41 results about "Information Criteria" patented technology

The invention relates to a method, a device and equipment for identifying abnormal pollution discharge based on multi-source data, and a medium, and belongs to the technical field of environment monitoring. The method comprises the following steps: collecting working condition electricity consumption historical data and pollutant discharge historical data; extracting at least two key variables from the working condition electricity consumption historical data and the pollutant emission historical data, taking the key variables as endogenous variables, constructing a vector autoregression model according to the endogenous variables, and determining a lag order of the vector autoregression model by using an information criterion method; processing the working condition electricity consumption historical data and the pollutant emission historical data to obtain a training set, and training a vector autoregression model by using the training set; obtaining input data according to a to-be-predicted moment and the lag order, inputting the input data into the trained vector autoregression model, outputting a predicted value of a key variable at the to-be-predicted moment, and obtaining a corresponding measured value; and judging whether pollution discharge is abnormal or not according to the predicted value and the measured value of the key variable, and analyzing an abnormal reason.

The invention discloses a water conservancy spraying area identification system based on a laser radar, and the system comprises a data acquisition and modeling module which completes the distortion removal of a point cloud, generates a signed distance field and a reflection intensity body, and builds tangential, normal and auxiliary normal coordinates; the section slicing module obtains a section point set; the feasibility feature and anisotropy Morse function module constructs a scalar field and generates a gradient field; the complex construction module is used for establishing a two-dimensional Morse-Smal complex to obtain an initial spraying area mask and a section boundary candidate; the boundary completion module is used for realizing a homotopy-preserving closed boundary according to the distance field gradient; the persistent coherence screening module is used for obtaining a stable section boundary; and the section calculation and coverage initialization module is used for completing section type fitting and information criterion selection and outputting an operation section size vector and a strip coverage initialization parameter. According to the method, the sprayable area and the section size are integrally identified, the boundary is continuous and consistent, and the method adapts to complex working conditions.

The invention discloses a sluice multi-monitoring-point deformation monitoring model construction method, which comprises the steps of S1, constructing a group of bidirectional analysis frameworks, and forming a two-dimensional panel by fusing time sequence data and sluice cross section data; s2, expanding the two-dimensional panel containing the time dimension and the cross section dimension to a geographic space attribute level to construct a three-dimensional panel; s3, on the basis of three-dimensional panel data representation, performing clustering analysis according to the space-time similarity of the monitoring points, constructing an information criterion to evaluate candidate clustering numbers, performing parameter calibration by adopting a random coefficient model, constructing a Gaussian mixture model in each cluster as a multi-monitoring-point model, then estimating undetermined model parameters through the random coefficient model, and performing multi-monitoring-point monitoring on the multi-monitoring-point model. The method is used for constructing a deformation monitoring model of the sluice gate and carrying out fitting and prediction capability evaluation. According to the method, collaborative modeling can be carried out on data of a plurality of monitoring points, deformation space correlation characteristics between adjacent monitoring points are represented, and the accuracy and effectiveness of sluice deformation prediction are improved.

The invention belongs to the technical field of uncertainty probability characterization analysis, and discloses a thermoplastic composite material leading edge structure design allowable value probability characterization method and system under a small sample condition, and the method comprises the steps: defining a plurality of candidate probability distribution models; fitting each model based on the original sample data and calculating an AIC value and a BIC value; a dynamic weight factor alpha is calculated according to the sample size n, and then a hybrid information criterion HIC value is calculated; generating a plurality of sample sets through Bootstrap self-service sampling, recalculating the HIC value on each sample set, and counting the selected optimal frequency of each model; and determining an optimal probability distribution model according to the frequency, wherein the optimal probability distribution model is used for representing a design allowable value. According to the method, the dynamic weight factor alpha is introduced, AIC and BIC criteria are effectively unified, optimal balance between prediction precision and model complexity is achieved under the condition of small samples, and engineering practicability and robustness are remarkably improved.

