36 results about "Data abstraction" patented technology

The invention discloses an internet of things multi-protocol fusion limo intelligent control system, which relates to the field of limo intelligent control and comprises a hardware irrelevant layer, a multi-protocol adaptation layer, a data abstraction layer, a programmable logic layer, a security and redundancy mechanism and an edge AI module. According to the invention, a physical bus is packaged into an independent module, an MCU / SoC and a protocol transceiver are built in, hot plugging is realized through a standardized PC I e / EdgeConn backboard, a core processing board only exposes a unified backboard interface, module drive localization execution is realized, master control firmware does not need to sense protocol details, only the module needs to be replaced when a protocol is newly added, and the master control firmware or upper software does not need to be modified. The defect that the whole vehicle firmware needs to be refreshed when a bus is replaced in a traditional system is overcome, each protocol corresponds to an independent drive micro-service deployed in a containerization mode, the drive micro-service analyzes a message into a unified data model and uploads the unified data model to a data abstraction layer, a downlink control instruction is packaged and issued according to a target protocol, and it is guaranteed that the protocols are diversified and interfaces are consistent.

The invention relates to the technical field of power distribution network monitoring, and discloses a power distribution network load monitoring system and method, and the system comprises a load data collection module which can carry out the hierarchical division of an operation region to obtain basic monitoring units and extract a collection interval; the multi-source data integration processing module collects multi-category data such as load power and voltage fluctuation according to a sampling period and generates time sequence integration data; the state feature construction module abstracts a load state according to time sequence data to obtain a model, and extracts multi-dimensional features such as capacity bearing and operation stability; the abnormal working condition discrimination and evaluation module identifies local anomalies according to the multi-dimensional features and determines overall anomalies; and the monitoring strategy regulation and control module dynamically adjusts monitoring parameters and monitors the real-time load state of the power distribution network in real time. According to the system, layered monitoring and abnormity identification of the load of the power distribution network are realized, and the monitoring accuracy and the dynamic regulation and control capability are improved.

The invention provides an undercarriage system scheme design method based on heterogeneous design simulation data management, which comprises the following steps of: firstly, analyzing, abstracting and standardizing different types of heterogeneous design and simulation data in a development process of an undercarriage system, and constructing a uniform abstract data model; secondly, according to the defined semantic rules, an undercarriage semantic ontology model is constructed, then the semantic rules are loaded through a semantic reasoning engine, and processed data are associated; according to the research and development process of the undercarriage system and the semantic association relationship between the data, data storage and display are realized; after data abstract analysis, semantic association and unified storage are completed, functional information, design information and simulation information are further derived based on an undercarriage semantic ontology and a reasoning result, so that an undercarriage system development scheme driven by requirements is constructed. According to the method, multi-source heterogeneous design and simulation data generated in the research and development process of the undercarriage system can be effectively managed, and a design scheme of the undercarriage system is constructed on the basis of the multi-source heterogeneous design and simulation data.

The invention discloses an equipment manufacturing energy consumption and state visual monitoring method based on an industrial internet of things, solves the problem of passive fixation of an industrial monitoring visualization technology in the prior art, and realizes autonomous and intelligent generation of an optimal visual view matched with a real-time working condition. The method comprises the following steps: collecting industrial field multi-source heterogeneous data in real time, abstracting a monitoring object as an entity node, and constructing a feature vector containing static and dynamic attributes; constructing and updating a dynamic knowledge graph based on nodes and features, wherein relation edges of the dynamic knowledge graph comprise dynamic edges generated by static connection and real-time analysis; inputting the atlas into a pre-trained multi-attention graph neural network to obtain a display priority score and a visual affinity matrix of each node; calculating coordinates of all nodes on an interface through a layout optimization algorithm in combination with the priority and the affinity matrix; and generating a visual drawing instruction according to the coordinates and the node attributes, and driving a display device to render a monitoring interface.

The invention discloses a nuclear power platform data edge processing method and a storage medium, the nuclear power platform comprises a platform layer and an edge layer, and the method comprises the following steps: data standardization: obtaining data and carrying out data standardization processing; data structuring: abstracting the standardized data into structured data; data sinking: sinking the structured data to the edge layer according to a sinking decision, and processing the structured data by the edge layer; and data synchronization and feedback: synchronizing the data processed by the edge layer so as to realize nuclear power platform data edge processing. Through edge processing of the nuclear power platform data, the delay of data transmission is reduced, the burden of the central server is reduced, the response speed and reliability of the system are improved, and the operation efficiency and safety of the nuclear power station are effectively improved.

