32 results about "Loose coupling" patented technology

The invention discloses a dynamic self-verification method based on service flow-test flow homologous heterogeneous execution. The method comprises the following steps: uniformly describing service logic and a corresponding interface test contract through a strong type of domain specific language (DSL); the method comprises the following steps: simultaneously compiling homologous DSL descriptions into stateful business workflows executed by a process engine and stateless and combinable atomic test contract sets called by a test executor by utilizing a double-path compiling engine; during operation, loose coupling synchronous triggering and execution of a service flow and a test flow are realized based on a high-throughput event bus, and real-time and high-fidelity contract verification, performance monitoring and chaos probe testing are performed on cross-system interface calling in a production environment. According to the invention, the traditional post-event and off-line integration test is converted into the in-event, on-line and continuous system endogenous capability, and the industrial problems that the complex enterprise system group integration test is difficult and the operation and maintenance monitoring is passive are effectively solved.

The invention discloses an application software architecture for encapsulating a task management and synchronization mechanism of an operating system, and belongs to the technical field of industrial real-time control. According to the architecture, an operating system task management and synchronization mechanism is modularly packaged into three types of objects: an employee type for packaging specific business execution logic; the pipeline class is used for packaging inter-task communication and synchronization functions; and the behavior class is used for packaging an operating system task entity and is responsible for scheduling employee objects contained in the behavior class to be executed orderly. The three types of objects are created, assembled and started through a factory mode according to configuration information to form a complete real-time task system. According to the method, through object-oriented packaging and configuration assembly, the problems of tight task coupling and difficult code maintenance in traditional process-oriented programming are solved, loose coupling and high cohesion among task modules are realized, the configurability, maintainability and expandability of industrial real-time software are remarkably improved, and the method is suitable for various real-time operating system environments.

The invention discloses a loose coupling unified task arrangement method supporting multi-dimensional expansion, which comprises the following steps of: firstly, defining a transmission agreement of task arrangement: constructing a unified intermediate representation structure for describing an arrangement process, the intermediate representation structure being used for recording a process topological structure, a node type and parameter configuration; secondly, on the basis of the intermediate representation structure, an executable directed acyclic graph is automatically constructed by adopting a child node matching construction algorithm based on queues and backtracking; and thirdly, executing an arrangement process based on the constructed directed acyclic graph: starting from all write nodes, recursively calling a data processing method of a father node until reaching a read node, and executing a complete data processing link in a bottom-up manner. Through a unified intermediate representation structure and a queue backtracking matching algorithm, automatic topology construction and dependency management of a complex process are realized, flexible expansion of node types is supported, development complexity is reduced, and adaptability, efficiency and maintainability of a data management system are improved.

The application is specifically a DPI shunt device rule management system and method, the system comprising: a service configuration table, wherein each service configuration item has a unique service ID; a rule configuration table, wherein each rule configuration item has a unique rule ID; a rule hash table, which quickly retrieves the rule ID by hashing the rule matching content; a policy control server, which issues service configurations to the DPI shunt device through a remote procedure call (RPC) interface, and the DPI shunt device splits the service configurations into one or more rule configurations according to a splitting algorithm; and a Redis database, which is used to persistently store the contents of the service configuration table, the rule configuration table and the rule hash table. The application has the advantages that: the table items of each management module are designed to be loosely coupled, allowing them to be independently developed and debugged; and the dual-ID management mode has good scalability and can be compatible with traditional single-ID rule management systems.

The invention relates to the technical field of data processing, and provides a multi-protocol data acquisition method and device, electronic equipment and a readable storage medium. The method comprises the following steps: firstly, sending a first driving request to an independently running acquisition driver for discovering an acquisition driver configured in an acquisition scheduling framework; then receiving driving protocol information sent by the acquisition driver based on the first driving request; in response to a received task request sent by the acquisition driver, issuing task metadata in a target structured format to the acquisition driver according to the driver protocol information; and finally, receiving the acquisition data of the target structured format acquired by the acquisition driver based on task metadata acquisition. The protocol compatibility of the data acquisition system can be improved, loose coupling of the data acquisition system is realized, the expandability of the system is improved, the maintenance cost of the system is reduced, and the reliability of the whole acquisition system is improved.

