41results about How to "Ensure accelerated performance" patented technology

The invention relates to the technical field of model construction, in particular to an obstetric delivery decision model construction method based on a knowledge graph. The method comprises the following steps that multi-dimensional information collection is conducted on a pregnant woman to be delivered, individualized data are obtained, and the multi-dimensional information comprises a medical record text, a laboratory examination result, an ultrasonic image and fetal heart rate monitoring data; performing standardization and structural processing on the individualized data to generate a standardized multi-modal data set; performing entity identification and relation extraction processing on the standardized multi-modal data set according to a preset obstetrical domain ontology model to obtain an initial knowledge graph; and carrying out priori knowledge-based mapping expansion on the initial mapping knowledge domain to obtain a multi-modal mapping knowledge domain. Through an attention mechanism and a decision support interface, the interpretability of the model is enhanced, and doctors can understand and trust decision suggestions.

The invention relates to a power grid business intelligent auditing method based on AI enhancement, and the method comprises the following steps: S1, collecting data from a power grid SCADA system and a financial system, integrating historical and real-time data, and constructing an auditing knowledge graph; s2, constructing an instruction recognition model by using RoBERTa to analyze a natural language instruction of an auditor, and combining knowledge graph information with a language model to improve the understanding and reasoning ability of an instruction background; s3, constructing a multi-task deep learning framework to realize processing and fusion of multiple auditing tasks; s4, on the basis of a natural language instruction input by an auditor, extracting query targets, conditions and parameters involved in the instruction according to the instruction recognition model, obtaining corresponding auditing data in the database, and obtaining intention classification; and S5, calling a corresponding auditing model based on the intention classification result and the auditing data, and obtaining an analysis result. According to the invention, efficiency and precision in power grid business auditing work are effectively improved.

The invention discloses a wireless charging intelligent regulation and control method and system, and relates to the technical field of wireless charging regulation and control, and the method specifically comprises the following steps: obtaining the charging demand information of each robot in real time, and generating a task priority coefficient and an electric quantity demand index of each robot; and constructing a wireless charging power distribution evaluation model for the generated task priority coefficient and the electric quantity demand index of each robot, generating a priority distribution coefficient of each robot, and evaluating the wireless charging demand level of each robot. Dividing each robot into a high-demand robot, a normal-demand robot and a low-demand robot according to an evaluation result; and according to the division result of each robot, corresponding power distribution strategies are adopted for the robots with different wireless charging demand levels. According to the method, the problems of wireless charging demand difference identification and dynamic allocation of the robot in the intelligent warehousing system are solved, and an efficient and accurate charging resource optimization configuration effect is achieved.

The invention provides a work plan and task management system, which belongs to the technical field of work plan tasks and comprises a portal layer, an application program interface gateway layer, a business service layer, a public service layer and a data layer. And the application program interface gateway layer realizes intelligent management of the interface through a self-adaptive division module, wherein the intelligent management comprises dependency analysis, granularity calculation, aggregation optimization and recombination execution. The business service layer focuses on work plans and task management, and realizes collaboration through flexible interface interaction. The public service layer provides basic support for user authentication, log recording, message queue, service registration, configuration management and the like. And the data layer adopts a diversified storage strategy, integrates a relational database, a cache database, a time sequence database and distributed file storage, and meets data processing requirements of different scenes. The problem that in the prior art, interface definition and calling relations between micro-services are difficult to manage in a self-adaptive mode is solved.

The invention provides a fault simulation test method and device, electronic equipment and a storage medium. In the fault simulation test method, when a test instruction sent by a user is received, test parameters are configured based on the test instruction; wherein the test parameters comprise a test type, a fault parameter and a test target. And based on the test type and the fault parameter, configuring a network rule through a flow controller, and generating a fault simulation scene. And injecting the fault simulation scene into the test target, and carrying out performance test on the test target. And monitoring performance indexes of the test target in the test process, and generating a test report of the test target. Therefore, by utilizing the method provided by the invention, different types of network delays or faults, such as delay increase, packet loss and connection interruption, can be accurately injected into the target server, so that the stability and the performance of software or service under bad network conditions are effectively evaluated, and the flexibility and the expandability of the system are ensured at the same time.