Telecommunications systems, computer devices and systems, and methods are provided. One example method, performed by a data abstraction layer of a telecommunications system for a network, includes accessing data elements associated with a network function of the network, performing abstraction on the data elements to generate one or more data products, receiving a customer request for a service on the network, identifying the data product related the service, and provisioning the identified data product to the customer. The method may further include identifying proprietary data and private data of the data elements based on a predefined policy and removing proprietary data and private data before performing abstraction.

The invention discloses a hybrid expert model deployment method and device based on a heterogeneous hardware equipment cluster, equipment and a medium, and relates to the technical field of artificial intelligence, and the method comprises the steps: carrying out the data abstraction processing of all equipment in a target heterogeneous hardware equipment cluster, so as to obtain an initial equipment description vector, compiling the target expert model file into a binary intermediate representation; generating a hardware probe for the target equipment based on the hardware type of the target equipment in the target heterogeneous hardware equipment cluster and the binary intermediate representation, and obtaining a real-time equipment description vector based on hardware; determining a hardware sensing routing weight of each device in the target heterogeneous hardware device cluster based on the real-time device description vector, and determining a to-be-activated expert model file from the target expert model file based on the sensing routing weight; and distributing the to-be-activated expert model file to the target heterogeneous hardware equipment cluster for deployment and installation. Therefore, heterogeneous deployment of the hybrid expert model can be realized through unified arrangement.

The application provides a kind of detection method and system of Internet of Things equipment credential trigger type backdoor, comprising: the binary code library is formed by decompressing device firmware;Executable file in code library is analyzed reversely, and backdoor verification key code is positioned using string comparison function and backdoor model feature;The code segment related to backdoor verification key code is stripped from original program, and complete credential generation process code is obtained;Each credential generation process code is subjected to data abstraction and process modeling processing, forming formal description of verification credential, and the semantics and uncertainty of each input variable in formal description are calculated;According to the semantics and uncertainty of each input variable, it is judged whether the verification credential is predictable, and whether the device has a backdoor that can be triggered by predictable credential.The application can automatically detect whether there is a backdoor triggered by credential in the device through static analysis of binary program in Internet of Things firmware.

The invention relates to a computer architecture (110) enabling data transfer between several functional units. A functional unit supply structure (111) provides several functional units. A selection structure (112) provides a selection of functional units and linking information, the selected functional units being linked to perform predefined data processing. An interface structure (113) provides an interface for data transfer between linked selected functional units, this interface being designed to transform at least a portion of the output data content of the selected functional unit into input data of the other functional unit, according to the selected functional units and the linking information. [Abstract figure] Fig. 1

A system, method, and apparatus for generating a data visualization are disclosed. The method includes (i) obtaining a natural language query, (ii) determining an intent of the natural language query, (iii) generating one or more data requests to selected one or more data sources, the one or more data requests based at least in part on the intent, (iv) abstracting result data to obtain a data abstraction, and (iv) generating one or more data requests to selected one or more data sources based at least in part on the data abstraction. The result data is responsive to the one or more data requests, and (v) generating a visualization for the result data based at least in part on the data abstraction.

The invention discloses a big data management method and system based on semantic analysis, and relates to the technical field of big data management. A multilayer semantic abstract data organization structure comprising a base layer, a mode layer and a statistical layer is constructed, real-time semantic analysis is performed on input data, and semantic descriptors are generated; the high-frequency data instance in the base layer is stored to a corresponding level according to a central semantic index, a semantic folding operation is triggered based on a preset rule, the high-frequency data instance in the base layer is automatically abstracted into a modeled representation of a mode layer, and similar representations in the mode layer are further aggregated into an aggregated representation of a statistical layer; when the query is responded, the system analyzes a query intention and determines a target data abstraction level, high-level aggregation or modeling representation is restored into bottom-layer detailed data as required through reverse expansion operation, and a query result is generated; through the multi-layer semantic abstract data organization structure and the data folding operation, the storage efficiency of mass complex data and the performance of multi-granularity intelligent query are effectively improved.

The invention relates to a computer implemented method for efficient access to a database via a functional model-based procedure by translating a sequence of operations that are stored in a memory and that is described in the procedure that is specified using a generalized programming language into an intermediate database abstraction layer access language, whereby the intermediate data abstraction layer not being SQL. The method comprises: generating by a processor: an initial builder corresponding to an original originating source, zero or more intermediate operations and a terminating operation; storing in the memory, the sequence of operations, and translating the builder or parts of the builder is translated to at least one operation using the intermediate database abstraction layer access language upon specifying the terminating operation.