The invention relates to the field of electric vacuum devices, and provides a method, a device and equipment for evaluating the service life of a space traveling wave tube, and the method comprises the steps: generating an equivalent comprehensive stress profile according to an on-orbit working profile of the space traveling wave tube; under the action of the equivalent comprehensive stress profile, continuously collecting a life characteristic parameter set; performing data preprocessing on the life characteristic parameter set, performing stable degradation interval identification on the obtained purification parameter set, and determining a starting moment of a stable degradation interval; performing filtering smoothing processing on the purification parameter set from the starting moment, inputting obtained smooth degradation data into a preset machine learning model library, determining an optimal degradation function according to an information criterion, and obtaining a life point estimation value based on the optimal degradation function; and obtaining life interval estimation according to the life point estimation value. According to the invention, the problems of long test period and low accuracy of space traveling wave tube life evaluation in the prior art are solved, and rapid and accurate prediction of the long life of the space traveling wave tube in limited test time is realized.

The invention belongs to the technical field of system modeling, and relates to a multi-innovation OIF Elman network modeling method based on measurement data, which comprises the following steps of: constructing a mathematical model of an OIF Elman network, and obtaining a to-be-estimated parameter matrix of the network by defining a parameter matrix; a sliding window mechanism is introduced, a to-be-estimated parameter matrix is utilized to define a multi-information criterion function, and a parameter estimation sub-algorithm for estimating the OIF Elman network weight is constructed; establishing a dynamic factor adjustment strategy by adopting an exponential decay and minimum value limitation mode; in the parameter estimation sub-algorithm, a dynamic factor adjustment strategy is introduced, and a gradient parameter estimation algorithm is constructed; and estimating a parameter matrix of the OIF Elman network model through a gradient parameter estimation algorithm based on the acquired input and output measurement data. The method is suitable for online modeling, and is superior to the existing method in the aspect of modeling precision.

The invention discloses an uncertainty tracing and self-adaptive decision-making system and method based on a time sequence evolution causality graph, and relates to the technical field of artificial intelligence and data science crossing, and the method comprises the steps: receiving multi-source event observation data, and carrying out the preprocessing; estimating the real time posteriori distribution of the event through a probabilistic time warping module; real time posterior distribution is used as input, a causal graph structure is learned, and the existence probability and effect posterior distribution of edges are jointly estimated; performing uncertainty decomposition on a causal learning result, constructing an uncertainty evolution tree, and tracing the contribution of multi-source uncertainty; dynamically adjusting a decision threshold based on an uncertainty tracing result, and realizing a conservative decision through quantile calculation; and planning a data acquisition action based on a value information criterion, and generating and executing a directional acquisition instruction. According to the method, the probabilistic time warping and dynamic uncertainty tracing mechanism is introduced, so that the timestamp deviation in the multi-source industrial data can be effectively corrected, and the uncertainty source can be analyzed.

The application discloses a video point information standard management method and system based on reverse geographical joint checking, which comprises the following steps: performing joint checking on name semantics and geographical semantics through reverse geographical analysis results of latitude and longitude, and calculating geographical consistency scores; and based on the geographical consistency scores and the standard entity set, automatically completing the missing device attributes. Through multi-field preprocessing + candidate slot map joint modeling, the application can fully utilize the complementary information in the original code, attributes, latitude and longitude, and installation description; the BERT-BiGRU-CRF model and its training process are disclosed through prompting enhancement, which can avoid the problem of insufficient disclosure of the entity recognition model, and improve the entity recognition accuracy of non-standard point names; through the dynamic graph updating mechanism, preferably in the incremental updating mode, the application can continuously absorb new aliases, new device types and artificial review results, and avoid long-term static failure of the knowledge graph.

A method of training an artificial intelligence model includes obtaining a target matrix, obtaining a library matrix including a library of candidate model terms associated with a plurality of candidate models, and generating, based on the target matrix and the library matrix, a coefficient matrix including estimated model coefficients. The coefficient matrix is pruned based on a pruning matrix of coefficient pruning thresholds to generate at least one sparsified coefficient matrix. One or more of the at least one sparsified coefficient matrix is selected as a candidate to be used by the artificial intelligence model in predicting behavior of a physical system.

