Brain's perceptual and cognitive system
By constructing a brain-based perception and cognition system, the problems of incomplete data preprocessing and rudimentary access control were solved, enabling efficient and secure multi-source data processing and management, and improving the system's scalability and operational efficiency.
Patent Information
- Application Number
- CN202611158064.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-07-31
- Publication Date
- 2026-11-17
AI Technical Summary
Existing data sensing systems cannot meet the requirements of efficiency, security, and scalability. They suffer from incomplete data preprocessing and coarse-grained access control, resulting in low data processing efficiency and high security risks.
The brain perception and cognition system is constructed, including a source system adaptation module, a data perception and preprocessing module, a task management module, a security authentication module, a quality assessment and state perception module, and a dynamic rule learning and optimization module, so as to realize data preprocessing at the perception end, refined permission management, and full-process security control.
It significantly improves the compatibility of multi-source data integration, reduces downstream processing load, lowers data security risks, improves operation and maintenance efficiency, and ensures data reliability and consistency.
Smart Images

Figure CN123116236A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of data perception and processing technology, specifically to a brain-based perception and cognition system. Background Technology
[0002] As digital transformation deepens, business systems in enterprises and institutions are becoming increasingly complex, and data sources are exhibiting diversified and heterogeneous characteristics, commonly including API interface data, IoT device data, and enterprise-level database data. To achieve effective utilization of various types of data, there is an urgent need to build a perception and cognition system capable of coordinating multi-source data.
[0003] However, existing data sensing systems can no longer meet the requirements of efficiency, security, and scalability in practical applications, and there are many problems that urgently need to be solved, as detailed below:
[0004] 1. Lack of data preprocessing, increasing downstream processing load: Existing systems often directly transmit collected data to the data warehouse without preprocessing at the sensing end. For example, an organization's customer information data is scattered across multiple business systems. System A stores customer phone numbers as encrypted strings, System B stores plaintext phone numbers, and System C contains some abnormal data with empty phone number fields. Existing systems directly collect this data and connect it to the data warehouse, requiring the data warehouse to develop additional processing logic for encryption / decryption, outlier filtering, and format standardization. This reduces data processing efficiency by more than 40% and easily leads to data inconsistency issues.
[0005] 2. Coarse-grained access control, high data security risks: Existing systems mostly focus on access control at the system login level, failing to set granular permissions for the entire process of data collection, processing, and viewing. For example, in a company's sales data awareness system, ordinary operators could view all customers' sensitive sales data and arbitrarily configure collection tasks. There have been security incidents where operators misconfigured tasks, leading to the collection of core customer data to unauthorized systems, resulting in serious data leakage risks. Summary of the Invention
[0006] To address the aforementioned technical problems of incomplete data preprocessing and coarse-grained access control, this invention provides the following technical solution:
[0007] The brain's perceptual and cognitive system includes:
[0008] The source system adaptation module parses source system information, matches docking rules, and builds an adaptation channel;
[0009] The data perception preprocessing module, based on the adaptation channel of the source system adaptation module, collects multi-type source system data after it has been connected to the source system adaptation module, and outputs structured data after parsing, cleaning, encoding and standardization preprocessing.
[0010] The task management module configures tasks based on the adaptation channel requirements of the source system adaptation module and the preprocessing requirements of the data perception preprocessing module. It also adopts and executes tasks based on the corresponding scheduling strategy, monitors the data during task execution, and outputs management data.
[0011] The security authentication module, based on the control data from the task management module, performs full-process security control over system operations, including identity authentication, permission allocation, and authentication auditing.
[0012] The quality assessment and status awareness module collects structured data from the data awareness preprocessing module and operational data from the source system adaptation module, data awareness preprocessing module, task control module, and security authentication module. It assesses data quality and system status and displays the results visually. At the same time, it outputs the quality assessment results and operational data of each module.
[0013] The dynamic rule learning and optimization module, based on the quality assessment results of the quality assessment and state awareness modules and the operation data of each module, autonomously learns and optimizes adaptation rules, preprocessing rules and control strategies.
[0014] The data sensing preprocessing module includes:
[0015] After receiving the adaptation channel output by the source system adaptation module, the data acquisition unit selects the corresponding acquisition method based on the source system type obtained by parsing the source system information, and temporarily stores the acquired data in the buffer.