A computer-implemented method according to one embodiment includes creating a data abstraction that includes a set of memory resources and variables associated with the memory resources. The computer-implemented method also includes implementing the data abstraction within a programming language environment. A system according to another embodiment, includes a processor. The system also includes logic that is integrated with the processor, executable by the processor, or integrated with and executable by the processor. Moreover, the logic is configured to: create a data abstraction that includes a set of memory resources and variables associated with the memory resources. The logic is further configured to: implement the data abstraction within a programming language environment.

The embodiment of the invention provides a recursive query processing method and a related device, and relates to the technical field of big data, the method comprises the following steps: when a statement to be processed is determined to be a recursive query statement, extracting a table name and a recursive condition in the recursive query statement; according to the table name and the recursion condition, constructing a directed graph, and based on the directed graph, obtaining the maximum depth of nodes and the total number of paths; splitting the directed graph into a plurality of sub-graphs according to the maximum depth and the total number of paths; and determining a link from a starting node to a target node according to the plurality of sub-graphs, and outputting the link. According to the method, the recursive query table data is abstracted into the directed graph, so that the success rate and efficiency of recursive query can be improved.

The invention relates to the technical field of processors, in particular to a tile block instruction set architecture and a processing method, and the tile block instruction set architecture comprises a tile block parallel processing system control instruction, a tile block descriptor register configuration instruction, a tile block data handling instruction, a tile block data copying instruction and a tile block arithmetic instruction. According to the tile block instruction set architecture processing method, one or more tile block operation units, one or more tile block storage management units and one or more tile block calculation task scheduling and synchronization units are combined into a tile block parallel processing system, so that tile block instructions are allowed to be processed in parallel, and tile block-based data operation is completed. The architecture aims at solving the problems that when an existing processing system processes two-dimensional data blocks, instruction overhead is too large, data abstraction is insufficient, and the coupling degree of software and hardware is too high.

The invention discloses a system and method for uniformly analyzing and rendering Gerber and ODB + + in real time, and the method comprises the following steps: S1, analyzing Gerber format data or ODB + + format data, and uniformly converting each layer graph into a polygon expression; s2, organizing the polygon according to a layer sequence, and triangularizing the polygon into mesh format data; and S3, performing real-time laminated rendering on the mesh data by using a GPU through a graphic rendering pipeline. Compared with the prior art, the method has the advantages that through the innovative unified data abstraction and GPU acceleration rendering technology, various pain points of an existing PCB data processing method in the aspects of precision, memory, efficiency and uniformity are effectively solved, and a more advanced and practical solution is provided for the field of PCB design, manufacturing and detection.

The invention discloses a user consumption behavior prediction method under cross-tenant privacy protection and a related product. After a consumption prediction request of a tenant for a target user is received, interaction data and current geographic position data of the target user are firstly obtained. Converting the interaction data into a first semantic note for describing the consumption motivation of the user by using a large language model, and abstracting the geographic position data into a regional commercial attribute tag which does not contain specific street information to obtain a first tag; vectorizing and splicing the first semantic note and the first tag to form a target feature vector; and then a reference vector with high similarity with the target feature vector is retrieved in a public vector database, group features associated with the reference vector are combined, reasoning analysis is performed through a large language model, and finally a consumption intention prediction result of the target user is obtained. According to the invention, on the premise of ensuring the privacy security of the user, the cross-brand data can be effectively utilized to carry out precision marketing.

The invention relates to a platform data file distributed management method and system. The invention aims to solve the problems of load imbalance, low data reliability, difficulty in file management and the like in traditional distributed file storage. The management method is realized through a distributed development kit, and specifically comprises the following steps of: segmenting a file into a plurality of blocks, and storing the file blocks on different servers; performing multiple backups on each file block to enhance the reliability of the data; abstracting metadata of the file into a directory tree for management so as to simplify file management operation; a unified file management function is provided for the outside through the distributed development kit, and a user only needs to interact with the kit and does not need to care about bottom storage details. The method has the advantages of being high in fault tolerance, suitable for batch processing, suitable for storing large files and the like, can run on a low-cost machine, effectively reduces hardware cost, and meanwhile improves the concurrent access capacity of a system and the high availability of data. The management system comprises a file segmentation module, a file storage module, a backup module, a metadata management module, a distributed development kit and the like so as to realize the management method.

