30 results about "Correlation method" patented technology

The invention discloses an industrial product quality risk association method based on a graph neural network, and relates to the field of industrial manufacturing, and the method comprises the steps: collecting multi-source industrial data, carrying out the preprocessing of the multi-source industrial data, obtaining a standardized data table, and constructing an industrial element heterogeneous graph based on the standardized data table; inputting the industrial element heterogeneous graph into a graph attention network basic model for node representation learning, and setting a supervised learning task taking a product batch node in the industrial element heterogeneous graph as a prediction target and taking a historical quality detection result as a supervised label; and introducing a meta-path self-learning layer in a message passing process of the graph attention network basic model. According to the method, deterministic prediction is upgraded into probabilistic inference, internal confidence degree measurement is provided for each risk prediction value, and automatic, accurate and interpretable correlation analysis and traceability positioning of the industrial quality risk are realized.

The invention discloses a power plant 5G private network elastic security data analysis method and device based on data mining, and relates to the technical field of power grid security data analysis. The method comprises the steps of defining a power grid security data analysis target and an analysis range; establishing a data acquisition and transmission link, and acquiring operation parameters, operation logs and fault record data of key equipment of the power grid in real time; the acquisition server transmits the acquired data to the security access area; the collection server preprocesses the received data to obtain a time sequence data set of the security data; performing preliminary correlation analysis on the time series data set by adopting a Spearman rank-based correlation method to obtain correlation among a plurality of security data state indexes; based on the correlation between the indexes, obtaining a time series data set corresponding to the strong association security data indexes; and based on the time sequence data set, mining the association rule of the security data by adopting a time sequence FP-Growth algorithm to obtain a strong time sequence association rule. According to the invention, the utilization efficiency of the security data can be improved.

The embodiment of the invention provides an investment product data association method and system and electronic equipment, and relates to the technical field of computer application. The investment product data association method comprises the following steps: acquiring a plurality of investment product data records in a plurality of data sources; for each investment product data record, extracting a key field of the investment product data record, the key field being used for identifying an investment product corresponding to the investment product data record; calculating the similarity of key fields of different investment product data records; under the condition that the similarity of the key fields of the different investment product data records reaches a preset similarity condition, establishing an association relationship of the different investment product data records; and merging the investment product data records with the association relationship to obtain a merged data record. Through the investment product data association method and system and the electronic equipment provided by the embodiment of the invention, the accuracy of data association of the same investment product can be improved.

The invention discloses an optical observation correlation method based on mixed multi-hour ephemeris, and belongs to the technical field of space target monitoring. Aiming at the problem that the correlation performance is reduced when the ephemeris quality is unstable due to dependence on a single ephemeris in the prior art, the method comprises the following steps of: firstly, constructing a mixed set containing multi-source ephemeris; secondly, intelligently selecting a leading ephemeris and a lagging ephemeris which are closest in time for each observation arc section and each target; then, the two ephemeris are used for association respectively, and union sets of results are obtained, so that an initial association set is expanded; and for the generated fuzzy correlation, performing primary disambiguation based on the comprehensive confidence, and performing secondary accurate disambiguation on the residual fuzzy condition by comparing the weighted angular distance residual error root-mean-square value of the whole arc section. According to the method, the dependence on single-term and single-source ephemeris precision is effectively reduced, available information is fully utilized, and the correlation robustness and accuracy of optical observation data under the complex ephemeris condition are remarkably improved.

The invention provides a product full life cycle data model construction and association method, system and device and a medium, and relates to the field of weapon equipment digitization. According to the method, a data model architecture is determined by analyzing light weapon full life cycle data, and a unified modeling language is used for carrying out model building on a meta-model layer, an abstract model layer and an object instance layer. On the basis of a constructed light weapon development and production process data model, a light weapon product data association rule is constructed from the perspective of application. Based on the constructed association rule and the light weapon product data model, combining the data with the association rule, constructing a light weapon product data association relationship, and associating the light weapon product data through unique identifiers such as a product code, a batch number, a system code and the like. The method provides a basis for application of quality problem tracing, rapid fault diagnosis and positioning, product digital resumes, knowledge maps and the like in the whole life cycle, and the light weapon whole life cycle data acquisition efficiency can be effectively improved.

