34 results about "Sequence diagram" patented technology

The invention relates to the technical field of knowledge maps and intelligent decision, and discloses an intelligent decision support method and system based on enterprise knowledge maps, and the intelligent decision support method comprises the steps: collecting original data to obtain a standardized data set; performing knowledge extraction and fusion, and obtaining a domain knowledge graph by adopting an ontology evolution mechanism; symbol rule extraction and vector representation learning are carried out, and a two-way attention fusion mechanism is adopted to obtain an inference result and an explanation link; performing time sequence correlation analysis, and obtaining a decision sequence diagram by adopting a time decay attention mechanism; evaluating the supply chain toughness by adopting a health state prediction method and a network topology vulnerability analysis method to obtain a toughness enhancement scheme; performing multi-objective optimization sorting to obtain a recommended decision scheme; and performing knowledge verification by executing feedback, and obtaining an optimized knowledge graph by adopting an incremental updating method. According to the method, symbol reasoning and vector reasoning can be fused, and accurate and interpretable intelligent decision support is provided.

The invention discloses a power distribution network equipment fault root cause analysis data modeling method, relates to the technical field of power distribution network fault analysis, and aims to solve the technical problems of low fault root cause analysis accuracy and slow response under high-frequency topology dramatic change and edge terminal computing power fluctuation of an existing power distribution network. Comprising the following steps: S1, acquiring equipment operation data and real-time topology information by a heterogeneous edge terminal of a power distribution network, constructing a dynamic lightweight time sequence diagram model with online self-adaptive capability according to an edge terminal computing power high-frequency dynamic topology dramatic change scene, and outputting a fault root cause coarse positioning result; s2, the heterogeneous edge terminal uploads topology increment information and a coarse positioning result in a grading manner based on topology change intensity, wherein the topology increment information is structured data of local topology change of the power distribution network; and S3, the cloud platform updates the global time sequence diagram according to the topology change intensity in a grading manner. The method has the advantages of improving the fault root cause analysis accuracy and the response speed.

The invention relates to a rolling bearing fault diagnosis method based on multi-sensor fusion and a GAF-CNN. The method comprises the steps of collecting a plurality of one-dimensional time sequence signals of a rolling bearing through a plurality of sensors; converting each one-dimensional time sequence signal into a corresponding two-dimensional time sequence diagram by adopting a Grubrum angle field image coding algorithm; inputting the plurality of generated two-dimensional sequence diagrams into a pre-established neural network model, and extracting a feature vector from each two-dimensional sequence diagram in parallel by the model so as to output a plurality of feature vectors; performing feature fusion on the plurality of output feature vectors to generate a fusion feature; and finally carrying out fault diagnosis on the rolling bearing based on the fusion feature. The invention aims to solve the technical problems of low diagnosis reliability caused by incomplete information of a single sensor and insufficient diagnosis precision caused by insufficient feature extraction when a one-dimensional time sequence signal is directly input into a neural network in the prior art.

This application discloses an interface dependency parsing and automated testing method, device, and storage medium, comprising: extracting the minimum path set of the target medical business scenario from the interface call sequence diagram based on the target medical business scenario; using the minimum path set as the test path, parsing the data dependency relationship of each interface in the test path, extracting data flow rules to generate test data; calling the interfaces of the test path according to the test data to generate integrity verification results; comparing the integrity verification results with the expected behavior of the interface call sequence diagram to locate abnormal test nodes; and identifying changes in the interface dependency relationship of abnormal test nodes. This application automatically constructs and dynamically updates the interface call sequence diagram from real call logs through a time sequence pattern mining algorithm, replacing manual dependency analysis. When business logic changes, the test path and data flow rules are re-extracted to ensure that the test plan is updated synchronously with the business, thereby reducing the maintenance cost of test scripts.

The invention discloses a transformer substation E-key lock management method and system. The method comprises the steps that an operation order is converted into a structured sequence diagram through operation sequence modeling; dynamic permission activation is realized based on a state machine model; performing five-prevention logic real-time verification in linkage with a substation monitoring system; digital signature and bidirectional authentication are adopted to ensure instruction security; local decision execution is realized through intelligent verification of the lockset; a sequence state is automatically advanced based on an operation progress. The system comprises a management background, an intelligent mobile terminal and an intelligent electronic lock, and a complete closed-loop management system is formed. According to the invention, deep binding of lock operation and substation operation processes is realized, the technical problems of extensive management and disjunction with safety regulations of a traditional lock are solved, and the operation safety, the operation and maintenance efficiency and the system reliability are remarkably improved.