The invention discloses an autonomous mobile platform adaptive control method and system fusing multi-modal feedback and visual thinking chain reasoning, and the method comprises the steps: abstracting environment data collected by a plurality of sensors and perceptron in real time into a structured expression through an expert abstraction unit, and inputting the structured expression and user voice instruction information into a large language model; and the large language model generates a segmented dynamic reinforcement learning dense reward function. A mixed attention mechanism of visual thinking chain reasoning is adopted, multiple candidate paths and corresponding sub-target image sequences are generated through autoregression, and visualization is carried out through a human-computer interface. A user can feed back a correction instruction through voice and feed back a correction target visually. And finally, performing feature alignment on the multi-modal feedback of the user, updating the dense reward function and re-planning the path. The system can also record a user feedback result and train a personalized user preference model. The method has the advantages of strong generalization ability, high feedback efficiency and the like.

This invention relates to the field of multidimensional data processing technology. It discloses a method and system for constructing a multidimensional data model for thermal power equipment. The method includes: collecting basic and indicator data of the thermal power equipment; using machine learning algorithms to identify data relationships between different business domains and constructing a multi-level data abstract model; defining the hierarchical structure and specific attributes in the data model, as well as the relationships between each layer; and dynamically updating the data relationships to adapt to changes in the data source, forming a data model that supports the entire lifecycle management of the equipment. This method combines advanced machine learning algorithms to identify data relationships between different business domains to support comprehensive equipment management.

The application belongs to the fields of artificial intelligence, computer vision, natural language processing and the like, and is a document-level image-text review sentiment classification method fusing common sense knowledge. First, review data collected from the Internet is abstracted into structured data, and after data preprocessing work such as Chinese word segmentation and image normalization, the feature representation of the review data is obtained by sequentially passing through single-modal sentiment feature representation learning and multi-modal sentiment feature fusion, and finally the fusion features are input into a classifier or a multi-layer perceptron is directly used to obtain a sentiment probability distribution. The method can effectively improve the accuracy of sentiment classification on the one hand by fusing common sense knowledge to strengthen text semantic understanding, and on the other hand by introducing related image information to assist in improving text sentiment expression.

The application discloses a kind of software defect automatic positioning methods based on heterogeneous graph representation learning.The method is first obtained from the required data in project, by abstracting various data into different types of vertex and edge, constructs a heterogeneous graph;Then the information in heterogeneous graph is aggregated by three modules, and the feature representation of each method in source program is obtained;Finally, a prediction model is constructed based on convolutional neural networks (CNN) to calculate the probability of defects existing in each method, and the defect positioning result is obtained.At the same time, the application introduces a resampling strategy, which effectively overcomes the class imbalance problem in the problem.Through the method of the application, the effectiveness of feature learning and the accuracy of defect positioning can be effectively improved, which has very important practical significance for reducing software development and maintenance cost.

The application relates to the technical field of processors, in particular to a tile block instruction set architecture and a processing method, wherein the tile block instruction set architecture comprises a tile block parallel processing system control instruction, a tile block descriptor register configuration instruction, a tile block data carrying instruction, a tile block data copying instruction and a tile block arithmetic instruction; the tile block instruction set architecture processing method is to combine one or more tile block operation units, one or more tile block storage management units and one or more tile block computing task scheduling and synchronization units into a tile block parallel processing system, so as to allow parallel processing of tile block instructions and complete tile block-based data operation; the architecture aims to solve the problems of excessive instruction overhead, insufficient data abstraction and excessively high coupling degree of software and hardware of an existing processing system when processing a two-dimensional data block.

The application discloses a traffic trajectory anomaly detection method and device based on neighborhood reconstruction and graph contrast learning, and the method comprises the following steps: a traffic trajectory network is constructed by selecting a Porto data set and performing data preprocessing, trajectory data of each vehicle is abstracted as a node in a graph, and the relationship between vehicles is represented by an edge; a neighborhood reconstruction module is used to encode the receiving domain of the node by using a graph neural network, the neighborhood structure of the node is reconstructed in a dimension reduction space, and the complex relationship and attribute information between the vehicles in the traffic trajectory are captured; a graph contrast learning is used to maximize the similarity of positive samples and minimize the similarity of negative samples, so that the learning ability of the model for traffic trajectory abnormal behaviors is enhanced, and the detection accuracy of abnormal trajectories is improved; an abnormal score function is defined to calculate the abnormal score of each trajectory, and a threshold of the abnormal score is set based on historical data and error analysis; finally, the abnormal score is ranked, and the trajectories with high scores are marked as possible abnormalities, so that the effective detection of traffic trajectory abnormal behaviors is realized.