This invention relates to the field of artificial intelligence compiler technology, and discloses a method and system for rewriting heterogeneous partitioning of neural networks based on graph intermediate representation. The method includes: obtaining the intermediate representation of the high-level computational graph of the neural network, the hardware configuration of the heterogeneous hardware platform, and the description of the heterogeneous backend devices; extracting subgraph nodes and tensor edges of core operators from the intermediate representation of the high-level computational graph; constructing a cost model based on the hardware configuration; solving the heterogeneous backend device allocation result for each subgraph node; rewriting the allocation result back to the intermediate representation of the high-level computational graph in the form of attributes; generating a wrapper function marked with heterogeneous backend devices; and rewriting the computational graph structure inside the target forward function. This invention realizes hardware-aware graph-level automatic partitioning, tightly coupling the partitioning result with the high-level IR, and connecting to backend code generation through a unified wrapper function interface, solving the problems of existing technologies that rely on static rules, lack quantitative cost models, and have loose coupling between partitioning results and backend calls.

The embodiment of the application discloses a FPGA matrix transposition method and device based on AXI protocol. In the field of data transposition for SAR imaging processing on satellite, the AXI protocol is used, and a mixed coupling scheme combining a loose coupling scheme and a tight coupling scheme is proposed to reduce resource usage in the matrix transposition process. The mixed coupling scheme combines external storage of the tight coupling and internal transposition of the loose coupling, and more storage space is used than the tight coupling, but the mixed coupling scheme is suitable for a larger image size like the loose coupling scheme. Compared with the loose coupling scheme, the mixed coupling scheme reduces the cache resources, so that the limited cache resources improve more efficiency. In the actual application process, the mixed coupling scheme has certain adaptability.

The invention discloses a multi-enterprise loose coupling MDP Lagrange collaborative optimization system and method, and belongs to the technical field of port supply chain collaborative scheduling. According to the method, an optimization framework combining LC-MDP sub-models, Lagrange relaxation and a global coordination function C is adopted, cross-enterprise services are decomposed into a plurality of LC-MDP sub-models capable of being solved independently, global constraints are processed through Lagrange relaxation, multipliers are iteratively updated, strategy fusion and conflict resolution are achieved by means of the global coordination function C, and therefore the cross-enterprise services are obtained. And dynamically adapting the supply chain disturbance in combination with a three-layer architecture and a feedback mechanism. According to the invention, extensible collaborative optimization of large-scale multi-enterprise business is realized, distributed training and real-time scheduling are supported, the problems of resource conflicts and dynamic decision making are effectively solved, and supply chain collaborative efficiency and stability are improved.

The present application relates to a memory-enhanced deep personalized input method system based on end-side large language model, and belongs to the technical field of artificial intelligence. The present application solves the problems of the prior art, such as the loose coupling of the model and the input core process, the shallow personalization mechanism, the delay and privacy risk caused by cloud deployment, and the insufficient efficiency of the end-side operation that cannot meet the real-time input requirements. The system includes a front-end real-time interaction module, a stylized large model post-training module, a hierarchical memory mechanism module, and an end-side reasoning optimization module. The present application can realize low-delay real-time text generation under the condition of highly limited mobile terminal resources, and at the same time, through continuous modeling of user input behavior, it can complete deep personalization adaptation and ensure the end-side retention and privacy security of user input data throughout the process.

The invention discloses a multi-agent collaborative industrial artificial intelligence algorithm evaluation system and method, and aims to solve the problems of disordered evaluation process, insufficient compliance guarantee and poor system expansibility in the prior art. The system adopts a six-level layered architecture, the core is a navigation agent, and the navigation agent monitors the task state in a global data layer, performs dual verification in combination with national standard rules, and schedules data division, script writing, algorithm parameter adjustment, report generation and other special agents for sequential execution only when the state is compliant. And each agent interacts through the standardized JSON and calls a tool based on MCP packaging. According to the invention, automation, ordering and national standard compliance of the evaluation process are realized, flexible expansion is supported through loose coupling design, and comparability and traceability of the evaluation result are guaranteed.