[0016] The format parsing unit identifies the data format type based on the data collected in the buffer of the data acquisition unit, calls the corresponding parsing algorithm to perform structured parsing of the data, and generates structured data.
[0017] The data cleaning unit optimizes the quality of the structured data from the format parsing unit based on preset cleaning rules.
[0018] The encoding standardization unit calls a preset encoding conversion rule library to perform unified encoding processing on the data after the data cleaning unit has optimized its quality.
[0019] As a preferred embodiment of the brain perception and cognition system described in this invention, the source system adaptation module includes:
[0020] The source system information parsing unit receives basic information from external source systems, performs standardized parsing of the information, and generates a source system feature dataset.
[0021] The adaptation rule matching unit matches the corresponding docking rules from the preset adaptation rule library based on the source system feature dataset of the source system information parsing unit; if no completely consistent rule is matched, a basic adaptation framework is automatically generated.
[0022] The interface / protocol adaptation unit constructs a dedicated adaptation channel based on the docking rules or basic adaptation framework output by the adaptation rule matching unit.
[0023] As a preferred embodiment of the brain perception and cognition system described in this invention, the task management module includes:
[0024] The task configuration unit provides a visual configuration interface, which supports users in configuring the adaptation channel requirements of the source system adaptation module and the preprocessing requirements of the data perception preprocessing module to complete the task configuration and output task information.
[0025] The task scheduling unit receives task information output by the task configuration unit and adopts the corresponding scheduling strategy based on the task type.
[0026] The task execution unit receives the scheduling strategy from the task scheduling unit, calls the source system adaptation module and the data perception preprocessing module to execute the task, and provides real-time feedback on the task execution progress and status.
[0027] The task status management unit receives the task execution progress and execution status feedback from the task execution unit, and classifies and manages the task status.
[0028] The task monitoring unit collects the task execution progress and status feedback from the task execution unit in real time and generates visual charts.
[0029] As a preferred embodiment of the brain perception and cognition system described in this invention, the security authentication module includes:
[0030] The user authentication unit uses a multi-factor authentication method of username + password + verification code to verify the identity of users logging into the system;
[0031] The permission allocation unit, after the user identity authentication unit completes the identity authentication, uses a role-based access control model to pre-set multiple types of roles and, in conjunction with the management data from the task management module, assigns differentiated permissions to different roles.
[0032] The operation authentication unit, after completing the differentiated permission allocation by the permission allocation unit, receives user operation requests in real time, verifies the validity of the user access token and the scope of permissions, and performs authentication operations.
[0033] The security log auditing unit records all user operations after completing the authentication process of the operation authentication unit.
[0034] As a preferred embodiment of the brain perception and cognition system described in this invention, the quality assessment and state perception module includes:
[0035] The data quality acquisition unit collects structured data output from the data perception and preprocessing module in real time and extracts key data quality indicators.
[0036] The data quality assessment unit evaluates the key data quality indicators of the data acquisition unit based on preset quality thresholds, generates a data quality assessment report, and outputs the quality assessment results.
[0037] The system operation status acquisition unit collects real-time operation data from the source system adaptation module, data perception and preprocessing module, task control module, and security authentication module.
[0038] The status visualization unit visualizes the quality assessment results of the data quality assessment unit and the operating data of each module of the system operation status acquisition unit.
[0039] As a preferred embodiment of the brain perception and cognition system described in this invention, the dynamic rule learning and optimization module includes:
[0040] The learning data acquisition unit collects the adaptation channel records of the source system adaptation module, the preprocessing logs of the data perception preprocessing module, the quality assessment results of the quality assessment and status perception module, and the control data of the task control module in real time, extracts key data, and forms a learning dataset.
[0041] The rule analysis and optimization unit calls machine learning algorithms to perform in-depth analysis on the learning dataset of the learning data collection unit, explore the potential for rule optimization, and output the optimized rules.
[0042] The rule verification and update unit generates candidate solutions from the optimized rules of the rule analysis and optimization unit, verifies them in a simulated environment, compares the data status before and after optimization, and automatically synchronizes them to the rule base of the corresponding module after verification.