The application relates to a top-adjusted optical module data processing method, which comprises the following steps: S1, continuously multiple bit top-adjusted signals are represented as a top-adjusted data matrix through data abstraction; S2, the top-adjusted data matrix is subjected to a transpose operation to obtain a top-adjusted data transpose matrix; S3, a feature matrix is set, the feature matrix is subjected to a multiplication operation with the top-adjusted data transpose matrix to obtain a target matrix; S4, elements in the target matrix are subjected to a summation operation, and whether the top-adjusted data is a characteristic similar code of a preamble of a current data frame is judged according to a summation operation result; if yes, data recovery and synchronization processing are carried out; otherwise, the current data frame is discarded. The top-adjusted optical module data processing method realizes the output and identification of the characteristic similar code of the preamble through a matrix operation mode, improves the anti-interference ability, data processing ability and transmission reliability of top-adjusted optical module communication, and reduces the packet loss rate of top-adjusted optical module communication.

A computing device and a method to generate a decentralized delivery scheme for multivehicle in an interested area may comprise the following steps. First, the computing device may receive map data, travel request data, and service vehicle data. Second, the computing device may determine one or more constraints based on the travel request data. Third, the computing device may abstract the travel request data, the service vehicle data, and the map data into a request-vehicle graph comprising nodes and edges. Fourth, the computing device may trim the request-vehicle graph into partial request-vehicle graphs for each service vehicle. Fifth, the computing device may encode the partial request-vehicle graphs through a graph neural network (GNN). Sixth, the computing device may train the GNN to predict actions for each service vehicle. Finally, the computing device may instruct the service vehicles to operate based on the actions.

This invention belongs to the field of software development technology, specifically relating to a method and system for designing analysis step-related data structures applied to structural finite element software development. Step 1: Considering the business background of structural finite element calculation and analysis, the analysis step-related data is abstracted at the data level, abstracting the finite element model into data objects and analysis steps. Step 2: Based on the data objects from Step 1, data object management requires data abstraction and data structure design to form a data object model. This data object model should include a unique identifier for the analysis step, an analysis step status identifier, and an analysis step data manager. Step 3: Based on the data object model from Step 2, according to the usage requirements of data objects in structural finite element analysis calculations and the user interaction logic at the UI level of the structural finite element analysis calculation software, the interface encapsulation design of the data object model is carried out. This invention addresses the problems of status and data management of analysis step-related data in the existing field of structural finite element model data structure design.

The invention discloses a distributed soft bus communication method and system based on multi-hop path matrix optimization, and belongs to the technical field of communication, and the method comprises the following steps: obtaining multi-dimensional link state information through a lightweight detection message, and forming structured link measurement data; abstracting link measurement data into a cost vector, and mapping the cost vector into a matrix element in a weighted combination manner, so as to obtain a computable multi-dimensional cost matrix; cost function weights in the multi-dimensional cost matrix are set according to different service scenes, an optimal path is calculated by using an improved Dijkstra algorithm or an incremental graph algorithm, a stability control mechanism is introduced, and a result is dynamically updated; and an optimal path result is injected into the SoftBus routing layer through an interface and is bound by the session management layer, and inter-application communication is established. According to the method, conversion from a single index to multi-target constraint is realized in a path selection link, and the communication efficiency and reliability of the distributed soft bus are improved on the premise of keeping transparency.

The present application relates to a kind of equipment time series data storage model construction method and system suitable for coal mine multi-disaster fusion analysis, belong to coal mine safety disaster analysis technical field, to solve the problem of data effectiveness, history deficiency in prior art.The method first abstracts equipment time series data into point type data mode and line type data mode, and constructs the data storage conceptual model including monitoring time series data model, alarm time series data model, personnel positioning monitoring data model, monitoring time series slice data model and alarm time series slice data model.Database system using NoSQL technology, design the database table into conceptual model, and create the corresponding wide table structure, take column-based storage mode, set the compression strategy of column, realize long-term online storage and efficient reading of data.The equipment time series data storage model constructed by the present application can effectively reduce the complexity of data access, reduce data load and operation amount.

The embodiment of the invention provides a robot learning data processing method and device based on human behavior data, and the method comprises the steps: converting multi-source heterogeneous human behavior data into a unified intermediate behavior representation which describes a human operation intention through a key point pose; based on the kinematic model and the kinetic model of the target robot body, the intermediate behavior representation is mapped into a robot action sequence, and the robot action sequence meets the joint limit, moment constraint and contact stability requirements of the target robot body. By abstracting human behavior data into intermediate behavior representation irrelevant to a specific acquisition mode, unified processing of multi-source heterogeneous data is realized, and bidirectional mapping from human intention to robot action is realized based on kinematics and dynamics models. Through decoupling data acquisition, representation and mapping processes, the problems of data splitting, domain difference and insufficient generalization ability caused by body difference between human behavior data and a robot body in the prior art are solved.