The invention discloses an internet of things-oriented federated learning anti-poisoning attack method. The method comprises the following steps of S1, initializing an internet of things federated learning system; s2, performing mask processing on the privacy data of the Internet of Things participants; s3, the cluster server collects mask data provided by all the Internet of Things participants and aggregates the mask data; and S4, identifying malicious poisoning participants of the Internet of Things through robust clustering. According to the method, an advanced machine learning algorithm and a privacy protection technology are introduced, a more effective anti-poisoning protection measure is provided for the Internet of Things system, the security and robustness of the Internet of Things system are remarkably improved through successful application of the method, and an important guarantee is provided for coping with potential hostile attacks and data tampering.

The invention relates to interface generation, in particular to an interface generation system based on a large language model (LLM), which comprises a data acquisition layer for performing data extraction and parameterized packaging on interface description data received by a user layer and converting the interface description data into a structured data set; the parameter processing layer is used for classifying the structured data set output by the data acquisition layer according to an input purpose, inserting a corresponding prompt template, automatically generating context variables required by each interface module in an interface generation process by AI, and carrying out association matching on each interface module and parameters thereof; the model processing layer summarizes the operation of each interface module, converts the operation into operation description, inserts the operation description into a corresponding prompt template, and selects the most suitable operation from the operation description according to the description of each interface module; according to the technical scheme, the defect that in the prior art, it is difficult to generate interface codes meeting service requirements on the basis of limited interface description can be effectively overcome.

The invention discloses a decentralized identity authentication method using a DID technology, and the method comprises the following steps: S1, generating a public key and a private key by a user, and generating an identity identifier by using the public key; s2, the user combines the identity data with the identifier, encrypts the identity data and the identifier, and stores the encrypted identity data and identifier on the block chain; s3, the third-party authentication mechanism issues a verifiable certificate based on the identifier; s4, the user manages the verifiable certificate, signs the identity verification request by using a private key, and sends the signed identity verification request to the verifier; s5, the verifier verifies the validity of the signature through the public key on the block chain; s6, recording verification time, a signature and a result, and storing the verification time, the signature and the result on the block chain; s7, the user can recover the lost private key or identity identifier through multiple signatures or other mechanisms; and S8, providing a user-friendly interface, supporting cross-platform migration of identity data, following open standards and protocols, and establishing a transparent governance mechanism. According to the method, the block chain and the advanced encryption technology are combined, and safe storage and privacy protection of identity data are realized.

The invention discloses a multi-agent architecture-based previous insurance symptom key information structured extraction system, which comprises a previous insurance symptom key information extraction system and a multi-agent architecture-based previous insurance symptom key information extraction system, the previous insurance symptom key information extraction system comprises a data preprocessing module, a task allocation and management module, a disease diagnosis and recognition module, a treatment record extraction module, an information integration and structuring module and a feedback optimization module. Each module is independently responsible for a specific function, the flexibility and expandability of the system are ensured, natural language processing and machine learning are applied, semantic information of a medical text can be deeply understood, key information can be accurately recognized and extracted, information sharing and cooperative work are achieved through an efficient communication mechanism, and the system is high in practicability. Complex processing tasks are coped together, and the cooperative mode not only improves the processing efficiency, but also enhances the overall performance and robustness of the system.