[0043] Compared with existing technologies:
[0044] By constructing a flexible adaptation framework and a rich adaptation rule base, the compatibility of multi-source data integration is significantly improved, effectively reducing the expansion costs of adapting to new source systems. Data preprocessing is moved to the sensing end, completing operations such as format parsing, cleaning, and encoding standardization, significantly reducing the load on downstream data processing stages. Multi-factor authentication and role-based fine-grained permission control are adopted, combined with full-process operation authentication and log auditing, significantly reducing data security risks. Integrated management and control of task configuration, scheduling, status management, and monitoring are achieved, breaking the disconnect between task management and monitoring and improving operational efficiency. A real-time data quality collection and evaluation mechanism is established, synchronously sensing and visually displaying the system's operating status, ensuring data reliability, and providing solid support for subsequent decision-making and applications. Attached Figure Description
[0045] Figure 1 This is a schematic diagram of the overall framework of the present invention;
[0046] Figure 2 This is a schematic diagram of the source system adaptation module framework of the present invention;
[0047] Figure 3 This is a schematic diagram of the data perception preprocessing module framework of the present invention;
[0048] Figure 4 This is a schematic diagram of the task management module framework of the present invention;
[0049] Figure 5 This is a schematic diagram of the security authentication module framework of the present invention;
[0050] Figure 6 This is a schematic diagram of the quality assessment and state perception module framework of the present invention;
[0051] Figure 7 This is a schematic diagram of the framework of the dynamic rule learning and optimization module of the present invention. Detailed Implementation
[0052] To make the objectives, technical solutions, and advantages of the present invention clearer, the embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.
[0053] This invention provides a brain-based perception and cognition system; please refer to [link / reference]. Figure 1 ,include:
[0054] The source system adaptation module parses source system information, matches docking rules, and builds an adaptation channel;
[0055] The data perception preprocessing module collects data based on the adaptation channel of the source system adaptation module, and outputs structured high-quality data through parsing, cleaning, encoding and standardization.
[0056] The task management module configures tasks, schedules execution, and monitors status based on the adaptation requirements of the source system adaptation module and the preprocessing requirements of the data perception preprocessing module.
[0057] The security authentication module, based on the control data from the task control module, performs full-process security control of identity authentication, permission allocation, and authentication auditing for all operations.
[0058] The quality assessment and status awareness module collects preprocessed data from the data awareness and preprocessing module, as well as operational data from each module, to assess data quality and system status and visualize them.
[0059] The dynamic rule learning and optimization module learns and optimizes adaptation rules, preprocessing rules, and control strategies autonomously based on the quality assessment results of the quality assessment and state awareness modules and the operation data of each module.
[0060] Please see Figure 2 The source system adaptation module includes:
[0061] The source system information parsing unit receives basic information from external source systems, including system name, system code, data storage type (database / message queue / interface), transmission protocol (HTTP / MQTT, etc.), network address, etc., performs standardized parsing on the information, and generates a source system feature dataset, which contains key information such as system type identifier and data transmission method identifier.
[0062] The adaptation rule matching unit, based on the source system feature dataset from the source system information parsing unit, matches the corresponding docking rules from the preset adaptation rule library. The adaptation rule library pre-stores docking parameter templates and adaptation logic for various types of databases (MySQL, Oracle, etc.), interfaces (RESTful, SOAP, etc.), message queues (Kafka, RabbitMQ, etc.), and protocols (MQTT, TCP, etc.). If no completely matching rule is found, a basic adaptation framework is automatically generated for users to supplement personalized parameters.
[0063] The interface / protocol adaptation unit, based on the docking rules or basic adaptation framework output by the adaptation rule matching unit, constructs a dedicated adaptation channel to achieve a stable connection with the source system. For database-type source systems, it adapts the corresponding driver and configures connection parameters; for interface-type source systems, it configures the interface path, request parameters, and authentication information; for message queue or protocol-type source systems, it configures parameters such as queue address, port, and protocol version, and completes connection testing.
[0064] Please see Figure 3 The data perception preprocessing module includes:
[0065] After receiving the adaptation channel output by the source system adaptation module, the data acquisition unit selects the corresponding acquisition method based on the source system type obtained by parsing the source system information, including offline acquisition (full / incremental database acquisition, timed interface acquisition) and real-time acquisition (database change capture, real-time protocol transmission, message queue subscription), starts the acquisition task, collects data from multiple types of source systems after being connected by the source system adaptation module, and temporarily stores the collected data in the buffer.