The application discloses a track correlation method based on improved topology, which comprises the following steps: S1, converting two radar target tracks into a common coordinate system; S2, constructing a topology sequence of the radar target tracks; S3, combining historical topology information to calculate the topology average distance of the target at the current moment; S4, using the OSPA distance to measure the correlation degree of the target topology set to obtain a correlation matrix; and S5, using the Hungarian algorithm to perform optimal distribution on the correlation matrix to obtain a correlation result. The track correlation method based on the improved topology can effectively reduce the influence of random errors on the track correlation algorithm and improve the correlation performance.

The application discloses a kind of equipment fault diagnosis methods based on time series data analysis in industrial production in industrial scene, steps include: the data in time series database generated when industrial production system runs is preprocessed;Effective sample is extracted from the data after pre-processing, and multiple measuring point sample time is synchronized;Effective sample data is analyzed according to the characteristics of occurrence frequency, occurrence time, duration, and the correlation of the measuring points is obtained, so as to extract relevant measuring points;According to the data of relevant measuring points, the fault attribution factor graph is inferred.The application uses the data mining method related to big data to mine the correlation of equipment failure from massive sensor data, uses the related method in the field of artificial intelligence, carries out single-source, multi-source time series analysis on historical sensor data, locates one or more fault positions corresponding to abnormal data in combination with the correlation relationship mined.

The application provides a satellite in-orbit anomaly and space environment correlation method based on density distribution drift, comprising the following steps: S1: determining a typical repetitive anomaly of an in-orbit spacecraft, on the basis of which, it can be considered that the event has consistent environmental correlation characteristics, so that it can be statistically analyzed together and further analyzed; S2: using the occurrence of the abnormal event as a condition, the space environment data is divided into two parts, and the probability density distribution of the two parts is counted; S3: according to the difference between the probability density distribution of the two parts in S2, the influence of the space environment is formed to form an explainable correlation characteristic, and the correlation analysis is completed. The method is suitable for analyzing the in-orbit repetitive anomaly phenomenon, and the possible correlation characteristics between the change characteristics of the space environment and a kind of repetitive in-orbit anomaly phenomenon are analyzed by using the distribution characteristic construction method in statistics.

The application discloses an industrial product quality risk correlation method based on a graph neural network, relates to the field of industrial manufacturing, and comprises the following steps: collecting multi-source industrial data, preprocessing the multi-source industrial data, obtaining a standardized data table, constructing an industrial element heterogeneous graph based on the standardized data table, inputting the industrial element heterogeneous graph into a graph attention network base model for node representation learning, setting a product batch node in the industrial element heterogeneous graph as a prediction target, and setting a historical quality detection result as a supervised label of a supervised learning task, and introducing a meta-path self-learning layer in a message passing process of the graph attention network base model. The application upgrades deterministic prediction to probabilistic inference, provides an inherent confidence measure for each risk prediction value, and realizes automatic, accurate and interpretable correlation analysis and traceability positioning of industrial quality risks.

The present application relates to the technical field of smart grid, disclose a kind of multi-factor associated return home degree winter power load pre-control method and system, method includes: the historical power consumption data obtained in advance is transformed to obtain power load data, and power load data is identified to obtain power consumption problem;The influencing factor is obtained by analyzing power consumption problem, and the strong correlation factor data matrix is obtained by analyzing influencing factor, and the correlation influence coefficient is obtained by analyzing strong correlation factor data matrix;The trained power load prediction model is obtained by iterative training to power load prediction model, and the current power consumption data obtained in advance is input into trained power load prediction model to obtain load prediction data;The red yellow green classification area is obtained by dividing power grid;Power load pre-control scheme is formulated, so as to carry out power consumption prevention and control management.The present application can solve the problem that power grid overload is difficult to prevent and control during return home degree winter and power consumption prediction is not accurate enough.