The invention provides a visual network AI calculation method and system based on SIP signaling and a video gateway. The method comprises the following steps: acquiring a real-time video stream address of camera equipment; accessing and processing a video stream by using a video gateway; carrying out algorithm processing on the frame image by using an AI analyzer; according to an analysis result of the AI analyzer, under the condition that the event meets a preset condition, the event data are stored in a receiving center, and the event is pushed to a specified subordinate system by utilizing a pushing center; under the condition that the number of camera devices or the concurrency of video streams is increased, dynamically expanding service nodes of a video gateway and an AI analyzer; and under the condition of equipment failure or network abnormity, a failover mechanism is automatically triggered. Dynamic expansion of the video gateway and the AI resolver is realized through a containerization technology, and high availability and stability of the system can be ensured to be kept even under the condition of high concurrency in combination with a load balancing strategy.

The invention relates to a neural network model construction method and device, computer equipment and a storage medium. The method comprises the following steps: acquiring a selection operation on a neural network parameter value to be selected in a neural network parameter option component on a configuration interface, and storing the selected neural network parameter value; obtaining structure configuration parameters on the configuration interface, and performing network layer structure configuration according to the structure configuration parameters; configuring the network layer according to the network layer configuration data and the stored neural network parameter value to obtain a creation logic of the network layer; and creating a network construction class, and constructing a neural network model through the network construction class according to the stored neural network parameter value, the configured network layer structure and the creation logic of the network layer. By adopting the method, the construction process of the neural network model can be simplified, and the model construction efficiency is improved.

The invention belongs to the technical field of automobile data storage, and provides a data storage method and system based on field-level association storage, and the method comprises the steps: building a storage structure for data to be stored based on fields in a distributed database, and associating the fields; pre-processing the associated storage data, and performing timing data compression and duplicate removal on the pre-processed storage data; and optimizing a query path for the preprocessed storage data based on index establishment and a caching technology. According to the method, the data structure is remarkably simplified by adopting a field-level associated storage mode. According to the method, a lengthy and complex table structure in a traditional data model is disassembled, fields are used as basic units for storage, and an efficient association index is established according to the logic relation between the fields. The data storage efficiency is greatly improved, the data writing and reading speed is remarkably increased, and a solid foundation is laid for subsequent data processing and analysis.

The invention discloses a software research and development method based on a PBC assembled architecture, and the method comprises the steps: providing an overall view of a business field oriented to a manager or an architect, and meanwhile, generating linkage with a specific business field to realize a basic drill-down capability; a default engineering scaffold or a user-defined engineering scaffold is supported to generate a code framework; the precipitated asset capabilities of technical components, business components and the like are reused; executing service component custom configuration and generating a service related code path and basic function logic; according to the current situation and the service characteristics, the application selects to add code logic under the condition of cutting or not adjusting stock code logic, so that the scenarized ability is improved; realizing assembly and arrangement of technical components and business components, and performing scene interface development through a visual designer; and breaking through with a DevOps tool chain. According to the technical scheme, an enterprise can accelerate asset precipitation, the delivery efficiency is improved, and the capacity of quickly responding to market requirements is achieved.

The invention discloses a business process optimization execution management system, which comprises a process design module used for designing a business process through a graphical interface; the process execution engine is used for automatically executing a business process according to a pre-designed process model; the task management and distribution module is used for managing and distributing tasks in the process to related participants; the flow monitoring module is used for monitoring the execution condition of the flow in real time, providing flow state, progress and processing time information and supporting setting of monitoring indexes and an early warning mechanism; and the process optimization module is used for improving and optimizing the existing process according to the result of the process analysis, including adjusting the process structure, optimizing the process rule and improving the automation degree of the process. The system supports graphical process design, has a powerful process execution engine and a real-time monitoring capability, can perform deep analysis and optimization according to process execution data, and helps enterprises to realize continuous improvement and efficient operation of business processes.