[0066] The format parsing unit, based on the data collected in the buffer of the data acquisition unit, identifies the data format type (such as JSON, database table, binary, etc.), calls the corresponding parsing algorithm to perform structured parsing of the data, extracts field information, data type and business meaning from the data, and generates structured data;
[0067] The data cleaning unit optimizes the structured data in the format parsing unit based on preset cleaning rules. This includes outlier filtering (removing data that exceeds a reasonable range, such as data where the customer's age is greater than 150 years old), null value handling (marking null values in key fields or filling them based on business rules), and deduplication (de-duplication based on primary keys or key fields and retaining unique data).
[0068] The encoding standardization unit calls a preset encoding conversion rule library to perform unified encoding processing on the data cleaned by the data cleaning unit; for example, it unifies the encoding of customer gender fields such as "1 / 0", "male / female", and "M / F" in different systems into the standard format of "male / female"; at the same time, it decrypts encrypted fields to ensure that the data meaning is consistent.
[0069] Please see Figure 4 The task management module includes:
[0070] The task configuration unit provides a visual configuration interface, allowing users to configure the adaptation channel requirements of the source system adaptation module (such as basic information (system name, task name, task code)) and the preprocessing requirements of the data perception preprocessing module (such as acquisition parameters (acquisition type, synchronization method, execution time), preprocessing rules (cleaning rules, encoding rules) and output parameters (output address, format)); it also supports the classification and management of tasks based on business domain and data theme.
[0071] The task scheduling unit receives task information output by the task configuration unit and adopts the corresponding scheduling strategy based on the task type (real-time / offline). Real-time tasks are scheduled immediately based on an event-triggered mechanism, while offline tasks are scheduled based on Cron expressions or timed rules. It also supports setting task priorities, with higher-priority tasks occupying resources first.
[0072] The task execution unit receives the scheduling strategy from the task scheduling unit, calls the corresponding units of the source system adaptation module and the data perception preprocessing module to execute the task, and provides real-time feedback on the task execution progress (such as collection progress, cleaning completion rate) and execution status (running, completed, failed).
[0073] The task status management unit receives status information from the task execution unit and classifies and manages task status; it supports task online / offline operations, displaying an offline button when the task status is online and an online button when the task status is offline; it marks abnormal tasks (such as data collection failure or connection interruption) and triggers alarms.
[0074] The task monitoring unit collects execution process data in the task execution unit in real time, including data collection volume, processing rate, synchronization frequency, number of errors, etc., and generates visualization charts such as transmission efficiency trend charts and synchronized data volume bar charts; it also supports jumping to the details page to view the full running data of the task.
[0075] Please see Figure 5 The security authentication module includes:
[0076] The user authentication unit uses a multi-factor authentication method of username + password + verification code to verify the identity of users logging into the system. It also supports encrypted storage of user information and generates a temporary access token after successful verification.
[0077] The permission allocation unit, after user authentication is completed by the user authentication unit, uses a role-based access control model to pre-define multiple roles (administrator, operator, viewer, etc.) and assigns differentiated permissions to different roles, including source system configuration permissions, task configuration permissions, data viewing permissions, log auditing permissions, etc.; it also supports custom role and permission fine-grained configuration.
[0078] The operation authentication unit, after completing the permission allocation by the permission allocation unit, receives user operation requests (such as configuring tasks, viewing data, and deleting tasks) in real time, verifies the validity of the user's access token and the scope of permissions; at the same time, it intercepts and records unauthorized operations, and performs secondary confirmation for sensitive operations (such as deleting tasks and modifying authentication information).
[0079] The security log auditing unit records all user operations after completing the authentication process of the operation authentication unit, including the operation module, operation type, operation name, operator, operation time, and operation result; it also records security events during task execution (such as permission interception and abnormal access); and supports log querying, filtering, and exporting to preserve audit traces.
