24 results about "Data scatter" patented technology

A hardware accelerator and acceleration method based on a VisionTransformer neural network, the accelerator is to deploy the VisionTransformer neural network on a ZYNQ development platform; the acceleration method is: an ARM processor stores a feature picture into a DDR memory, the read data is dispersed to an input cache and a weight cache, the processed feature picture is input to an on-chip cache unit, the processed data is sent to a PL end, the hardware IP of the PL end is configured, and read-write operation is performed at the same time, the final calculation result is obtained, written into the DDR memory, the data in the DDR memory is taken out and probability operation is completed, the probability operation result is transmitted to a PC, a PetaLinux operating system is transplanted to the hardware accelerator system of the VisionTransformer neural network, and a prediction result of inference is obtained from the PC; the neural network structure is optimized through the parallel method of multiple input and multiple output channels, the calculation speed is fast, the hardware resource occupation is low, the recognition accuracy is high, and the image classification task can be efficiently completed.

This invention relates to the field of reservoir engineering management technology, specifically a reservoir lifecycle management system based on spatiotemporal matrix and digital twin. The system includes a digital twin model construction module, a spatiotemporal matrix module, a data fusion module, a lifecycle simulation module, a real-time data acquisition and update module, a model dynamic adjustment module, and a management decision support module. The system uses a spatiotemporal matrix to uniformly organize and manage multi-source data throughout the reservoir's lifecycle, utilizes a digital twin model to construct a virtual representation of the reservoir, achieves bidirectional synchronization through data fusion, and dynamically adjusts model parameters based on real-time data, ultimately generating lifecycle management decision outputs. This invention solves the problems of scattered data and static models in existing technologies, improving the accuracy and reliability of reservoir management.

This invention discloses a multi-level approval management method for material requisition applications across multiple business types. Through a hierarchical grid design, it achieves five-level approval (workshop planner → deputy workshop director → workshop director → equipment department staff → equipment department head) and full-chain operation tracking. Simultaneously, it significantly improves approval efficiency by using front-end color-coding to distinguish status and a one-click filtering function. The system covers multiple business scenarios, completely recording the personnel, address, and time of each operation, achieving audit-level traceability. It also integrates with SAP orders, contract numbers, and inventory, solving problems such as chaotic material requisition processes, inefficient approvals, fragmented data, and difficult cost accounting in traditional methods. This invention is suitable for equipment and material management scenarios in large industrial enterprises. It has been practically implemented in state-owned steel plants for many years.

The application discloses a kind of construction method and system of port production factor digital twinborn model, specifically relates to port digital twinborn technical field, for solving the problem that multiple source factors are difficult to unify;The application constructs the unified object merging of port production factor, space-time alignment, relationship identification and incremental updating mechanism, realizes the unified digital mapping and dynamic correlation expression of shore bridge, field bridge, transport vehicle, berth, container location, ship and operation task, solves the problem of scattered data, identification fragmentation, space-time inconsistency and operation chain difficult continuous tracking, and through unified object key, relationship graph, state check and local update, improve the accuracy, real-time and stability of port production monitoring, collaborative scheduling and operation control.

The present application belongs to the technical field of medical health, and particularly relates to a youth physique dynamic monitoring and intervention method and system based on multi-modal fusion. Physique data is collected and processed through a stereoscopic vision sensor array and a deep learning algorithm, multi-modal psychological data is collected through an AI drawing interactive process, and a personalized intervention scheme is generated through cross-modal semantic fusion and intelligent decision-making, thereby solving the problems of single data dimension, imperfect evaluation system, homogeneous intervention and scattered data of a traditional monitoring system, and realizing full-process intelligent and personalized dynamic management of youth physique health.

This invention provides a method, apparatus, and storage medium for optimizing petrochemical equipment data, belonging to the field of petrochemical technology. The method includes: importing multiple equipment identifiers and multiple original multi-source heterogeneous datasets of petrochemical equipment; and performing labeling processing on each original multi-source heterogeneous dataset of petrochemical equipment according to each equipment identifier to obtain labeled multi-source heterogeneous datasets of petrochemical equipment. This invention achieves a closed-loop construction from raw industrial data to high-quality domain model training data and a steady improvement in the quality of generated data. It can transform the raw, messy, multimodal data of the petrochemical industry into high-value training datasets with clear structure, accurate labeling, and controllable quality, laying a solid data foundation for building large-scale professional petrochemical models and solving the problems of industry data dispersion, labeling difficulties, low quality, and knowledge gaps.