The invention discloses a sequence diagram simulation execution visualization method and system based on UML (Unified Modeling Language) specification driving, and relates to the technical field of UML / SysML (Unified Modeling Language / SysML) modeling and simulation verifier.The method comprises the following steps: firstly, converting a SysML sequence diagram model into an Interaction semantic structure conforming to a UML standard; then, a simulation engine performs semantic verification on the structure and compiles the structure into an executable simulation step network with a clear dependency relationship; then, the engine performs event-driven dynamic execution according to the network in the step, and issues an execution state event through an event bus; and finally, subscribing the events by a user interface, and displaying the execution processes of message transmission, lifeline activation and complex control flow in real time in an animation form. According to the method, high conformity and execution efficiency of simulation to the UML standard are ensured through strict semantic conversion and compilation type execution architecture.

Embodiments of the present application disclose a test case construction method and related device, the method comprising: obtaining a sequence diagram; generating an interaction directed graph corresponding to the sequence diagram; a node in the interaction directed graph corresponding to message or fragment condition information in the sequence diagram, an edge in the interaction directed graph corresponding to an order relationship between messages in the sequence diagram, or an order relationship between fragments, or an order relationship between messages and fragments; for each edge in the interaction directed graph, determining a traversal path of the edge in the interaction directed graph; the traversal path comprising a first sub-path from a starting point of the interaction directed graph to the edge, and a second sub-path from the edge to an ending point of the interaction directed graph; for each determined traversal path, determining a case key element according to respective contents of respective nodes on the traversal path; and constructing a test case corresponding to the to-be-tested software based on the case key element. The method can automatically construct a test case, improve the construction efficiency of the test case, and is less likely to miss a test scenario.

The present application relates to the technical field of vehicle video data analysis, and discloses a vehicle video data analysis method based on traffic responsibility definition. The method first builds a video library containing various traffic scene videos, and each video has an event sequence diagram. Then, the same scene event sequence diagrams are merged into a comprehensive sequence diagram. Next, the responsibility requirements and analysis requirements of the accident are obtained, and the comprehensive sequence diagram is selected and the basic sequence diagram is extracted according to the requirements, and the part to be identified is labeled. Then, a responsibility judgment model is constructed, and after the complete sequence diagram is obtained by analyzing the part to be identified, the key frame nodes are checked, and the nodes with a dispute probability greater than a critical threshold are defined as dispute nodes and a label reminder is generated. In addition, when the video library is built, the video is collected and preprocessed, and the key frame nodes are extracted by the optical flow method or the convolutional neural network model. The method can improve the efficiency and accuracy of traffic responsibility definition.

The invention relates to the field of fault diagnosis, in particular to a port ship loader dynamic monitoring and multi-source information fusion fault diagnosis method, which comprises the following steps of embedding physical constraints into input high-frequency original data through a deep diagnosis module, generating theoretical signals, and determining a fault problem in a mode of constructing a loss function. In consideration of possible causal connection from the past to the present, the causal reasoning module constructs a full-connection time sequence diagram, deletes false association through conditional independence test, obtains a directed causal diagram, and traces a fault propagation path. A fault component identified by the deep diagnosis module is mapped to a fault propagation path, and a finally determined fault problem is judged based on a consistency verification method. According to the method, physical constraints are embedded in the diagnosis process, excessive dependence of a traditional data-driven fault diagnosis method on prior data is overcome in combination with causal reasoning, and fault diagnosis participates in combination with a physical mechanism.

The application discloses an AI-driven construction drawing compliance review system and method, relates to the technical field of construction drawing review, and initializes a construction element database based on input construction drawings and a BIM model, automatically generates a construction sequence diagram by using an AI engine, and synchronously generates simulation scene parameters; a graph construction module automatically loads matched construction process specifications and process standards to construct a specification graph, converts the specification graph into a standardized graph format, and automatically calls the specification graph by using a simulation engine to compare a component behavior and a space state in real time; if there is a construction anomaly in the simulation process, a corresponding process node is automatically marked to form a violation event set. The review system realizes the paradigm change of construction drawing review from "static rule matching" to "dynamic process evolution", and significantly enhances the ability to understand complex construction logic and space behavior and the real-time response capability.

