configured by a management background, so that the adjustment of the permission is more convenient and lightweight.

The application provides a data analysis method and device and electronic equipment, and relates to the technical field of data processing.In the application, first, a crash test data set for stress testing of multiple measured baseboard management controllers (BMC) is acquired; then, based on a data level to which each crash test data in the crash test data set belongs, a target analysis rule corresponding to the data level is selected from an analysis rule set, and each crash test data is analyzed based on the target analysis rule to obtain each target analysis data; finally, each target analysis data is iteratively analyzed, and an analysis result corresponding to each target analysis data is output. In this way, a large amount of crash test data generated by BMC failure can be automatically collected, and the analysis rate of the crash test data is improved.

The invention provides a fluid identification method coupling reservoir physical constraint generation and horizon adaptive interpretation, and aims to cooperatively relieve tendency deviation caused by non-uniform sample distribution in fluid classification from binary dimensions of data representation and model decision. According to the method, a table data generation model (REaLTabFormer) based on Transform is introduced to generate a physical authenticity-guaranteeing sample, and fluid category distribution is improved from the data level. Compared with a traditional ensemble learning fixed weight fusion mode, the method has the advantages that the self-adaptive interpretation scheduler is used for sensing and recognizing feature states of different stratums, and a gating strategy aiming at heterogeneous judgment interpretation elements is dynamically generated, so that multi-solution interference caused by stratum heterogeneity is inhibited. In combination with an asymmetric reward mechanism based on geological risk compensation, the penalty weight of sparse fluid identification errors is increased through an unbalance coefficient, and prediction deviation generated when data distribution of a traditional model is uneven is effectively overcome. According to the invention, an efficient intelligent scheme is provided for oil gas evaluation in a complex work area.

This application discloses a data analysis method, apparatus, device, storage medium, and computer program product to address the problem that existing data analysis methods cannot accurately match and quantitatively correlate discrete noise events with continuous, macroeconomic economic indicators within a unified spatiotemporal framework. This makes it difficult to scientifically reveal the intrinsic relationship between the two at the data level, thus failing to provide direct decision support for precise intervention in urban governance. The method includes: spatially aligning collected economic data according to a preset geographic grid to determine the economic data corresponding to each geographic grid; acquiring noise data corresponding to each geographic grid; determining the association topology between the geographic grids based on the noise data; generating an association matrix based on the temporal changes of the economic data corresponding to each geographic grid and the association topology; and generating a management strategy for each geographic grid based on the association matrix.

The application discloses a data security risk identification method based on data attribute classification and grading. The method solves the problem that the existing security risk identification of data through automatic grading cannot accurately calculate the damage degree of data security time and cannot meet the data classification diversity requirement. The method comprises the following steps: collecting target information system data, pre-processing the collected data, judging the attributes of the data, classifying the data, setting data levels, establishing a support vector machine, training the support vector machine with the training data after the data level marking, obtaining a risk identification model, and identifying the security risk of the collected data with the risk identification model. The application utilizes data attribute classification and grading, adopts the support vector machine classification method to judge the data level, identifies the data security risk level, and accurately determines the security of the system. The method is simple in operation, strong in data interpretability, good in anti-interference effect, and can obtain relatively accurate data results.

The present application belongs to the technical field of industrial process monitoring and fault diagnosis, and proposes a heavy medium coal preparation process fault detection method based on dynamic and static mixed graph, which can be called as HG-STAN based on dynamic and static mixed graph. The method comprises: collecting key variable data of heavy medium coal preparation process and preprocessing; constructing a static graph based on process prior knowledge, and learning dynamic association from data by using self-attention mechanism to form a dynamic and static mixed graph; constructing a time-space synchronous graph attention autoencoder (STGAAE) to synchronously aggregate the spatial neighborhood information and historical time information of the nodes; training the model by using a combined loss function containing reconstruction loss and graph structure sparsity regularization; establishing a multi-level monitoring system of data level statistics and graph level statistics; determining the control limit of each statistic by kernel density estimation to realize online fault detection. The present application significantly improves the accuracy and sensitivity of heavy medium coal preparation process fault detection.

The invention discloses an electric energy quality real-time evaluation method based on dynamic empowerment and data level double-track fusion, which relates to the technical field of electric energy quality evaluation, and comprises the following steps: obtaining an actually measured data sequence and a real-time numerical value of an electric energy quality monitoring index; calculating a basic weight through the information entropy of the actually measured data sequence, and performing correction by using a state variable weight factor determined based on the ratio of a real-time numerical value to a preset safety limit value to obtain a dynamic weight; obtaining grading threshold values of different evaluation grades, and calculating a weighted distance between the real-time numerical value and the grading threshold value and a preset ideal solution based on the dynamic weight; based on the weighted distance, the relative closeness degree of the real-time numerical value and the grading threshold value is quantified and serves as the real-time adaptation degree and the grade adaptation degree; and determining a real-time electric energy quality evaluation grade by comparing the real-time adaptation degree with the grade adaptation degree. The method is used for solving the problems that in the prior art, index abnormal state response is insufficient, and assessment results lack a unified reference standard.

The invention discloses a model updating method and device, equipment and a medium, and belongs to the field of artificial intelligence. The method comprises the steps of screening out target data serving as model increment information according to input to-be-updated data; predicting a target parameter subset related to the target data in the generative model according to the target data; and finally, performing parameter updating on the target parameter subset by using the target data to obtain an updated generative model. According to the method, through a two-stage screening mechanism, valuable new knowledge is screened on a data level, a sensitive updating area is dynamically positioned on a model parameter level, high-cost global model updating is converted into accurate local optimization, computing resource consumption is reduced, the model can efficiently and frequently integrate the new knowledge, and therefore the timeliness of the new knowledge is kept; and the updating range is limited to the parameters related to the newly added knowledge, so that the interference on mastered knowledge of the model is avoided, and the balance between the accuracy of knowledge fusion and the stability of the model is realized.

The invention discloses a dynamic data supervision method, device and equipment based on an intelligent agent, and relates to the field of data security governance, and the method comprises the steps: carrying out the keyword extraction of public health data and multi-source network public opinion data through a pre-training language model, and calculating the similarity between a to-be-graded keyword obtained after extraction and a public opinion keyword; performing risk assessment on the public health data to obtain an inherent risk value; calculating the product of the source weight of the public opinion data and the keyword similarity score to obtain the source credibility; based on the time interval between the publishing time and the current time of the public opinion data and the exponential decay function, generating a failure decay factor, and calculating the product of the failure decay factor and the source credibility to obtain a public opinion similarity score; and fusing the public opinion similarity score and the inherent risk value, and determining a target data level based on a target risk interval where the fused comprehensive risk value is located. According to the invention, the timeliness and accuracy of public health data classification can be improved.

The application discloses a forest fire risk prediction method based on time sequence characteristics and an integrated model, and relates to the field of forest fire risk prediction. From the data level, the training set random downsampling algorithm is adopted; from the algorithm level, the more minority class prediction model (BRF) and the setting of each model super parameter are adopted; from the evaluation level, the more minority class prediction evaluation index is selected, and the effectiveness of the fire risk prediction model under the unbalanced data is comprehensively evaluated. By extracting the average value and trend value time sequence dynamic characteristics of the meteorological and combustible materials within 16 days, the cumulative effect of the fire risk is effectively described, and the identification accuracy of the potential high-risk fire area is improved. Through the base model combination of XGBoost, LightGBM and balanced random forest and the fusion of the logistic regression meta model, the advantages of different models are integrated, and compared with a single model, the prediction accuracy is improved.