The application discloses a large-scale knowledge graph construction method for a financial field, relates to the technical field of knowledge graphs, and aims to solve the diversified large-scale knowledge graph construction problem caused by the heterogeneity and independence of large-scale field knowledge. The application comprises the following steps: S1, acquiring financial and economic data from an open knowledge base, performing concept extraction and knowledge extraction, and then constructing a large-scale knowledge graph; S2, constructing a common-sense knowledge base based on the large-scale knowledge graph; S3, expanding the common-sense knowledge base by using the large-scale knowledge graph; and S4, performing common-sense knowledge base-large-scale knowledge graph joint representation learning by using a graph representation learning module. The application integrates multiple data sources and various types of knowledge in the financial field, solves the problems of data dispersion, redundancy and inconsistency existing in the key knowledge base in the financial field, provides a more comprehensive and consistent knowledge view, realizes the sharing and interaction of large-scale data, and makes the query and reasoning of financial knowledge more efficient.

The application discloses a rigid contact line abnormal data detection system and method based on artificial intelligence, and relates to the technical field of abnormal data detection.The system comprises a full-life-cycle data construction module, a wear law mining module, a real-time data acquisition and matching module, and an abnormality determination and early warning archiving module;the method constructs a full-life-cycle database, mines wear laws of the same section and the same working condition, matches real-time and historical data, adopts real-time speed increase comparison and future trend prediction to determine abnormalities in two dimensions and trigger graded early warnings.The application solves the problems of the prior art, such as scattered historical data, static determination, and passive operation and maintenance, realizes precise early warning of potential abnormalities, promotes the shift of operation and maintenance from passive remediation to active prevention, and meets the demand of intelligent preventive maintenance of rail transit.

The application discloses a stope overburden load three-dimensional model construction method, relates to the model construction technical field, and comprises the following steps: obtaining a sample data scatter diagram based on sample data of an overburden to be detected; obtaining a first stress threshold value and a second stress threshold value based on the sample data scatter diagram; obtaining a stress representation value based on the first stress threshold value and the second stress threshold value; obtaining a stress change function based on the stress representation value; obtaining generated data points based on the overburden to be detected; obtaining a first detection point based on the generated data points; obtaining a real-time stress value based on the first detection point and the stress change function; obtaining a function fitting stress based on the stress change function and the real-time stress value; and generating a stope overburden load three-dimensional model based on the function fitting stress. The application is used to solve the problem of low stress acquisition accuracy of an instrument at an undetected position in the prior art model construction technology, thereby improving the accuracy of the stope overburden load three-dimensional model construction.

The application discloses a data security union processing method and system, belongs to the technical field of data management, solves the problem that due to the inherent defect of hash algorithm, hash collision cannot be completely avoided, the introduction of cross-shard retrieval will destroy the shard parallelism, significantly increase the merging complexity, and lead to the problem that the shard number matching may cause the actual associated data to be scattered in different shards due to hash collision and cross-shard retrieval is needed. Including the following steps: dynamic sharding, according to the data heat of the primary key. Through the dynamic sharding, distributed bloom filter conflict pre-checking and local redirection query technology, combined with multi-thread parallel processing and real-time load balancing optimization, the proportion of cross-shard retrieval caused by hash collision is limited, the hash collision rate is reduced while the shard parallelism is ensured, the super large data set is merged, and the problem that the shard number matching may cause the actual associated data to be scattered in different shards due to hash collision and cross-shard retrieval is needed.

The application discloses a big data center construction project data management system, method, equipment and medium, relates to the technical field of big data management, and comprises a big data center module, a data management module and a system integration module. The method disclosed by the application realizes the orderly management and collaborative circulation of construction project data between different operation levels and processing stages by constructing a data management scheme combining layered bearing and partition processing, forms a unified data processing link for data collection, storage, analysis and access, realizes the centralized access and management of business systems and real-time data at the plant end level, the company level and the group level through a unified data integration mechanism, reduces the situation of scattered data storage and repeated processing, improves the overall collaboration of construction project data management and the flexibility of data utilization, and is suitable for the management requirements of big data center construction projects.

The present application relates to the field of urban planning management, and more particularly to a multi-modal planning research data processing method and system based on street view analysis, in response to a street view analysis instruction for a planning research project, a planning semantic target associated with the project is issued to a mobile terminal; based on the issued result, the mobile terminal is used to execute acquisition of field collection data; the field collection data is acquired from the mobile terminal; according to the field collection data, the chain street view analysis result is associated and fused with the corresponding research track, multimedia data and attribute information according to the time stamp and positioning information; the processed data is visualized and displayed, including displaying the spatial distribution and statistical chart of the chain street view analysis result on an electronic map; based on the fused data, a structured data packet and a research report conforming to a preset format are generated and output. The pain points of scattered data and low efficiency of manual integration are solved.