The invention relates to the technical field of knowledge base fusion, in particular to a knowledge base data fusion association method and system based on an AI large model, and the method comprises the steps: firstly collecting original data through a connector, then carrying out cleaning and duplicate removal processing, then carrying out vectorization alignment on field names, extracting entities and attributes through the AI large model, and carrying out data fusion association on the entities and attributes; according to the method, entity conflicts are solved through a density clustering algorithm and an entity similarity threshold value, a relation triple is generated, unified identifiers are distributed for entities and attributes for standardization processing, data conflicts are detected through variable coefficients, parameters are dynamically adjusted to optimize a fusion strategy, and finally a graph database is utilized to construct a vector index and provide a graph downloading interface. According to the method, on the basis of realizing data fusion, the problem that reasons are difficult to discriminate and solutions are difficult to give in a targeted manner when data conflicts occur in an existing database fusion method is ingeniously solved, the accuracy and reliability of knowledge base data fusion are remarkably improved, and the method has important innovative significance and practical value.

The application discloses a dynamic information association method, comprising the following steps: 1) receiving a new information record, and determining a similar information entity set according to a static attribute; 2) if there is no similar information entity, a new information entity is created, and the process is ended; if there is a similar information entity, a history information record of a dynamic attribute of the similar information entity is obtained; 3) a prediction model for each dynamic attribute of the information entity is constructed, a reasonable value range of the dynamic attribute is determined, a correlation feature of the prediction model of each dynamic attribute of the information entity is constructed, and a prediction value range of the dynamic attribute of the similar information entity is calculated; and 4) if there is no similar information entity, a new information entity is created; if there is a similar information entity, the new information record is associated to the most matched candidate information entity. According to the application, the change trend of the dynamic attribute and the correlation between the dynamic attributes are utilized to associate information from different data sources and / or different time points, so that the accuracy and recall rate of the association are improved.

The application provides a relaxation and creep performance correlation method, and belongs to the field of new materials and is used for developing new materials by testing related activities. First, a high-temperature relaxation test is carried out to obtain corresponding data of temperature-stress-creep relaxation rate. Second, a high-temperature creep test is carried out to obtain minimum creep rate data. Finally, the third step is data processing, and a correlation model is constructed. The application aims at the shortcomings of long data obtaining time and large sample consumption of existing creep performance data, and proposes to use the characteristics of short stress relaxation test, i.e., more creep information and less sample consumption, to quickly obtain a large amount of stress relaxation performance data, and then combine the correlation method shown in the application to quickly and massively obtain the creep performance, so that the service performance can be evaluated.

The application discloses an AI auxiliary-based natural resource factor cross-layer index correlation method and system, which comprises the following steps: firstly, a set containing multiple remote sensing images is acquired, and the set and to-be-matched natural resource factor information are loaded into a pre-trained multi-modal ground feature correlation model to respectively obtain remote sensing image features and description features; and a target remote sensing image is determined by calculating a matching degree. Different layer data is obtained by separating a target image layer, layer features are extracted, and correlation with factor index features is established, thereby constructing a cross-layer index correlation model, and effective correlation of natural resource factor cross-layer indexes is realized.

The application discloses a stratum correlation method and device. The method comprises the following steps: performing continuous wavelet transform on selected well logging curves of a target well to obtain a modulus distribution diagram of the continuous wavelet transform coefficients at each depth point under each set scale; obtaining a modulus maximum line from the modulus distribution diagram; taking the position of the selected well logging curve when the modulus maximum line takes the minimum value of the scale variable as a singular value position; determining a to-be-correlated depth interval of the target well according to the singular value position and a correlation depth interval of a standard well determined in advance; obtaining a first data point sequence of the selected well logging curves in the to-be-correlated depth interval and a second data point sequence of standard well logging curves of the standard well in the correlation depth interval; determining a path between the first data point sequence and the second data point sequence according to a set rule; determining the distance of each path; and obtaining a stratum correlation result of the target well and the standard well according to the path with the shortest distance. The method can realize automatic fine stratum correlation.