The invention provides a resume data processing method and device, equipment and a storage medium, relates to the field of data processing, can automatically solve the problem that information expression in a resume is inaccurate and nonstandard, and greatly improves the resume processing efficiency. According to the specific technical scheme, the method comprises the steps of firstly performing word segmentation on target text information in a resume, then obtaining a prediction label representing whether each segmented word is an information entity and representing the position of each segmented word through a pre-training classification model, and screening and splicing to-be-processed information entities according to the prediction labels; and then determining a target standard entity in combination with the similarity between the information entity to be processed and a preset standard information entity, and finally completing the standardization of the target text information to obtain standardized resume data.

The invention relates to an auditing project data management method for digital intelligent power grid construction, which is applied to a data auditing device, the device comprises a power grid cloud server and a project auditing sub-terminal, and the method comprises the following steps: obtaining a user operation instruction of the project auditing sub-terminal through the power grid cloud server, generating user operation record information based on the user operation instruction; constructing an audit resource knowledge base, wherein the audit resource knowledge base comprises project archive information, standard data resource information and basic regulation retrieval information; extracting the user operation record information through an SQL (Structured Query Language) on the basis of the regulation retrieval information and the standard data resource information to obtain feature data information; and carrying out association relationship analysis on the feature data information to generate audit evaluation information. Through the content, high efficiency, intelligence and standardization of data management of the digital intelligent power grid construction auditing project are realized, powerful support is provided for auditing work, auditing quality and efficiency can be improved, and healthy development of a digital intelligent power grid is promoted.

This invention provides a radar motion parameter estimation method and apparatus based on information criterion values. It acquires raw radar echo data, performs range processing and target extraction to obtain a target slow-time complex sequence. Based on the target slow-time complex sequence, it constructs multi-order motion candidate models, calculates the residual sum of squares, and calculates the information criterion value based on the residual sum of squares. The optimal model order is determined based on the information criterion value, and the target radial motion parameters are inverted and output based on the phase coefficients corresponding to the optimal model order. Instead of pre-fixing the use of a first-order, second-order, or third-order model, it determines the optimal model order after constructing multi-order candidate models and evaluating them using the calculated information criterion value. This approach adaptively determines the appropriate motion model order for the target based on the phase evolution characteristics of the signal itself, and completes motion parameter estimation accordingly. This ensures estimation accuracy while suppressing unnecessary model complexity, thus improving its engineering application value.

This invention discloses a method for verifying and confirming numerical prediction thrust spectrum results of propellers, belonging to the field of computational fluid dynamics (CFD) result verification and confirmation. While maintaining consistency between the physical model and boundary conditions, this invention encrypts the CFD calculation settings with joint discrete parameters and quantitatively characterizes simulation errors and uncertainties based on Richardson extrapolation. This invention employs the Lomb-Scargle periodogram method to directly process non-uniformly sampled data, avoiding spectral distortion caused by resampling interpolation and improving the accuracy of spectrum estimation. This invention automatically selects the optimal AR model order through information criteria, models colored noise backgrounds in actual engineering, and achieves adaptive noise modeling. This invention simultaneously evaluates the statistical significance of the global maximum peak value and each expected harmonic frequency in a single Monte Carlo simulation cycle. This invention uses a block bootstrap method to calculate the frequency band power confidence interval, maintaining the local correlation of the time series and achieving reliable confidence interval quantification.

The present invention relates to the field of computer technology, and discloses a method, apparatus, device, and medium for standardizing address information. The present invention obtains the original text of the address information; performs a candidate entity search on the original text to obtain corresponding candidate entities, wherein the candidate entity search includes one or more of a graph database query and a regular matching; and performs a dynamic planning optimal entity path search on the candidate entities based on a pre-created initialization algorithm model to obtain a standardized result of the address information. The present invention solves the problems of poor scalability and low query efficiency on the data side, and low accuracy of address standardization on the model side, thereby optimizing the effect of address information standardization.

The invention provides a coal inspection information processing method, system and device based on artificial intelligence, and relates to the technical field of coal detection information processing, and the method comprises the steps: obtaining original coal inspection information of a client; on the basis of a preset database, supplementing the original coal inspection information to obtain complete coal inspection information; inputting the complete coal inspection information into a preset inspection model to obtain standard coal inspection information output by the preset inspection model; the standard coal inspection information is information sorted according to a preset format; the preset inspection model is a large language model; and recording the standard coal inspection information in a business system. The method has the advantages of high data processing efficiency, high processing result accuracy and low processing cost.