This invention relates to the field of optical module communication technology and discloses an online dynamic adaptive compensation method for optical module performance degradation. It solves the technical problems of insufficient trend prediction capability, lack of adaptive dynamic adjustment capability, data dispersion, and lack of global coordination in existing optical module performance degradation monitoring and compensation methods. This invention acquires and preprocesses multi-source heterogeneous parameters, extracts core degradation features using a self-supervised model, and then uses a degradation trend prediction network to obtain core performance prediction values. This effectively captures the mutual influence between multiple processes and the evolution law of parameters, significantly improving the accuracy of performance degradation trend prediction. Furthermore, this invention automatically calculates the optimal compensation factor vector through a compensation decision network and uses proportional-integral-differential calculations to precisely adjust the production equipment. It can dynamically adjust the compensation amount according to the actual degradation state, and collects the actual values ​​and improvement rates after compensation to update each network, forming a closed-loop adaptive optimization.

This application belongs to the field of data processing technology and discloses a payroll management method, apparatus, system, equipment, and medium based on a feedback mechanism. The method includes: receiving user-configured payroll base parameters and rate base parameters through payroll base parameter controls and rate base parameter controls provided by a front-end interactive interface; responding to the triggering of a payroll calculation instruction, determining whether the target month's payroll data has been imported at a preset payroll calculation node; if so, calculating directly using the imported payroll data; if not, calculating by combining the configured payroll base parameters, rate base parameters, and attendance data obtained through the front-end interactive interface; receiving adjusted payroll data for the target calculation month uploaded by the user through the front-end interactive interface before a future payroll calculation node, and recalculating based on the preliminary calculation results to provide payroll compensation. This solves the technical problems of low efficiency, error-proneness, data fragmentation, and inflexible adjustments in existing technologies.

This invention discloses a method, system, device, and storage medium for dynamic tracking of multi-source indicators. The method includes: standardizing multi-source business and financial data from different insurance business systems; simultaneously calculating core insurance expense difference indicators and multi-dimensional derived analysis indicators based on the standardized insurance business and financial data; visually displaying the core insurance expense difference indicators, multi-dimensional derived analysis indicators, related deviation information, and attribution conclusions in real time; and monitoring the real-time visualized content according to insurance business early warning rules, and generating early warning information when anomalies are detected. This invention solves the problems of scattered business and financial data, inconsistent definitions, low efficiency and high error rate of manual processing, limitations of single-dimensional indicator calculation, and delayed monitoring and early warning. It improves the ability to integrate and standardize multi-source data, enhances the depth of simultaneous calculation and analysis of multi-dimensional indicators, and improves the timeliness of dynamic visualization tracking and early warning.

This invention relates to an artificial intelligence-based early pancreatic cancer screening system, belonging to the field of early pancreatic cancer screening technology. To address the problem of existing screening systems having scattered and incomplete data collection, leading to inconsistent data and affecting screening results, this invention establishes a database containing pancreatic cancer disease characteristics. The database encompasses multi-dimensional feature information, presenting corresponding pancreatic cancer characteristic data according to different stages of disease development. This AI-based early pancreatic cancer screening system extracts key features through a feature extraction module, greatly reducing data redundancy and significantly improving data processing efficiency. The predictive analysis module uses data analysis and algorithms to perform authenticity screening on the extracted raw data and then performs early pancreatic cancer screening on the authentic data, avoiding misjudgments caused by invalid data interference and improving the quality and accuracy of early pancreatic cancer screening data.

The present application belongs to the technical field of oil extraction equipment information processing method, and discloses an oil pumping unit operation data processing system based on edge cloud cooperation, comprising a plurality of edge end nodes deployed at different oil pumping well sites, each edge end node being in communication connection with a cloud server, the edge end node and the cloud server being in communication connection with a visualization platform, the visualization platform receiving and displaying working condition diagnosis results; each edge end node is used for collecting operation data of a local oil pumping unit, constructing a local training data set, training a local working condition diagnosis model, obtaining a local model weight and uploading it to the cloud server; the cloud server aggregates each local model weight received, generates a global model weight, and issues it to each edge end node to update the local working condition diagnosis model of each edge end node, and the diagnosis model output obtains a visual working condition diagnosis result. The system solves the problem of scattered pumping unit indicator diagram data, which is difficult to model and diagnose.

