13 results about "Process configuration" patented technology

The application provides a configuration-driven DL4NLP model automatic training and deployment method, which comprises the following steps: parsing a YAML configuration file and verifying parameters to generate a unified process configuration; automatically loading data and performing preprocessing according to the configuration to generate a model input tensor; instantiating model components through a component factory and assembling the model components into a complete network model; automatically configuring a training environment to train and evaluate the model to obtain an optimal model; exporting the model into a TorchScript format binary file; and automatically generating a standardized deployment package based on a template engine. The application abstracts the whole process of model development and deployment into a configuration-driven standardized pipeline by parsing the YAML configuration file, automatically performs data loading and preprocessing, model component instantiation and assembly, training optimization and deployment package generation, and solves the problems of low development efficiency, long cycle and technical gap between training and deployment caused by repeated code writing and manual conversion and deployment for different tasks in traditional model training and deployment.

The application discloses a kind of low code industrial control method and equipment, method includes: based on standardization CSV configuration mode, communication object is configured, obtains communication object configuration file;Based on standardization CSV configuration mode and standardization JSON parameter, process node is configured, obtains process node configuration file;Communication object configuration file and process node configuration file are parsed, obtain the communication instance corresponding to industrial equipment, and node instance for controlling industrial equipment, process configuration file includes the node order between multiple node instances;Communication instance converts the control logic of node instance into control instruction according to node order, and through control instruction, industrial equipment is dynamically controlled and process start-stop management is carried out.The present application can reduce the configuration complexity, improve the configuration efficiency, scalability and debugging efficiency.

This invention provides an online approval method and system for automatically generating write-off materials through process configuration, belonging to the field of online approval technology. Specifically, it includes: generating write-off materials to be approved using contract and loan information; forwarding the write-off materials to approvers; determining the urgency assessment value of the write-off materials based on the remaining approval time, the number of remaining approvers, and the number of approval projects each remaining approver is responsible for; determining the base review time for the write-off materials based on the type of write-off materials, loan amount, and urgency assessment value; when the approver's review time exceeds the base review time, constructing the approver's review detail based on the focus, review time, and review time of different pending review positions during the approval process; and determining whether the review is complete based on the review detail and generating completed write-off materials, thereby further improving the efficiency and accuracy of the review.

This invention discloses an intelligent financial risk control system based on ABTest technology, belonging to the field of financial risk control technology. It includes a process configuration module, a strategy management module, an ABTest management module, a decision execution engine, a verification and analysis module, a machine learning model integration module, and a real-time monitoring and early warning module. The system allows for the configuration of risk control decision-making processes in a visual manner and supports the flexible integration of various strategy tools and machine learning models. Utilizing ABTest and a champion-challenger (Champion Challenger) mechanism, it conducts parallel testing and scientific evaluation of online strategies and challenger strategies. Based on statistical indicators and model performance indicators, it automatically analyzes and recommends the optimal strategy, achieving intelligent switching and iteration of strategies. This invention features high flexibility, scalability, and intelligence, improving the accuracy, efficiency, and system stability of risk control decisions. It is suitable for the risk management and business optimization needs of financial institutions in rapidly changing business environments.

This application provides an AI-based social security intelligent service management method based on process configuration. By establishing a material-item association matrix and dynamically adjusting the necessity weight of associations based on historical case data, and accurately deriving the list of missing materials for subsequent nodes based on the actual nodes and process dependencies covered by the user's submitted materials, rather than relying on simple comparisons using fixed templates, this significantly improves the accuracy and timeliness of material correction guidance. Furthermore, this invention combines image quality assessment, local enhancement processing, and the write-back update of structured field information. This allows areas with problems such as blurriness, occlusion, and glare to re-participate in OCR recognition and update downstream data after targeted enhancement, avoiding the duplicate submission problem caused by traditional global rejection mechanisms. This ensures the continuity of material review and data consistency, constructing a closed-loop management mechanism of "recognition-verification-control," significantly improving the automation level of social security intelligent service management.

The application discloses an intelligent agent scheduling method and device, equipment and a computer readable storage medium, applied to the field of artificial intelligence, comprising: when receiving a service request submitted by a service system, generating a preset process identifier for the service request; loading process configuration according to the preset process identifier, determining the target intelligent agent and the intelligent agent calling sequence; the process configuration is used for defining the calling sequence and parameter mapping rule of the intelligent agent; according to the intelligent agent calling sequence, each target intelligent agent is called in turn, and based on the ability configuration associated with the target intelligent agent, a task execution result is obtained by executing the task; the ability configuration is used for defining the specific execution ability of the intelligent agent. The application separates the process configuration and the ability configuration, so that the arrangement of the business process and the specific execution ability of the intelligent agent are independent of each other, the coupling degree between modules is effectively reduced, and the scalability, maintainability and reusability of the system are improved.