The invention discloses a parallel simulation construction and operation method for complex scene multi-domain loose coupling segmentation, and belongs to the field of combat simulation, and the method comprises the following steps: 1, starting a scenario editing tool, editing a complex scene scenario file, dividing scenario entities into different combat domains based on combat domain loose coupling segmentation, enabling each combat domain only to have less interaction, and enabling each combat domain to have less interaction; storing the scenario file; 2, setting a multi-domain interaction message, defining a message structural body in a multi-domain interaction middleware MDI, and adding a message sending function and a message processing function in a related Agent; thirdly, starting a scenario operation monitoring tool, opening a complex scene scenario file, and distributing one or more operation nodes for each combat area; fourthly, complex scene scenario operation is started, synchronous operation is achieved based on multi-node and multi-domain simulation, and message transmission among multiple domains is transmitted through MDI. According to the invention, the simulation calculation amount and the communication requirement of each simulation node are reduced, so that the simulation realization of a multi-node complex scene has operability.

The invention discloses a large model application computing power-model-application break-through method. The break-through method comprises the steps that an application building platform, a knowledge base platform, a large model square, a tool platform and a computing platform are built; designing a loose coupling module; independent life cycle management; the computing power, the model and the application are communicated from bottom to top; isolating a self-built tenant system from a data surface; modularized supply of application functions is realized; and zero-shutdown knowledge updating. The problem that the whole package must be restarted during upgrading in the prior art is solved.

The invention provides a distributed asynchronous communication method and device, computer equipment and a storage medium, and is applied to the field of finance or medical health, and the method comprises the steps that a first entity receives a task instruction, generates or updates an instance file according to preset resource information and the task instruction, and submits the instance file to a shared warehouse, the instance file comprises an expected state field and an observed state field, and the first entity only modifies the expected state field; when monitoring that the shared warehouse is changed, the second entity pulls and analyzes the instance file from the shared warehouse to obtain an analysis result; the second entity generates an execution result after executing the task according to the analysis result, modifies the observation state field according to the execution result, and submits the modified instance file to the shared warehouse; and when the first entity monitors that the shared warehouse is changed, the modified instance file is pulled from the shared warehouse, and an execution result is analyzed and obtained. Cross-network and cross-organization loose coupling asynchronous communication capable of carrying out bidirectional service communication is realized.

The invention discloses an intelligent retrieval method based on industrial intelligence data and electronic equipment, and aims to solve the three problems of high information noise interference, low entity and relationship extraction precision and performance bottleneck caused by tight system coupling in the prior art. According to the system, a modular loose coupling design is adopted, functions such as noise recognition, entity extraction and relation mining are decoupled into independent service units, and cooperation is carried out through a standardized interface. The core is to construct an integrated intelligent processing mechanism, so that noise filtering and semantic analysis share features in a unified framework, collaborative optimization is realized, and error accumulation is avoided while the data quality is improved. The system has high expandability and maintainability, supports independent upgrading of modules and on-demand expansion and contraction of resources, remarkably improves intelligence processing precision and efficiency, and provides high-quality and high-credibility intelligence support for enterprise strategic decision making.

The invention provides a knowledge base management and retrieval system and retrieval method based on VS Code double plug-ins. The knowledge base management and retrieval system comprises an LjCode plug-in and an LjKB plug-in which are independently deployed on a VS Code client, and a Dify platform which provides bottom support. The LjCode plug-in and the LjKB plug-in realize loose coupling collaboration through a VS Code standardized communication interface, and the LjCode plug-in and the LjKB plug-in establish data interaction with a Dify platform through a standardized API (Application Program Interface). According to the method, the LjKB plug-in is configured, the limitation that a traditional auxiliary programming plug-in only completes error correction is broken through, personal knowledge base retrieval and development integration of an integrated development environment are achieved, multi-language multi-format personal base creation and sharing are supported, and the Dify native support format range is expanded; a high-quality vertical built-in knowledge base of various different code languages is built in through an LjCode plug-in, and a bottom layer development scene is covered in a full range; and double plug-ins of the LjCode plug-in and the LjKB plug-in are designed to be loosely coupled and can operate independently or cooperatively, so that secondary development is facilitated.