The present specification provides a high-precision stratigraphic correlation method, device and electronic equipment, which constructs a filtering factor through well logging data, and then extracts a high-frequency seismic data body and a low-frequency seismic data body from a seismic data body by using the filtering factor. Through time-varying seismic wavelet deconvolution on the high-frequency, medium-frequency and low-frequency seismic data bodies, a target data body with a "virtual resolution" and a much wider frequency band than the seismic data body can be obtained, so that the profile data of the target data body can be used to more accurately distinguish the interwell lateral characteristics, and the stratigraphic correlation is adjusted based on the interwell lateral characteristics to realize high-precision stratigraphic correlation. The high resolution and lateral variation characteristics of the target data profile are used to finely carry out stratigraphic correlation, thereby providing protection for fine exploration and development of oil and gas fields, especially development geology research.

The application relates to the technical field of knowledge base fusion, and particularly relates to a knowledge base data fusion correlation method and system based on an AI large model. First, original data is collected through a connector, then washing and deduplication processing are carried out, then field names are vectorized and aligned, an AI large model is used to extract entities and attributes, entity conflicts are solved through a density clustering algorithm and an entity similarity threshold, relationship triplets are generated, uniform identification is assigned to entities and attributes for standardized processing, data conflicts are detected through a coefficient of variation, and parameters are dynamically adjusted to optimize a fusion strategy, finally, a vector index is constructed through a graph database, and a graph download interface is provided. On the basis of realizing data fusion, the application skillfully solves the problem that in the existing database fusion method, when facing data conflicts, it is difficult to specifically identify the reasons and give solutions, significantly improves the accuracy and reliability of knowledge base data fusion, and has important innovative significance and practical value.

The present application relates to the technical field of data analysis, and particularly relates to a multi-modal data adaptation and feature correlation method for intelligent insight, comprising the following steps: standardizing and arranging structured data fields in data sources of different systems to form a structured field set, identifying field drift patterns, and generating a structure drift label set; guiding semantic flux analysis on content related to the structured field in unstructured modal data, and extracting core semantic candidate units; finally screening out a standardized target semantic unit set; performing graph structure alignment on the standardized target semantic unit set and the structured field set to obtain a field evolution graph, and performing dynamic adaptation based on field inheritance relationships and time sequence evolution paths in the field evolution graph. The present application effectively excludes pseudo-related expressions that are only close in static semantics but have inconsistent contexts, and improves the authenticity and business effectiveness of semantic unit screening.

The application provides an improved FMEA method based on interval two-type fuzzy sets and DEMATEL-VIKOR hybrid multi-criteria decision, and on the basis of traditional FMEA, a more perfect IMS key hardware risk assessment model is constructed by improving related methods. The failure modes involved are obtained by the fault tree analysis method. The application combines the interval two-type fuzzy sets and the FMEA to solve the problem of inaccurate RPN value in the traditional FMEA; the similarity aggregation method is used to calculate the comprehensive weight of the experts in combination with the personal background of the experts; the combination weight method is used in the risk factor aspect to improve the analysis precision, the weight of multiple indexes is comprehensively considered, and the contribution of each factor to the overall risk can be more comprehensively evaluated; and the DEMATEL-VIKOR hybrid multi-criteria decision method of preference relation analysis fully considers the correlation between factors and the comprehensive influence of multi-criteria, and provides more comprehensive and intuitive decision support.