The invention discloses a traffic flow prediction method based on a complex quantum time sequence chart neural network, and belongs to the technical field of artificial intelligence. The method comprises the following steps: firstly, constructing a time sequence traffic map and extracting an adjacent matrix, a Laplacian matrix and a time sequence characteristic matrix; then, constructing a quantum propagation evolution matrix based on a Laplacian matrix, and mapping real number features to a complex number field by using a full-connection network; performing feature propagation through a quantum evolution operator, and extracting spatial features by adopting a complex graph convolution operation; a deep network is constructed by stacking a plurality of complex quantum graph convolutional layers, and finally, a prediction result is generated through a full connection layer after an absolute value of an output complex feature is obtained. According to the method, the spatial-temporal evolution rule of the traffic network is effectively and uniformly modeled through complex field calculation and a quantum evolution mechanism, the prediction precision and the calculation efficiency are remarkably improved, and reliable technical support is provided for an intelligent traffic system.

Embodiments of the present application disclose a sequence diagram-based test scenario generation method, device, equipment, medium and product. The method comprises: acquiring a design sequence diagram and a preset sequence diagram standard. The design sequence diagram comprises interactive elements for describing message passing and control flow between objects. The sequence diagram standard is used for describing rules and requirements for drawing a sequence diagram. Element relationship extraction is performed on the design sequence diagram according to the sequence diagram standard, to obtain mutual relationships between interactive elements in the design sequence diagram. An interactive directed graph for describing interactive element flow is constructed according to the mutual relationships. Each operation path described by the design sequence diagram is extracted from the interactive directed graph, and a target test scenario corresponding to the design sequence diagram is generated according to each operation path. The technical solution of the embodiments of the present application realizes standardization and automation of test scenario generation, guarantees coverage and accuracy of the test scenario, and reduces dependence on experience of test personnel.

The application discloses a two-dimensional code full life cycle tracking management system based on dynamic encryption, relates to the technical field of two-dimensional code tracking management, and achieves dynamic generation and security binding of two-dimensional code content by constructing a target object life cycle state sequence diagram, dynamically generating a two-dimensional code coding index based on a current state, a device environment and time factors, and combining a multi-factor asymmetric encryption mode; meanwhile, legality verification of two-dimensional code access behavior is carried out through path consistency verification and state chain continuity judgment, and two-dimensional code effectiveness is judged in combination with an environment deviation detection mechanism; compared with the prior art, the application can effectively prevent two-dimensional code forgery, replay and unauthorized access, improve the uniqueness, verifiability and security of two-dimensional codes in the full life cycle, and is suitable for high-security product tracking, anti-fake authentication and multi-stage identity recognition scenes.

The invention provides an emergency operation program risk assessment method considering time limitation. The method comprises the following steps: step 1, constructing an event sequence diagram-fault tree ESD-FT model; 2, constructing three general sub-models, and respectively describing dynamic changes of available time EoAT, response time EoRT and human error probability EoHE under time limitation; step 3, constructing an event sequence diagram-fault tree-dynamic Bayesian network ESD-FT-DBN comprehensive model; and 4, carrying out Monte Carlo simulation. The method overcomes the defects of an existing static risk assessment method in the aspect of processing time dynamic factors, the time dependence relation of all steps in the emergency operation process is brought into the risk analysis category, the assessment process can reflect the change of available time, response time and human error probability along with the accident process, and the risk assessment accuracy is improved by combining a Monte Carlo simulation method. And comprehensive evaluation of the probability of success, failure or timeout of the emergency operation under the emergency condition is realized.