[0080] Please see Figure 6 The quality assessment and status awareness module includes:
[0081] The data quality acquisition unit collects data output from the data perception and preprocessing module in real time and extracts key data quality indicators, including completeness (field non-empty rate), accuracy (data format compliance rate), and consistency (encoding uniformity rate).
[0082] The data quality assessment unit evaluates the quality indicators of the data collected by the data acquisition unit based on preset quality thresholds, generates a data quality assessment report, and marks unqualified data and their corresponding sources; it also supports quality anomaly alarms to remind users to handle them in a timely manner.
[0083] The system operation status acquisition unit collects real-time operation data of each module, including server CPU utilization, memory utilization, disk utilization, interface response time, task execution time, and number of real-time tasks executed.
[0084] The status visualization unit visualizes the data quality assessment results from the data quality assessment unit and the operational status data from the system operation status acquisition unit through dashboards, trend charts, pie charts, and other formats. It also supports filtering and viewing by system, business domain, and time range, and triggers alarms for operational indicators that exceed thresholds.
[0085] Please see Figure 7 The dynamic rule learning and optimization module includes:
[0086] The learning data acquisition unit collects the adaptation records of the source system adaptation module, the processing logs of the data perception preprocessing module, the evaluation results of the quality assessment and status perception module, and the fault records of the task control module in real time. It extracts key data, including the reasons for adaptation failure, the types of preprocessing anomalies, the quality non-compliance items, and the causes of task failures, to form a learning dataset.
[0087] The rule analysis and optimization unit calls machine learning algorithms to perform in-depth analysis on the learning dataset of the learning data acquisition unit to explore the potential for rule optimization; for adaptation rules, it identifies the characteristics of source systems that failed to match and generates new adaptation parameter templates; for preprocessing rules, it counts the frequency of anomalies in cleaning and coding standardization, and optimizes filtering thresholds, coding mapping relationships, etc.; for task management rules, it analyzes the failure patterns of tasks and adjusts scheduling priorities and monitoring alarm thresholds.
[0088] The rule verification and update unit generates candidate solutions from the rules optimized by the rule analysis and optimization unit. These solutions are then verified in a simulated environment, and the data status before and after optimization (such as adaptation success rate, data quality compliance rate, and task operation stability) is compared. After successful verification, the solutions are automatically synchronized to the rule base of the corresponding module (such as the adaptation rule base of the source system adaptation module and the encoding conversion rule base of the data perception preprocessing module). The rule update log is recorded, and rollback operations are supported. For candidate solutions that fail verification, the problem points are marked and fed back to the learning data acquisition unit for re-analysis after data supplementation.
[0089] The workflow of the brain perception and cognition system described in this invention is as follows:
[0090] Step 1: The source system adaptation module establishes a stable connection with various source systems through information parsing, rule matching, and interface / protocol adaptation;
[0091] Step 2: The task configuration unit of the task management module configures the collection and preprocessing tasks based on business requirements, and the scheduling unit starts the tasks according to the configuration;
[0092] Step 3: The data acquisition unit of the data perception preprocessing module collects data based on the adaptation parameters, and outputs high-quality data after format parsing, cleaning and encoding standardization;
[0093] Step 4: The security authentication module performs identity authentication, permission verification, and log auditing for the entire operation process to ensure data and operational security;
[0094] Step 5: The quality assessment and status awareness module collects data quality and system operation data in real time, evaluates and visualizes the data, and achieves full-process awareness.
[0095] Step 6: The dynamic rule learning and optimization module collects the running data and quality assessment results of each module, analyzes and optimizes them to generate new rules, and after verification, synchronizes them to the corresponding module to complete the system's adaptive iteration.
[0096] Although the present invention has been described above with reference to embodiments, various modifications can be made and components can be replaced with equivalents without departing from the scope of the invention. In particular, as long as there is no structural conflict, the features in the disclosed embodiments can be combined with each other in any manner. The lack of an exhaustive description of these combinations in this specification is merely for the sake of brevity and resource conservation. Therefore, the present invention is not limited to the specific embodiments disclosed herein, but includes all technical solutions falling within the scope of the claims.