The present invention relates to the technical field of welding groove detection, and specifically provides a method for extracting multi-type welding groove features based on three-dimensional point clouds, including: 1. Measuring the three-dimensional point cloud data of the workpiece; 2. Filtering the three-dimensional point cloud data; 3. Constructing, clustering, merging, and classifying to obtain a clustering result; 4. Constructing a piecewise linear regression error function err(r) for each clustering cluster; 5. Minimizing the error function err(r) to search for the optimal r; 6. Comparing the information criterion value under the current number of segments with the previous value to determine whether to update the number of segments. If necessary, save the current optimal r and loop through steps 4 to 6; otherwise, return the optimal r to obtain a fitted curve; 7. Calculating the groove feature points. The present invention evaluates the quality of the model by using the AIC and BIC methods, defines a method for extracting multi-type welding groove features, and can obtain relatively accurate feature results for each type of welding groove.

The invention provides a business expansion work order information extraction method based on deep learning. The method comprises the steps of obtaining a business expansion work order image; preprocessing the acquired images to obtain a unified image; inputting the unified image into a text recognition model, wherein the text recognition model is used for performing text recognition on the unified image to obtain a predicted text sequence; according to a predefined information extraction mode, an input text is generated based on the prediction text sequence, the input text is input into an information extraction model, and the information extraction model is used for carrying out information extraction on the input text to extract structured key information; correcting errors in the structured information based on business rules and dictionaries, and standardizing the structured key information into a uniform format to obtain an extraction result. By applying the method, the character-level sequence relationship can be optimized, the text recognition robustness in a complex scene is improved, the text recognition accuracy is enhanced, and character recognition in multiple arrangement modes is supported; automatic key information extraction can be realized, and the working efficiency is improved.

This invention relates to the field of data processing technology, and discloses a backend data processing method, apparatus, device, and medium, comprising: standardizing and uploading the basic equipment information of the target device to a backend server; using a preset semantic network for semantic deconstruction and reorganization to generate a structured semantic knowledge graph, thereby deeply revealing the inherent logical relationships between data. Based on this graph, a business indicator heatmap is drawn in conjunction with real-time backend business data to intuitively present risks and performance bottlenecks; a real-time business recommendation strategy is generated in conjunction with customer data to achieve precise marketing decisions; simultaneously, the business logic is modeled and associated with business data to obtain a visualized business process graph, clarifying the business logic and data flow. Finally, the heatmap, recommendation strategy, and process graph are integrated to generate a unified business navigation interface. This invention improves the data integration capability and semantic parsing capability of complex business data during backend data processing.

The invention discloses a combined scene generation method, system, equipment and medium considering wind and light correlation, and relates to the field of new energy power generation dispatching, and the technical scheme is characterized in that the method comprises the steps: obtaining wind and light output data, carrying out marginal probability modeling through multi-distribution and kernel density estimation, and screening an optimal probability model of each variable; based on the optimal marginal distribution, constructing a wind and light output joint probability model by adopting a Copula function, evaluating and screening an optimal Copula model by utilizing a maximum likelihood estimation and information criterion (AIC / BIC), and generating a joint output scene considering correlation; an improved density-variance dynamic time warping clustering (DV-DTW k-Medoids) method is proposed, and a typical daily joint output scene set is generated; a density peak fast search clustering (CFSFDP) algorithm is combined with a DTW distance to generate a clustering decision diagram to identify an extreme scene. According to the invention, the problem of insufficient scene generation typicality caused by the fact that dynamic time similarity and initial center sensitivity are not considered when the wind-solar combined scene is generated in the prior art is solved.