The application discloses a kind of production line ERP-MES fusion's whole-process production report intelligent collection and unmarked material statistics method, and relates to the technical field of industrial manufacturing execution system and enterprise resource planning system data fusion.The application builds whole-process standardization data base, adopts contract number+contract hash value+plan number cross-system unique identification, realizes whole-link data and reverse trace-through;Through lightweight temporary table extremely simple reuse, CTE one-step calculation and batch direct change main table strategy, realize large data volume report second-level response;Establish 100% accurate identification and whole node statistics mechanism of unmarked material, adapt to the unified standard output of multiple business types.The application solves the pain points such as data dispersion, system barrier, unmarked material control loss, report generation slow, etc., and the statistical efficiency is improved by more than 95%, which has been landed in actual industrial production line MES system, with mature engineering value and full-industry promotion ability.

The application relates to a data asset supervision method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: acquiring asset business data of a heterogeneous system service; performing asset detection according to a plurality of preset asset detection models and the asset business data to obtain original asset data; acquiring multi-terminal data of an Internet of Things database; performing standardization processing on the original asset data and the multi-terminal data according to a preset rule library to obtain standardized asset data; and performing asset supervision on the standardized asset data according to a pre-trained abnormality monitoring model. The method can solve technical problems such as incomplete asset coverage, state monitoring lag, scattered data not shared, and disconnection between security management and asset monitoring in existing data asset management.

The application discloses a multi-line sample sending and test data management and retest closed-loop visualization method, and aims at process media such as acid liquid, passivation liquid and zinc liquid to realize quality closed-loop of one-time test, unqualified forced retest and secondary sample sending and inputting. The system automatically matches test indexes according to a unit, supports multi-parameter trend curve drawing, and records operation traces throughout the whole process. The method effectively solves problems of no retest, scattered data and invisible trend in traditional test, and significantly improves the standardization and traceability of metallurgical production line process medium control. The application has been used in a state-owned steel plant for many years.

The application discloses a data processing method, system and computer program product, the method comprises: acquiring multi-source business data and original voucher files; parsing and field extraction, constructing a unified structured data set with business matters as dimensions, including business matter records, voucher file records, association relationship records and state records; extracting matching fields to generate matching feature vectors; calculating association scores and updating association relationship records according to the comparison result with a preset threshold; performing data integrity check and anomaly identification; calculating automatic calculation values and determining confidence levels; receiving structured review instructions and updating state records; and storing and generating traceable archival data according to business matter identification. The application solves the problems of scattered multi-source business data, weak association, lagging anomaly identification and lack of unified archives for processing results, and improves the structured processing capability and result traceability capability.

The present disclosure provides a method, system and medium for intelligent typing of chikungunya virus, obtains CHIKV sequence data from a public database, performs quality control, uses an artificial intelligence model to extract typing reference sequences of CHIKV, and then compares the sequence data with the typing reference sequences one by one. If the comparison result of the sequence data and a certain typing reference sequence is similarity>99% and sequence coverage>99%, the sequence information of the typing reference sequence is supplemented to the sequence data, and a typing sequence database is generated. The present disclosure uses an artificial intelligence model to realize automatic extraction, semantic alignment and standardization of multi-source data, solves the bottleneck of data dispersion and metadata loss, realizes intelligent typing of chikungunya virus, encapsulates complex bioinformatics processes into a "one-key" online tool, significantly reduces the technical threshold, enables grassroots personnel to independently complete high-level analysis, and provides real-time, visual molecular evidence and risk prompts for epidemic prevention and control.

The application discloses a kind of distribution network equipment health degree evaluation and intelligent operation and maintenance method based on negative area integral model.The method is by building the dynamic negative area integral calculation model of fusion equipment outage, operating abnormity and multi-source data such as customer complaint, realize the accurate quantitative evaluation of the health state of line, distribution transformer and other equipment, and generate equipment panoramic health portrait for visual early warning.System is based on the evaluation result automatically triggers or intelligently recommends the operation and maintenance strategy such as differentiated patrol, maintenance, and pushes to the scene execution through mobile application.At the same time, the integral model is continuously self-optimized by using operation and maintenance closed-loop feedback data.The application effectively solves the problems of data dispersion, evaluation lag and decision-making relying on experience in traditional operation and maintenance, improves the initiative, accuracy and intelligent level of distribution network operation and maintenance, and guarantees power supply reliability and service quality.