The present application relates to the field of industrial data analysis, and more particularly to an industrial big data graph computing algorithm, the present application constructs a runtime sequence diagram for industrial equipment by obtaining equipment application data corresponding to a plurality of tasks in a predetermined period of an industrial plant, identifies a heavy operation time domain segment to extract a running state feature corresponding to the heavy operation time domain segment, evaluates a time sequence compactness representation value of the industrial equipment in combination with the running state feature and an average physical distance between the industrial equipment to divide the running state of the industrial plant, collects equipment working features of the heavy operation time domain segment according to the running state, and adaptively evaluates and analyzes the operation scheduling of the industrial plant. The present application improves the accuracy of the running state evaluation of the industrial equipment. At the same time, the present application can adapt to complex scene requirements, dynamically adapt the operation scheduling, realize dynamic balance of the operation efficiency and resource efficiency under different load scenes, and improve the continuity and economy of industrial production.

This invention provides an automatic UML sequence diagram generation method compatible with multilingual code analysis. By capturing the characteristics of C++ and Java code, it identifies the code types of each part of the mixed source code and analyzes different code types. For code marked as C++, regular expressions are used to match keywords, class information, and inheritance information of the C++ source code to be analyzed, and these are converted into parameter names. For code marked as Java, syntax and lexical analysis are used to extract class information and method call information of the Java source code to be analyzed. Corresponding Plant UML code is generated and drawn as a comprehensive UML sequence diagram. This invention's automatic UML sequence diagram generation method compatible with multilingual code analysis can intuitively reflect the functional and logical composition of mixed C++ and Java code, and intuitively reflect the interaction relationship between them. It avoids the need to view tedious source code to obtain program information and functional relationships, improving code readability and maintainability.

The invention discloses a time sequence diagram neural network security risk analysis method, system and device and a medium, mainly relates to the technical field of risk analysis, and is used for solving the problems that the existing scheme does not utilize the content of data, and the existing method has a multi-focus static structure, lacks adaptive capacity to dynamic evolution and is difficult to deal with real-time threats. Comprising the following steps: obtaining a trained time sequence diagram neural network by utilizing a diagram structure sequence, a constraint label set of each diagram structure in the diagram structure sequence and state data; acquiring multi-source data and a corresponding constraint label set in real time; calculating the risk level of the multi-source data according to the constraint label content in the constraint label set; and when the risk level is greater than a preset level, obtaining a graph structure sequence corresponding to the multi-source data and a set of all constraint labels corresponding to the graph structure sequence, and inputting the graph structure sequence and the set into the trained time sequence graph neural network to obtain prediction state data.

The application discloses an ArkTS distributed deadlock intelligent detection and repair method based on deep learning, constructs a multi-dimensional lock entity by establishing a mapping of distributed API and lock operation and an implicit synchronization mechanism of component life cycle and a decorator, generates a component state transmission directed graph, a distributed resource access time sequence diagram and a task scheduling dependency graph, and combines them into a global lock dependency graph atlas, detects a static deadlock path in combination with linear time sequence logic safety attribute, implants a hook function to obtain a lock state in an application runtime, detects dynamic deadlocks such as single-device, cross-device and state variable through a resource allocation graph, deploys a deep learning deadlock prediction model to realize runtime early warning, traces back a rule violation operation in reverse, analyzes and locates a root cause in combination with a deadlock type, and adopts strategies such as standardizing a lock application sequence, dynamically adjusting a task priority and the like to complete repair for distributed deadlocks, task scheduling deadlocks and state variable deadlocks, and significantly improves the reliability and development efficiency of a Hongmeng application.

The invention discloses a GPS track travel mode identification method based on semi-supervised double-graph learning, and the method comprises the steps: firstly obtaining GPS track data, segmenting the GPS track data according to a time sequence, and extracting point location features to obtain a plurality of track fragments; then, constructing a double-graph topological structure for each fragment, wherein the double-graph topological structure comprises a sequence graph for representing local time continuity and a learnable dependency graph for adaptively capturing non-local semantic correlation; carrying out self-supervised mask graph self-encoder pre-training by utilizing the unlabeled fragments, and learning space-time representation depending on perception; and finally, performing supervision and fine tuning on the pre-training model by using the labeled fragments, and outputting a travel mode prediction result. According to the method, the structured semantic information of the short-track fragments can be fully mined without a long-sequence track and massive labeled data, the defect that the recognition performance is seriously reduced in short-sequence and label sparse scenes in the prior art is effectively overcome, and the robustness and generalization ability of the model are remarkably improved.