The invention relates to the technical field of data processing and intelligent decision making, in particular to a dynamic business decision making system driven by data assets, and solves the problems that the business decision making system has obvious defects, the problem of data islands is serious, cross-department and cross-business data is difficult to fuse to form unified and reusable data assets, and the decision making view angle is one-sided. The method solves the problem that the existing decision-making process is mostly static or semi-static and seriously depends on historical data and preset rules, and an agile response cannot be made to a real-time changing business environment, and comprises a data capitalization layer which is used for being in butt joint with a multi-source heterogeneous data source. According to the method, a unified data capitalization layer and a modularized system architecture are constructed, multi-source heterogeneous data are integrated into standard and reusable high-quality data assets, loose coupling management of features, models and decision-making processes is achieved, the problem of data islands is solved, and the reuse value of the data and the flexible expandability of the system are remarkably improved.

The invention discloses a fluid-solid-heat intelligent loose coupling method, device, medium and product, and relates to the field of multi-physics field coupling calculation. The method comprises the following steps: constructing an appearance feature difference amplification model and an aerodynamic force intelligent prediction model for the tiny deformation of the aircraft; an aerodynamic heat intelligent prediction model under the variable-configuration variable-incoming-flow condition is constructed based on a local heat wall correction technology; and constructing an intelligent data transmission model oriented to the flow-solid interface non-matched grid. Executing multi-field coupling propulsion of dynamic time step based on the model; predicting aerodynamic force, structural deformation and aerodynamic heat distribution by using an intelligent model; and adaptively adjusting the time step length according to the change degree of the aerodynamic heat environment, if the change exceeds a threshold value, carrying out time domain propulsion by adopting numerical calculation and updating each intelligent model on line, otherwise, directly predicting the next time step by utilizing the intelligent model. According to the method, the coupling step length can be adaptively adjusted according to the change intensity of the physical field, and the calculation efficiency is remarkably improved on the premise of ensuring the precision.

This invention discloses a dynamic self-verification method based on the homogeneous and heterogeneous execution of business flow and test flow. The method includes: uniformly describing business logic and corresponding interface test contracts using a strongly typed Domain-Specific Language (DSL); utilizing a dual-path compilation engine to simultaneously compile the homogeneous DSL description into a stateful business workflow for execution by the process engine and a stateless, composable set of atomic test contracts for invocation by the test executor; and at runtime, achieving loosely coupled synchronous triggering and execution of business flow and test flow based on a high-throughput event bus, performing real-time, high-fidelity contract verification, performance monitoring, and chaos probe testing on cross-system interface calls in a production environment. This invention transforms traditional post-event, offline integration testing into in-process, online, and continuous system-native capabilities, effectively solving the industry problems of difficult integration testing of complex enterprise system groups and passive operation and maintenance monitoring.

This invention discloses a smart lock application software simulation system, comprising a hardware driver layer, a functional component layer, a bridging layer, an operating system layer, and an application layer. Each layer communicates via standardized interfaces. The hardware driver layer shields hardware differences, and the bridging layer handles cross-layer data adaptation, resolving the hardware-software coupling problem. This allows application development to be completely decoupled from the hardware foundation, and the code can be directly deployed after debugging, improving development efficiency and code reusability. The task isolation architecture allows the application layer to run in independent tasks, and the component-based design achieves loose coupling between functional modules. This avoids single-module malfunctions affecting the overall system while supporting dynamic module loading and updates, improving system stability and scalability. Furthermore, the combination of a visual development platform and an AI engine lowers the development threshold, and the remote monitoring and management functions further enhance system maintainability and management efficiency.

The invention relates to the technical field of computer software, in particular to a unified session control method and system based on page execution capability constraints, and is suitable for a multi-service page decoupling system of a loose coupling architecture. According to the method, pre-binding is carried out on a page capability version in a session suggestion generation stage to solidify an expected execution environment, after a user confirms, an execution token is utilized to further combine and bind the execution environment, and dual verification of version consistency and page instance consistency is carried out by taking pre-binding solidification environment information as a unique reference in an execution stage. And the business operation can be executed only after the verification is passed, and the execution condition is converted from dynamic judgment during operation to structured constraint during generation. The problem that cross-page execution environments are inconsistent is effectively solved, and session control safety and stability are improved.