The invention describes a computer-implemented method for planning and realizing a work process in engineering and production, comprising conducting a dialogue with a user, including displaying (130) known work steps of the work process and, for each of the work steps, displaying possible implementations of the work step; recording (140) a first selection by the user, wherein the first selection for one or more of the work steps includes one of the displayed implementations, the selected implementations being implementations currently used by the user; recording (150) a second selection by the user, wherein the second selection for each of the work steps includes one of the displayed implementations, the selected implementations being implementations desired by the user;Based on the first and second selections, determine (160) the additional work steps, implementations and technical resources required for the intended implementations that must be provided by the user; and display (170) the required work steps, implementations and resources.

The application discloses a process reuse method and device, electronic equipment and a nonvolatile storage medium. The method comprises the following steps: constructing a first target node object and a first transfer node object corresponding to the first target node object according to target node data; controlling the first transfer node object to execute business logic corresponding to the first target node object in sequence according to target relationship data, and setting a first target pointer to point to the first target node object being executed; in the case that a next target node object planned to be executed jumps to a sub-process, setting a second target pointer to point to a second target node object being executed in the sub-process, so as to realize reuse of the sub-process. The application solves the technical problems of tedious and complex process configuration, low process configuration efficiency caused by the fact that most of the existing process configuration, construction and execution methods do not support reuse mechanism.

This invention discloses a hierarchical control system and method with configurable state and event-driven chains, relating to the field of industrial automation control technology. The system includes a modular execution layer, a configurable aggregation engine, an event chain editor, and an event-driven execution layer. The modular execution layer completes atomic operations and publishes completion events through functional atomic modules. The configurable aggregation engine dynamically calculates and reports the system's aggregated state according to predefined rules. The event chain editor visualizes the orchestration process and generates independent configuration files. The event-driven execution layer loads files and is driven by event-driven processes. This invention achieves flexible state definition and visualized process configuration, realizing logical decoupling and improving system flexibility and maintainability.

This invention discloses a quality control method and system for the production process of rubber liners. It acquires production parameter data and process configuration parameters, generates process matching coefficients through cross-process linkage analysis, statistically analyzes defect distribution and maps it to a process control list to establish a hierarchical quality control scheme. Based on the process matching coefficients, it weights and adjusts the hierarchical quality control schemes to form a comprehensive control scheme, identifies deviation transmission amplification and complete absorption anomalies to generate detection anomaly alarms, and determines the graded quality control benchmark. It executes hierarchical corrections to generate monitoring feedback tables, tracks the narrowing rate differences between shifts to generate quality control trend charts, and triggers false-qualification warnings when detection parameters quickly reach the standard to form a qualification judgment level. It adjusts the detection frequency to generate process qualification certificates, screens multi-process synchronous critical batches to form quality stability labels, and outputs rubber liner production quality control instructions, realizing structured control of cross-process and cross-shift quality risks in the rubber liner production process.

This application provides a management system and method for nuclear power plant operation processes, relating to the field of nuclear power plant operation and maintenance technology. The management system for nuclear power plant operation processes includes: a data support module storing multiple inherent state sequences corresponding to various operation scenarios, each inherent state sequence comprising multiple sequentially set inherent states; a user application module for receiving template configuration instructions; and a configuration and driving module including a process configuration unit, which generates a corresponding state flow template based on the template configuration instructions and the corresponding inherent state sequences. The state flow template includes multiple sequentially connected custom node groups, each custom node group of the state flow template corresponding to each inherent state of the corresponding inherent state sequence. The configuration and driving module stores the business rules corresponding to the inherent states, generates process instances corresponding to the state flow templates, and advances process instances according to the business rules.

The application provides an additive manufacturing process parameter matching method and system, and relates to the technical field of metal additive manufacturing.The three-dimensional model of a part to be manufactured is introduced and converted into point cloud data, and standardized point cloud data is obtained through decentralization, scale normalization and uniform sampling; the local geometric features and global topological features of the data are extracted, and the entity structure of the three-dimensional model is partitioned; process parameters are matched in a mapping database based on the structure type, and when there are multiple groups of parameters, the optimal parameters are selected through a forming quality evaluation function, and the scanning path strategy is one-to-one corresponding to the structure type; the parameters are converted into control instructions to generate a process configuration file.The application realizes accurate matching of structure-level parameters, reduces manual intervention, improves forming quality and manufacturing stability, completes the automation of the closed loop from design to manufacturing, and is suitable for different additive manufacturing equipment.