The invention provides a vehicle unit co-simulation method and device, simulation equipment, a medium and a product. According to the method, an equipment execution sequence diagram and an FMU model corresponding to each target sub-simulation equipment are determined through main simulation equipment according to model connection relation data in a simulation process file, then a model construction request is sent to each target sub-simulation equipment, and the sub-simulation equipment constructs an FMU instance according to model information in the model construction request. And the main simulation device sends a simulation execution request to the target sub-simulation device according to whether the target sub-simulation device is a starting node in the device execution sequence diagram. And the sub-simulation equipment sequentially runs each FMU instance according to the simulation execution request and the model connection relation data in the model construction request, obtains simulation output data and sends the simulation output data to the main simulation equipment, and co-simulation of the vehicle units is realized. According to the scheme, the simulation process is controlled through the main simulation device, and the FMU instance is constructed by the sub simulation device, so that the simulation efficiency is improved.

The invention discloses an interface list generation method and system based on combination of a time sequence diagram and a large model, and the method comprises the steps: generating structured log information based on a target project file, and enabling the log information to be associated with interface information; the log information is filtered; generating a time sequence diagram including interface interaction information according to the filtered log information; performing semantic analysis on the filtered log information based on a large language model to generate functional description of each interface; and generating an interface list according to the time sequence diagram and the function description of each interface. Based on the data processing flow, the interface function description and the time sequence diagram are automatically generated, manual intervention and processing time are reduced, and working efficiency is greatly improved.

This application discloses a method and system for tracking cigarette product production information, relating to the field of tobacco technology. The method implements unique identification for each level of cigarette product through predefined coding rules; industrial IoT devices continuously collect the timestamps, status identifiers, and location coordinates corresponding to the codes to form a raw data stream; a stream processing engine based on Apache Flink performs real-time grouping, time-series alignment, and state merging on the raw data stream, outputting a structured event sequence; a graph database transforms the event sequence into a multi-dimensional relationship graph of coded nodes and process nodes, constructing a semantically complete full-process topology structure; a dynamic programming algorithm traverses the graph, extracting the complete flow path of any code according to the time sequence to generate a production trajectory chain; finally, a force-directed graph layout algorithm spatially rearranges the nodes and edges in the trajectory chain, achieving a natural balance in node distribution through physical simulation, and outputting an interactive visual graphic.

A method and a device for predicting an event to occur in a device are provided. The method includes identifying a plurality of events that occurred in the device; generating a plurality of sequence graphs based on a time-series order of the plurality of events; generating a sequence group graph by combining the plurality of sequence graphs based on relationships between nodes and edges in the plurality of sequence graphs; obtaining a computation graph by sampling the generated sequence group graph based on time-series order information of the plurality of events; and predicting the event to occur in the device using a graph neural network (GNN) based on the obtained computation graph.

The application discloses a GPS trajectory travel mode recognition method based on semi-supervised double-map learning, first acquires GPS trajectory data and segments according to time sequence, extracts point features to obtain a plurality of trajectory segments; then constructs a double-map topological structure for each segment, including a sequence map representing local time continuity, and a learnable dependency graph that adaptively captures non-local semantic correlation; then, a self-supervised mask graph autoencoder pre-training is performed on the unlabeled segments to learn a dependency-aware spatiotemporal representation; finally, the pre-trained model is supervised fine-tuned with labeled segments, and the travel mode prediction result is output; the application can fully mine the structured semantic information of short trajectory segments without long sequence trajectories and massive labeled data, effectively overcoming the defects of the prior art that the recognition performance seriously decreases in the short sequence and sparse label scene, and significantly improving the robustness and generalization ability of the model.

The application provides a natural gas pipeline third-party damage accident early warning method based on a dynamic Bayesian network. The method comprises the following steps: first, third-party damage accident factor identification; second, risk accident scene building; third, accident consequence analysis, an ESD is used to build a sequence diagram about the natural gas pipeline failure consequences caused by the third-party damage; fourth, node probability calculation; fifth, DBN dynamic quantitative modeling coupled with the ESD and the FT; sixth, third-party accident probability dynamic analysis based on the DBN model; and seventh, risk control measures are summarized according to the failure probability analysis results. The method can obtain the pipeline failure risk probability by calling historical data, calculating the occurrence probability of each factor by using a fuzzy comprehensive evaluation method, and reasoning the input data information of the Bayesian network as the data information, and the probability information is combined with the accident consequence analysis method to perform early warning.