This invention discloses a method and device for intelligent address completion, verification, and fraud analysis. The method includes semantic parsing and structural reconstruction of unstructured address information input by the user, outputting standardized address data with a unified format; using a large language model to perform compliance verification and semantic cleansing on the standardized address data; collecting the user's dynamic operation trajectory and filling logs during the address input process, and combining user attributes, machine learning algorithms, and the large language model to mine abnormal behavior patterns. This invention combines multi-pattern string matching, large language model semantic understanding, and machine learning behavior modeling techniques, combined with the data and business characteristics of the consumer finance field, to construct a complete implementation solution for standardizing customer address information and identifying fraud in the consumer finance field, addressing challenges such as incomplete and non-standardized address information, difficulty in verifying authenticity, and strong concealment of fraudulent behavior in high-concurrency scenarios.

The invention provides a government affair information processing method and device, and relates to the technical field of computers. The government affair information processing method comprises the following steps: acquiring to-be-processed government affair data; performing type identification on the to-be-processed government affair data to obtain a target data type; if it is judged that the preset basic data types include the target data type, obtaining target instruction information corresponding to the to-be-processed government affair data according to the target data type; inputting the target instruction information and the corresponding to-be-processed government affair data into a standard field generation model to obtain structured field information corresponding to the to-be-processed government affair data; the standard field generation model is obtained by training a basic large language model through training data of a basic data type; the training data is sample data with a structured text label; the sample data is an information pair composed of instruction information and input information. According to the method, the government affair information processing efficiency can be improved.

The invention discloses an automatic division method and system for a satellite constellation orbital plane, and relates to the technical field of spaceflight measurement and control. The method comprises the following steps: firstly, calculating an orbital plane unit normal vector based on a satellite orbit inclination angle and an ascending node right ascension, and performing first-level spatial pointing clustering by adopting a density clustering algorithm to form an orbital plane group; secondly, in the same orbital plane group, by taking a semi-major axis and an eccentricity rate as characteristics, automatically determining an optimal shell layer number by utilizing a Gaussian mixture model and a Bayesian information criterion, and realizing second-stage orbital energy shell layer division; and finally, in the same shell layer, sorting and interval statistical analysis are performed on the right ascension of the ascending node, recursive subdivision is performed through dynamic threshold segmentation in combination with a range-based dual-condition verification mechanism, and fine division of a third-level orbital plane is completed. The method realizes full-process automation from orbit data to division results, has the advantages of high division precision, strong adaptability, good engineering practicability and the like, and is suitable for operation and maintenance management of large-scale heterogeneous satellite constellations.

The present invention discloses a method for standardizing business activity information based on the BOR method, which relates to the field of information technology. It mainly aims to standardize business object information, achieve the standardization processing and high-quality conversion from business activities, business information to data elements, thereby facilitating the management and application of business data and serving as the basis for improving the level of business informatization and achieving high-quality development. The method includes: obtaining the business activity information to be processed; determining the attribute information corresponding to the business object in the business activity information; inputting the attribute information into a preset data entity classification model for entity classification to obtain the data entity corresponding to the attribute information; inputting the data entity into a preset subject domain classification model for subject domain classification to obtain the subject domain corresponding to the data entity; and then determining the standardized processing result corresponding to the business activity information. The present invention is applicable to the processing of business information.

According to the gas turbine unit operation condition division method based on the information criterion and the segmentation network provided by the invention, the problem that the time sequence of the power data sequence is not considered in the existing clustering model and classification method is solved, and the historical power data is subjected to recursive division by applying the information criterion; the stability change of the sequence can be accurately identified, and accurate division of different working conditions such as lifting load, stable load, variable load and the like can be realized; and the marked data set is utilized to train a segmentation network model with an encoder-decoder structure, an efficient working condition division model is constructed, real-time power data can be analyzed, a division result can be output, the accuracy, efficiency and adaptability of working condition division of the power curve of the gas turbine and the combined cycle unit are remarkably improved, and the working condition division efficiency of the gas turbine and the combined cycle unit is improved. Reliable basis is provided for analyzing operation monitoring data such as vibration, temperature and pressure under different working conditions, and the method has important practical engineering significance for realizing refined gas turbine state monitoring and fault diagnosis.