The present application relates to a kind of isochronous stratum automatic correlation method, device, medium and equipment, the method includes the following steps: obtaining a well profile, select several wells as standard well, and complete stratum division to the standard well;Optimizing the well logging curve that can effectively respond to stratum marker bed, sedimentary cycle, lithologic combination, and the preferred logging curve is pretreated;With the logging curve of standard well as data basis, in turn, complete the marker bed identification of well profile on the well to be compared, sedimentary cycle identification and lithologic combination identification;According to the linear solution of stratum boundary point except marker bed identification point, sedimentary cycle identification point, lithologic combination identification point is completed according to stratum superimposed pattern;The stratum automatic division result visualization of profile is carried out.The isochronous stratum automatic correlation method is fully and reasonably integrated into rich geological expert knowledge, with stratum superimposed pattern as constraint, based on the principle and step of geological expert stratum correlation, computer algorithm is designed, and stratum automatic correlation is realized.

The application discloses a multi-mode feature correlation method based on structured feature extraction, extracts point and line features from binocular images, estimates plane features through line features, and then constructs correlation between different features by using spatial geometric relations between features. Mainly includes the following steps: first, extract point and line features, then estimate the plane features of intersecting lines according to reliable line features, then construct the correlation between different features by using the spatial geometric relations between features and create a semantic map, finally combine the correlation with the back end to optimize the system. The application improves the algorithm for estimating the plane by using the line feature, gives the mathematical model of the spatial geometric relation between features, and uses the correlation for the back end optimization of the SLAM system, reduces the cumulative error of the system, and provides better positioning capability.

The invention discloses a sea-gas interface momentum flux calculation method coupled with boundary layer meteorological conditions, and relates to the technical field of marine atmospheric boundary layer flux observation, comprising: performing threshold screening and outlier removal processing on turbulence flux elements required by vortex correlation method calculation, and performing coordinate rotation on three-dimensional wind speed; processing the obtained sea surface fluctuation data to obtain a sea wave spectrum and wave parameters; on the basis of the Monin-Obhoff similarity theory, the Monin-Obhoff length, the atmospheric stability and the wind speed U10 of 10 meters on the sea are calculated; based on the data, splitting the friction speed into a wind speed item and a wave mixing item for fitting, and finally obtaining momentum flux; on the basis of guaranteeing the flux calculation precision, the requirement for an observation instrument is lowered, accurate calculation of the momentum flux of the air-sea boundary layer is achieved, and the requirements for momentum flux data in the fields of marine meteorological research, weather forecast, marine resource development and the like are met.

The invention relates to a method and an apparatus in which portions of adhesive (15) arranged on a substrate (12) are detected and / or tested during a production process, in which the portions of adhesive (15) are applied to the substrate (12), by means of a sensor (18, 29 A, 29 B) which generates measured values and is directed at the substrate (12), wherein the actual measured values (25) generated by the sensor (18, 29 A, 29 B) during the course of the method are evaluated, in particular by means of an analysis device, as to whether they have a measured-value group (21-24, 32) characteristic of a portion of adhesive (15). The method is characterized in that, within the scope of the evaluation, in the event of an actual measured-value group (21-24, 32) which could be characteristic of a portion of adhesive (15) being established, a correlation method is used to determine a measure of possible correlation - correlation measure - between on the one hand said actual measured-value group (21-24), or a value group derived from said actual measured-value group (21-24, 32), and on the other hand a target value group (28, 34) stored in a memory.

The invention belongs to the technical field of sea surface target association, and particularly discloses a sea surface target association method based on feature topological graph matching. The method comprises the steps of generating a feature topological graph of a to-be-associated target based on target sea surface distribution detected by a detector; two to-be-associated targets detected by different detectors form a to-be-associated target pair, and through rotation and translation, the number of reference target pairs meeting neighborhood overlapping in two feature topological graphs of the to-be-associated target pair is the maximum; the matching degree of the reference target pairs is evaluated according to the distance between two reference targets in the reference target pairs, and the matching degree of the target pairs to be associated is evaluated according to the matching degrees of all the reference target pairs; and finally, iteratively evaluating the matching degree of all the target pairs to be associated, and associating the target pair to be associated with the optimal matching degree as the same target. The sea surface target association method does not depend on absolute position information of the target, and is higher in adaptability and robustness.