The invention provides a low-code-platform-oriented plug-in dependency analysis and architecture extension method and system, and relates to the technical field of low-code platforms, and the method comprises the steps: converting a plug-in source code into an abstract syntax tree through static code analysis, extracting an interface contract, and verifying the compatibility; cyclic dependency is eliminated through dependency graph construction and topological sorting, and a compatible version is determined to generate a dependency relationship chain; and calculating a minimum influence surface based on the plug-in calling index table, constructing a function extension registration center and an interface directory, and realizing loose coupling calling between the plug-ins. According to the invention, the problem of dependency conflict of the plug-ins is solved, and the expansibility and stability of the platform are improved.

The invention belongs to the technical field of radar signal processing and high-performance computing, and particularly relates to a GPU + CPU-based heterogeneous information processing system and method.The GPU + CPU-based heterogeneous information processing system comprises a layered software system structure of a presentation layer, a domain layer, a persistent layer and a supporting layer, and loose coupling data interaction of components among layers and in the layers is achieved through a data management component; the presentation layer runs on a CPU main thread and is used for man-machine interaction and state monitoring; the domain layer runs on a CPU and is a core business logic layer, and the persistent layer runs on the CPU, is used for processing localized storage of data and comprises a process file storage module, an A display data storage module and a high-resolution file storage module; and the support layer runs on the CPU and provides bottom layer support service. According to the invention, deep collaboration of the CPU and the GPU is realized, and the overall efficiency, flexibility and reliability of complex radar signal processing are remarkably improved.

The invention discloses a multi-system fused wooden door manufacturing enterprise information system integration architecture and method, which adopts a layered loose coupling design, and comprises an enterprise operation layer which comprises an enterprise resource planning system for order and financial management and a product life cycle management system for product model and process management; the planning and scheduling layer comprises an advanced planning and scheduling system which serves as a planning decision center and carries out limited productivity simulation and optimized scheduling based on multi-source real-time data, and the manufacturing operation layer comprises a manufacturing execution system used for production execution and monitoring and a warehouse management system used for fine management of materials. By constructing a unified data model, adopting an event-driven loose coupling collaboration mechanism and establishing a middle platform for converging real-time manufacturing data, full-process digital communication and intelligent linkage from customer orders, product design, material supply, intelligent scheduling, manufacturing execution to warehouse logistics are realized.

The invention discloses an avionics equipment-oriented loose coupling automatic test system. The innovative architecture of the avionics equipment-oriented loose coupling automatic test system is mainly composed of three core parts, namely a virtual resource warehouse, a unified communication base and an application unit. The virtual resource warehouse is used for abstracting physical test resources and mapping the physical test resources into uniform logic resources to realize resource decoupling; the unified communication base is formed by fusing a data network, a control network and a storage network, provides a unified interaction channel for all elements of the system, and realizes interaction decoupling; and the application unit is constructed on the base and calls virtual resources to realize service functions of test control, execution, management and the like. According to the system, through the triple decoupling architecture, a test program can be developed independently, system functions can be recombined flexibly, different avionics devices and test environments can be adapted at small cost, and the development complexity and life cycle cost of the test system are reduced fundamentally.

The application provides a plug-in dependency analysis and architecture extension method and system for a low-code platform, relates to the technical field of low-code platforms, and comprises the following steps: converting plug-in source code into an abstract syntax tree through static code analysis, extracting an interface contract and verifying compatibility; eliminating circular dependencies through dependency graph construction and topological sorting, determining a compatible version to generate a dependency relationship chain; calculating a minimum impact surface based on a plug-in call index table, constructing a function extension registration center and an interface directory, and realizing loose coupling calls between plug-ins. The application solves the plug-in dependency conflict problem and improves the platform extensibility and stability.