The invention provides a dam safety monitoring information standardization integration method, system and device in a BIM environment and a storage medium, the IFC entity used for describing a dam safety monitoring system and a corresponding attribute set are expanded in a newly added entity definition mode, and a dam safety monitoring system information model can be constructed based on the expanded IFC standard; perfecting a dam project basic data system based on an IFC data format; according to the data model fusion method based on the cosine similarity algorithm provided by the invention, automatic and standardized storage of mass monitoring data can be realized, optimization of a monitoring data reading process under a BIM framework is facilitated, and consistency of information storage is ensured; according to the method, standardized storage of a dam safety monitoring system and monitoring data can be realized at the same time, construction of a dam project full life cycle data collection platform based on a unified data standard is facilitated, efficient data storage and sharing are realized, and a solid data basis is provided for BIM upper-layer application such as dam operation state analysis.

The invention relates to the field of power grid monitoring information processing, in particular to a power grid monitoring information standardized mapping method, device and equipment. Analyzing the monitoring information table and the telecontrol configuration description file, and respectively extracting a point number and a Chinese description and a point number and a data object reference name corresponding to each piece of monitoring information in the monitoring information table and the telecontrol configuration description file; the point number is used as an association key, and the Chinese description of the same monitoring information is matched with the data object reference name; for each piece of monitoring information, respectively carrying out similarity calculation on the Chinese description and the data object reference name of each piece of monitoring information and a standard Chinese description and a standard data object reference name of each piece of standard monitoring information in a standard information model library to obtain a corresponding Chinese description similarity and a corresponding data object reference name similarity; and on the basis of the double-similarity calculation result, judging standard monitoring information to which the monitoring information should be mapped, and establishing a standardized mapping relationship. And a unified and reliable data basis is provided for deep utilization of monitoring information in advanced applications such as risk early warning and fault diagnosis.

The application discloses a data center load prediction method based on ensemble learning, and belongs to the field of data center load prediction, and comprises the following steps: S1, collecting historical operation data of data center electric load and its influencing factors, and establishing a feature database; S2, constructing a SARIMA prediction model based on the feature database, and estimating parameters of the SARIMA prediction model according to information criteria AIC and BIC; S3, constructing an LSTM prediction model; S4, constructing an ensemble learning neural network dataset based on the SARIMA prediction model and the LSTM prediction model; S5, training the ensemble learning neural network dataset to generate an FC network of summation weights; and S6, obtaining a final prediction result through the FC network of summation weights of the SARIMA prediction model and the LSTM prediction model. The application can improve the accuracy of data center load prediction and is suitable for complex data center load prediction tasks.

The invention provides a method and system for automatically maintaining BIM model data based on a lightweight engine, and relates to the technical field of BIM, and the method comprises the steps: defining a BIM model attribute information standard of each stage of a target project; on the basis of the BIM model attribute information standard, quality inspection is conducted on all component data in the BIM model; the quality inspection comprises component classification coding inspection and component attribute inspection which are performed in sequence; component data passing the quality inspection are automatically added to BIM model data; and correcting the component data which does not pass the quality inspection according to the reason of failure, and adding the corrected component data to the BIM model data. According to the method provided by the invention, a large amount of BIM model data is quickly processed through a lightweight engine, the manual operation time is shortened, the working efficiency is improved, the accuracy and consistency of the BIM model data are ensured through an automatic checking and correcting mechanism, and the method can adapt to BIM model attribute requirements of different projects and is easy to expand and apply.

The invention discloses an automatic jurisdiction rule matching method based on a knowledge graph, and the method comprises the steps: inputting case information which comprises a case type, a mark amount, a residence of a party, and a cause of action; data preprocessing: standardizing the input case information; constructing a knowledge graph; carrying out dynamic weight calculation; carrying out multi-dimensional matching; and judging whether the matched jurisdiction courts have conflicts or not, if not, outputting the jurisdiction courts, and otherwise, solving the conflicts through a preset priority rule. According to the method, the knowledge graph is introduced to integrate legal provisions, cases and trial elements, so that structured expression of rules is realized; the priorities of different jurisdiction factors are quantified through a dynamic weight model, and the matching accuracy is improved; the method is suitable for intelligent determination of the jurisdiction in the case filing stage.