The invention relates to the technical field of network communication management, and discloses a system and a method for establishing a VPN (Virtual Private Network) data flow acceleration model, and the method comprises the following steps: distributing an IP (Internet Protocol) address for access equipment through a DHCP (Dynamic Host Configuration Protocol), and obtaining the IP address of a message queue telemetry transport protocol (MQTT) server; and the access device registers identity information through the MQTT and stores the identity information. And performing batch management on the access equipment according to the subscription mechanism of the MQTT server, wherein the management content comprises equipment configuration, operation state, terminal access and batch upgrading. Meanwhile, VPN data features and historical data features of the access device are obtained and stored, a VPN data flow classification model is established based on the features, and the VPN features of the device are substituted into the model to obtain a classification result. And finally, obtaining real-time performance data of a network link, determining an effective link according to a demand relationship between the real-time data and a classification result, and selecting an optimal path transmission mechanism based on the effective link so as to optimize the transmission performance of the VPN data stream.

The invention provides a hybrid development method and device, a computer program, a storage medium and computer equipment, which can call each function in a mobile application development framework on a development engine to realize hybrid development based on the development engine and the mobile application development framework. The hybrid development method comprises the following steps that mobile application development codes are packaged into an embedded package, and the embedded package is imported into the development engine; building a bridge file based on the development platform of the mobile application development framework, the bridge file being used for establishing communication between the development engine and the mobile application development framework; and calling the function in the embedded package by the development engine by utilizing the bridge file so as to realize mixed development based on the development engine and the mobile application development framework.

The invention discloses a DPU task scheduling method and a DPU task scheduling system.The method is used for scheduling tasks to a CPU side or a DPU side and comprises the steps that a task package is generated according to task information to be processed; calculating the priority of the task package according to the calculation amount and the deadline of the task package and whether special hardware for the task package exists on the DPU side; according to the priorities of the task packages, the task packages are stored to the positions corresponding to the priorities of the task packages in a scheduling queue, and the task packages are conveyed to the CPU side or the DPU side in sequence; and updating the scheduling queue according to an execution result of the CPU side or the DPU side on the received task packet. Through the scheduling method and the scheduling system, the overall performance of the system can be improved by effectively scheduling the tasks.

The invention belongs to the technical field of image data processing, and discloses a remote sensing image slice library image classification method, a remote sensing image slice library image classification model and an image sampling method. Comprising the following steps: separating an image in an image slice library into a background image and a target image by adopting a target and background separation strategy; extracting background features from the background image, extracting target features from the target image, and extracting overall features from the background image and the target image by adopting a contrast regularization processing method; and classifying the images according to the background features, the target features and the overall features. The problems of low image classification speed and low accuracy of the remote sensing image slice library are solved.

Embodiments of the present disclosure provide a wireless charging test method, device and equipment for a vehicle. The method comprises: determining a test function; generating a test instruction corresponding to the test function; sending the test instruction to the wireless charging panel; monitoring state information fed back by the wireless charging panel after receiving the test instruction; and determining whether the test function of the wireless charging panel passes according to the state information. In this way, a unified operation paradigm and data collection standard are provided for all wireless charging tests, differences in test results caused by inconsistent test steps, conditions or evaluation criteria are eliminated, the accuracy of test results is improved, and since the test instruction is automatically generated and sent by the test host and the state information is automatically monitored and determined, the test efficiency and automation level can be significantly improved.

The present invention proposes a UAV communication system and communication method based on multi-network integration, which relates to the technical field of UAV communication of multi-network integration, obtains multiple UAV self-organizing networks, and establishes a long-distance UAV communication network; obtains a first transmission path and a transmission update path, and performs real-time data transmission, thereby obtaining first path transmission information and second path transmission information; calculates path transmission coefficients respectively, and performs transmission state determination according to the path transmission coefficients; obtains dynamic variables of the UAV self-organizing network, performs dynamic comparison, performs path change processing according to the dynamic comparison result, calculates the path transmission coefficient of the transmission change path, performs transmission change comparison, and thereby determines whether to change again. The present invention is based on a multi-UAV self-organizing network communication method that combines super nodes with satellite communication networks, and has significant advantages in solving the long-distance and high-reliability communication needs of UAVs.