The invention provides a cross-medium unmanned aerial vehicle flight control system based on a modular loose coupling architecture. The system comprises two core modules, namely a task manager and an adapter. The task manager is an independent process based on a finite-state machine and is responsible for managing and controlling a task execution process and system state transition, and the internal state of the task manager covers waiting, take-off, multiple waypoint task modes, landing and completion states. The adapter serves as an independent process and is realized by adopting an adapter mode and interface-oriented programming, high-level control logic and bottom-level specific implementation are decoupled through abstract base classes and runtime polymorphisms, and a flight control adapter, a planner adapter and a positioning adapter are included in the adapter. According to the system, various message flows are transferred and converted through an adapter, data intercommunication among positioning, planning and control modules is realized, and opening and closing of adaptive channels are allowed to be controlled through non-compilation parameter configuration. According to the invention, the modularization degree, the expandability and the cross-platform adaptation capability of the unmanned aerial vehicle control system are improved.

The invention provides a processing architecture for super-large-scale brain-like intelligent heterogeneous fusion, which refers to a function partition and processing loop mechanism of a brain, and abstracts, isolates and schedules physical resources of a brain-like platform through a virtualization technology and a scheduling deployment tool. Furthermore, a calculation model and a calculation task provided by the user are deployed in a brain-like system, so that the flexibility, the safety and the hardware resource utilization rate of platform application are improved; on the basis of a platform integration development environment, rich software and hardware interfaces and a rapid design method, open application ecology is achieved, interactive communication, scheduling, integration and the like are carried out through simple, flexible and accurately defined module interfaces aiming at different applications by designing a coarse-grained and loosely-coupled integration architecture, and the complexity of underlying programming is reduced.

The invention discloses an embedded upgrading automation script execution engine and method based on a dynamic reconstruction function network, and belongs to the field of embedded system upgrading. The engine comprises a core execution layer, a scheduling coordination layer, a life cycle management layer and a monitoring verification layer. The core execution layer explains and executes the multi-language script through a plug-in executor; the scheduling coordination layer analyzes declarative configuration to define a script execution process, and schedules a task based on a directed acyclic graph; the life cycle management layer manages an upgrade process state through a state machine, and integrates a backup and rollback mechanism; and the monitoring verification layer executes verification rules, error processing strategies and resource monitoring. Each layer communicates through a predefined standardized interface, the engine realizes loose coupling integration of the script and the engine through a unified script interface specification, and the specification defines a standardized metadata description file and a life cycle hook for the script. Unified management, automatic execution and dynamic reconstruction of the multi-language scripts are achieved, and the reliability and flexibility of upgrading of the embedded system are improved.

This application relates to the field of computer software technology, and in particular to a unified session control method and system based on page execution capability constraints, applicable to loosely coupled multi-service page decoupling systems. This method pre-binds page capability versions during the session suggestion generation phase to solidify the expected execution environment. After user confirmation, it further binds the execution environment using an execution token. During the execution phase, it performs dual verification of version consistency and page instance consistency based solely on the pre-fixed environment information. Only after successful verification can business operations be executed, transforming execution conditions from runtime dynamic judgment to structured constraints at generation time. This effectively solves the problem of inconsistent execution environments across pages, improving the security and stability of session control.

The invention provides a large language model reasoning method and device and a storage medium, a large language model reasoning architecture comprises a plurality of function-decoupled and independently-expandable micro-services, and the micro-services are in communication cooperation through a predefined interface protocol, so that each service can be independently and transversely expanded according to an actual load and a resource demand, and the efficiency of the large language model reasoning architecture is improved. And the expandability and the reliability of the system are improved. During model reasoning, reasoning requests are received and analyzed through an arrangement scheduling service, the reasoning requests are disassembled into a series of standardized reasoning tasks, appropriate computing resources can be dynamically allocated to each task during reasoning, and global resource scheduling and task arrangement are achieved. Then, according to the reasoning task execution graph, a plurality of micro-services are scheduled for cooperative reasoning, and due to loose coupling among the micro-services, different instances can be scheduled to execute the reasoning task in parallel, so that the response time of the request is remarkably shortened, and the overall throughput and resource utilization rate of the system in a high-concurrency scene are improved.