The invention discloses an asynchronous track association method, device, medium and system, and belongs to the field of distributed multi-sensor multi-target tracking track management, and the method comprises the steps: S1, carrying out the association statistical test of an association period on two pieces of received asynchronous track data: if the association period ends and the track separation quality exceeds a threshold value, directly judging that the two tracks are not related, and if the track separation quality exceeds the threshold value, directly judging that the two tracks are not related; otherwise, carrying out re-checking period correlation statistical test; and S2, performing re-check period correlation statistical test on the two tracks passing through the correlation period: if the re-check quality of the tracks exceeds a threshold value after the re-check period is ended, the two tracks are considered to be correlated, and otherwise, the two tracks are judged to be uncorrelated. The technical problem of track error accumulation caused by time registration is avoided.

The invention relates to the technical field of intelligent traffic perception, in particular to a traffic multi-target association method based on spatial grouping and feature matching, which comprises the following steps: firstly, pre-classifying detected targets by adopting a spatial position grouping algorithm, and establishing a dynamic grouping model based on spatial density distribution and grouping relationship of the targets to effectively reduce target search space; then calculating a multi-dimensional correlation degree quantity, constructing a correlation matrix, comprehensively evaluating the motion consistency and feature similarity of the target, recording an optimal matching result for all successfully matched tracks in the same group, delaying to output the matching result, and when more than 50% of targets in the same group are successfully matched, outputting the matching result; or outputting the trace point with the most matching times in the track as the current matching result when the target matching times exceed 10 times. According to the traffic multi-target association method based on spatial position grouping and feature matching, the accuracy and real-time performance of multi-target association are remarkably improved by fusing target spatial distribution characteristics, motion features and delay output.

The invention relates to an imaging device (1) for measuring the dynamics of a sample comprising moving entities (2), the device (1) comprising: - a camera (3) capable of acquiring a first set of images, the camera being defined by an imaging time TCAM comprising an exposure time Texp and a dead time TP; - a pulsed light source (4) capable of generating light pulses that are intended to illuminate a measurement zone of the sample; - a synchronisation unit (5) configured to synchronise the emission of each light pulse with the exposure time of the camera according to a series of light pulses; - a unit for defining a series of light pulses (6) in order to define an increasing sequence of time intervals between two pulses; - an image construction unit (7) configured to construct a set of pairs of images in order to define an increasing sequence of time intervals between two images.

The invention discloses a low-orbit target association evaluation method based on optical observation, and belongs to the technical field of space target monitoring. In order to solve the problems of high false alarm rate and poor robustness caused by neglecting a distance factor in a traditional fixed angular distance correlation method, the method comprises the following steps: firstly, setting a nominal detection distance and a corresponding nominal angular distance residual error threshold; secondly, aiming at potential associated targets of different distances, dynamically calculating an exclusive associated threshold value according to a formula that an angular distance residual error threshold value = a nominal threshold value * (nominal distance / target distance), and reflecting a principle that an angular distance residual error caused by an ephemeris forecast error is inversely proportional to the distance; and finally, in order to prevent the correlation threshold of the extremely close distance target from being too wide, setting a maximum angular distance threshold for constraint. According to the method, the association confidence of different distance targets is unified through the distance self-adaptive dynamic threshold, the association false alarm rate is remarkably reduced, the tolerance to ephemeris forecast errors is improved, and the accuracy and robustness of optical observation data association are effectively improved.

A system and associated methods for organizing, representing, finding, discovering, and using data. Embodiments represent information and data in the form of a data structure termed a Feature Graph, that includes nodes and edges, where the edges serve to connect a node to one or more other nodes. A node in a Feature Graph may represent a variable, such as a measurable object, characteristic, or factor of a study. An edge in a Feature Graph may represent a measure of a statistical association between a node and one or more other nodes. Datasets that demonstrate or support the statistical association or measure the associated variable may be accessed through an identifier in a Feature Graph. An application may traverse a Feature Graph and aggregate and process data associated with a set of nodes or edges.