The present application belongs to the technical field of civil engineering monitoring, and discloses a performance verification method and system of a fiber strain sensor for highway infrastructure monitoring, which comprises the following steps: based on a layered material coupling mechanics model, the influence law of the change of coupling layer parameters on the strain transmission efficiency of the measured sensor is analyzed; based on a three-dimensional entity numerical model, the influence law of the layout process parameters of the measured sensor on the strain transmission rate is analyzed; the performance of the sensor is verified in a laboratory environment by using a scale model; and the performance of the sensor on the actual structure is verified through monitoring data on a real highway infrastructure or a full-scale model. The present application realizes multi-dimensional and progressive performance verification of the performance of the fiber strain sensor for highway infrastructure monitoring by constructing a complete performance verification system from a theoretical model to numerical simulation, to a scale model test and an entity structure verification.

The present invention relates to the technical field of backplane assembly, and in particular to an assembly method for a high-density power supply backplane, comprising determining a plug-in fluctuation parameter of a corresponding plug-in according to historical plug-in data and the plug-in update data of a single plug-in, and calculating a plug-in feeding parameter of the corresponding plug-in according to the plug-in usage data of the single plug-in; determining a matching response category of each plug-in for the power slot according to the plug-in fluctuation parameter and the plug-in feeding parameter; determining the number of power supply interfaces of the power slot corresponding to the single plug-in according to the matching response category of the single plug-in, and determining the number of first power slots, the number of second power slots and the number of third power slots respectively in combination with the number of plug-ins corresponding to each matching response category, so as to determine the number of power slots in the slot of the power supply backplane; and determining a corresponding arrangement method according to the number of plug-ins of each matching response category.

The invention discloses a function level defect positioning method based on a call graph and multi-level comparative learning. The method comprises the following steps: constructing a local call graph containing class nodes and internal function nodes and a global call graph covering all functions; performing graph embedding learning on the two calling graphs by utilizing Light-GCN to generate local and global embedding vectors of the function; meanwhile, a pre-training code model is adopted to extract a semantic embedding vector of the function code, and a pre-training language model is used to obtain a semantic embedding vector of the defect report; in order to optimize the performance of the model, a multi-level contrast loss function including local contrast loss, global contrast loss and semantic contrast loss is designed, and the matching relationship between a defect report and function embedding is effectively optimized by minimizing the loss function, so that the overall positioning capability is enhanced, and the positioning accuracy is improved. And the defect positioning accuracy of the model is obviously improved.

The invention belongs to the technical field of software development and operation and maintenance, and discloses a system for quickly responding to service iteration, which consists of a data source management module, a redis cache module, a configuration center module and an SPEL analysis and Naporn module. By introducing the configuration center system, centralized management and dynamic adjustment of business data output rules are achieved, the output rules are migrated to the configuration level from codes, the flexibility and maintainability of the system are greatly improved, when the rules are changed, only the configuration center needs to be updated, the codes do not need to be modified, the iteration speed is increased, and meanwhile the efficiency is improved. The historical record changing function of the configuration center not only facilitates tracking of rule changes and analysis of potential influences of the rule changes on services, but also can quickly roll back to a stable version when system instability or data glide is caused by configuration updating, so that risks are effectively reduced, and the mechanism ensures smooth proceeding of service iteration and provides powerful support for continuous optimization.

The present invention proposes a data access method and system applied to a teaching platform. The data access method applied to the teaching platform includes: configuring a Hadoop cluster of the teaching platform and defining a data stream; when there is a user's data access request on the teaching platform, using a streaming processing tool provided by writing into the Hadoop distributed file system to provide data information in a streaming manner, and displaying the data information on the user's smart device; recommending resource data based on the user's data access content combined with the user's learning history record; real-time monitoring of the data access operating parameters of the teaching platform, and judging whether to perform data access optimization according to the optimization strategy based on the data access operating parameters. The system includes modules corresponding to the steps of the method.