The application discloses a formal model of UML sequence diagram, a unified structure and a modeling tool thereof, and a method for keeping the robustness of the UML sequence diagram under the unified structure, realizes modeling and analyzing the sequence diagram from the structure aspect, can conveniently, quickly and effectively analyze, and can guarantee the robustness of the sequence diagram in software development, thereby effectively helping object-oriented developers to overcome the gap between the analysis model and the design model. In the unified structure modeling tool provided by the application, once a sequence diagram is constructed, it can be automatically converted into the unified structure and robustness analysis is performed.

The application provides a smart city multi-source data fusion intelligent analysis method and system, relates to the technical field of data processing, and comprises the following steps: performing time sequence unified coding on multi-source access data, and obtaining a multi-source time sequence access data set; constructing a multi-source data interaction time sequence diagram; identifying a data analysis task request, performing semantic structural analysis, obtaining a task description vector, and including a description target object, a task type, a task space range and a task time sequence window; obtaining a task related subgraph generated by a seed node set and a neighborhood node set; extracting a multi-source interaction data set and a multi-source non-interaction data set, performing data fusion on the multi-source interaction data set to obtain interaction fusion data; and feeding back to a request terminal. The application solves the technical problem that the prior art usually adopts a full data processing mode, which leads to a sharp increase in calculation load and limits the efficiency of smart city data analysis and decision support.

The invention provides a data modeling and optimizing system of a machine learning large model, which relates to the technical field of data processing and is used for constructing behavior sequences of different behavior types, identifying behavior fragments with time continuity, synthesizing the behavior fragments by setting a sliding window, obtaining a fragment combination set and analyzing a relative sequence of behavior occurrence time. Identifying a behavior time sequence evolution path, determining a trajectory overlapping ratio of different behaviors, calculating a behavior penetration degree between each pair of behaviors, extracting behaviors with the behavior penetration degree higher than a preset penetration threshold value in a behavior trajectory matrix, analyzing a connection frequency and a transfer position of each behavior, calculating an information concentration value of each behavior, and determining a behavior time sequence evolution path; screening each behavior in the structure sequence diagram to obtain a training reserved data set; according to the method, dynamic modeling with semantic consistency can be carried out on training data of a large model.

The invention discloses a network security artificial intelligence defense method and system and a storage medium, and relates to the field of network security, and the method comprises the steps: extracting event features from a multi-source service log and a security log, generating a first feature vector, aggregating the first feature vector into node features and edge features, and constructing a time sequence diagram; inputting the time sequence diagram into a dynamic graph neural network, predicting node features and obtaining abnormal indexes; calculating a node risk score and identifying a suspected attack chain and severity in combination with an abnormal index and an association relationship in the graph; carrying out local training on a plurality of mail area nodes based on local data and adversarial samples, obtaining a global model through federal weighted aggregation, and issuing the global model to each node; and constructing a system state vector according to the risk score, the attack chain severity, the service type, the asset value and the historical response, and outputting safety response actions such as speed limiting, blocking, isolating, freezing, current limiting or manual rechecking through a strategy network to realize intelligent cooperative defense of complex attacks.

The invention discloses a public area abnormal behavior processing method and device, equipment and a storage medium, and relates to the technical field of big data. The method comprises the steps of performing key point coordinate extraction on a user behavior video to obtain key point coordinate time sequence data of at least one current user, and constructing a space time sequence diagram for representing a current user behavior rule on the basis of the key point coordinate time sequence data; determining an abnormal behavior score of the current user according to the space sequence diagram based on a pre-trained space-time diagram convolutional network model; matching the key point coordinate time sequence data with a preset abnormal behavior rule to obtain an abnormal behavior detection result of the current user; and determining an abnormal behavior processing result of the public area according to the abnormal behavior score and the abnormal behavior detection result. According to the scheme, the original video stream is converted into the structured human skeleton data, and the spatial-temporal modeling analysis is performed on the structured human skeleton data by using the spatial-temporal diagram convolutional network model, so that the efficiency and accuracy of processing the abnormal behaviors in the public area are improved.