The application relates to the technical field of computers and discloses a general transmission control system bus signal simulation method.The method comprises the following steps: acquiring device node topology, protocol types and historical load data; constructing a bus network abstract model; extracting a message flow timing template; generating a concurrent message scheduling sequence conforming to arbitration logic; fusing a high-fidelity signal base; superimposing physical layer interference signals based on measured data; and injecting a measured system through a hardware-in-the-loop interface.The system comprises bus topology analysis, network modeling, timing template extraction, scheduling generation, signal fusion, interference injection and hardware output modules.The method significantly improves test authenticity, coverage and platform universality through joint modeling of logic and physical layers.

The invention relates to a training method and device for a hunting strategy network based on multi-agent reinforcement learning, and is applied to a liquid environment comprising at least two hunting persons and escapers. According to the method, respective action information of at least two purchasers can be obtained by using purse strategy networks corresponding to the at least two purchasers according to respective state information of the at least two purchasers and relative state information of the at least two purchasers and other purchasers and escapers at the current moment; and inputting the global information into the evaluation network to obtain score information for the respective action information of the at least two trainers so as to train the respective corresponding hunting strategy networks of the at least two trainers. A normal form of centralized training and decentralized execution is adopted, the stability and expandability of the algorithm in the complex environment are ensured, and efficient cooperative hunting of the bionic robotic fish in the complex environment is further achieved.

The present application provides a kind of coal mine water exploration intelligent management and control system and method, comprising: data acquisition module, for real-time acquisition in the process of coal mine water exploration multiple-source data;Data processing module, using microservice architecture and component design, for standardization integration and management to multiple-source data;Intelligent analysis module, for real-time analysis based on artificial intelligence machine vision technology to construction process video stream, including identifying drill rod quantity, personnel behavior and equipment state, and by three-dimensional modeling to real-time analysis drill hole trajectory;Intelligent early warning module, for according to real-time drill hole trajectory data and the dynamic comparison of preset design parameter, constructs multidimensional deviation early warning model, realizes risk classification early warning;Visual module;Management module.The present application realizes the whole process intelligent management and control of water exploration, effectively eliminates false exploration, leak exploration and other illegal behaviors, significantly improves the safety management level of coal mine water disaster prevention.

The invention relates to the technical field of talent management, and discloses a talent ability evaluation and post matching system based on a knowledge graph, and the system comprises a data processing module which collects individual information and target industry information, converts the individual information into an individual ability vector, and converts the target industry information into an industry demand vector; the graph construction module is used for constructing a cross-industry knowledge graph through multi-source knowledge fusion; the skill migration module is used for calculating skill migration complexity and a capability gap vector; the path planning module is used for generating a plurality of transformation paths and selecting an optimal transformation path based on a transformation path scoring function; according to the method, accurate matching of individual capabilities and post requirements is realized by constructing the cross-industry knowledge graph, and cross-industry skill migration paths can be identified, so that the system can discover potential matching opportunities which are difficult to identify by a traditional method, and the most suitable development paths are recommended for individuals.

The invention relates to the technical field of glass seal welding process optimization and defect prediction, in particular to a glass seal welding process optimization and defect prediction method based on deep learning. The method comprises the following steps: firstly, acquiring multi-source sensor electric signal data of a welding furnace in real time, and synchronizing and preprocessing to form a standard time sequence data sequence; and then, inputting the data sequence into a pre-trained multi-task deep learning model, and synchronously realizing dynamic process optimization and early defect prediction by the model through a shared feature extraction network and two parallel task branch networks. After the process is finished, the system associates an actual quality result with process data to form an incremental sample, and performs online fine adjustment on the model based on an intelligent trigger mechanism and an anti-forgetting algorithm, so that the system can adapt to changes of equipment and materials. According to the invention, the transformation of the glass seal welding process from fixed parameter control to real-time closed-loop intelligent optimization is realized, and the product yield, the process stability and the production intelligence level are effectively improved.