Claims
1. A brain-based perceptual and cognitive system, characterized in that, include: The source system adaptation module parses source system information, matches docking rules, and builds an adaptation channel; The data perception preprocessing module, based on the adaptation channel of the source system adaptation module, collects multi-type source system data after it has been connected to the source system adaptation module, and outputs structured data after parsing, cleaning, encoding and standardization preprocessing. The task management module configures tasks based on the adaptation channel requirements of the source system adaptation module and the preprocessing requirements of the data perception preprocessing module. It also adopts and executes tasks based on the corresponding scheduling strategy, monitors the data during task execution, and outputs management data. The security authentication module, based on the control data from the task management module, performs full-process security control over system operations, including identity authentication, permission allocation, and authentication auditing. The quality assessment and status awareness module collects structured data from the data awareness preprocessing module and operational data from the source system adaptation module, data awareness preprocessing module, task control module, and security authentication module. It assesses data quality and system status and displays the results visually. At the same time, it outputs the quality assessment results and operational data of each module. The dynamic rule learning and optimization module, based on the quality assessment results of the quality assessment and state awareness modules and the operation data of each module, autonomously learns and optimizes adaptation rules, preprocessing rules and control strategies. The data sensing preprocessing module includes: After receiving the adaptation channel output by the source system adaptation module, the data acquisition unit selects the corresponding acquisition method based on the source system type obtained by parsing the source system information, and temporarily stores the acquired data in the buffer. The format parsing unit identifies the data format type based on the data collected in the buffer of the data acquisition unit, calls the corresponding parsing algorithm to perform structured parsing of the data, and generates structured data. The data cleaning unit optimizes the quality of the structured data from the format parsing unit based on preset cleaning rules. The encoding standardization unit calls a preset encoding conversion rule library to perform unified encoding processing on the data after the data cleaning unit has optimized its quality.
2. The brain-sensory cognitive system according to claim 1, characterized in that, The source system adaptation module includes: The source system information parsing unit receives basic information from external source systems, performs standardized parsing of the information, and generates a source system feature dataset. The adaptation rule matching unit matches the corresponding docking rules from the preset adaptation rule library based on the source system feature dataset of the source system information parsing unit; if no completely consistent rule is matched, a basic adaptation framework is automatically generated. The interface / protocol adaptation unit constructs a dedicated adaptation channel based on the docking rules or basic adaptation framework output by the adaptation rule matching unit.
3. The brain-sensory cognitive system according to claim 1, characterized in that, The task management module includes: The task configuration unit provides a visual configuration interface, which supports users in configuring the adaptation channel requirements of the source system adaptation module and the preprocessing requirements of the data perception preprocessing module to complete the task configuration and output task information. The task scheduling unit receives task information output by the task configuration unit and adopts the corresponding scheduling strategy based on the task type. The task execution unit receives the scheduling strategy from the task scheduling unit, calls the source system adaptation module and the data perception preprocessing module to execute the task, and provides real-time feedback on the task execution progress and status. The task status management unit receives the task execution progress and execution status feedback from the task execution unit, and classifies and manages the task status. The task monitoring unit collects the task execution progress and status feedback from the task execution unit in real time and generates visual charts.
4. The brain-sensory cognitive system according to claim 1, characterized in that, The security authentication module includes: The user authentication unit uses a multi-factor authentication method of username + password + verification code to verify the identity of users logging into the system; The permission allocation unit, after the user identity authentication unit completes the identity authentication, uses a role-based access control model to pre-set multiple types of roles and, in conjunction with the management data from the task management module, assigns differentiated permissions to different roles. The operation authentication unit, after completing the differentiated permission allocation by the permission allocation unit, receives user operation requests in real time, verifies the validity of the user access token and the scope of permissions, and performs authentication operations. The security log auditing unit records all user operations after completing the authentication process of the operation authentication unit.
5. The brain-sensory cognitive system according to claim 1, characterized in that, The quality assessment and status awareness module includes: The data quality acquisition unit collects structured data output from the data perception and preprocessing module in real time and extracts key data quality indicators. The data quality assessment unit evaluates the key data quality indicators of the data acquisition unit based on preset quality thresholds, generates a data quality assessment report, and outputs the quality assessment results. The system operation status acquisition unit collects real-time operation data from the source system adaptation module, data perception and preprocessing module, task control module, and security authentication module. The status visualization unit visualizes the quality assessment results of the data quality assessment unit and the operating data of each module of the system operation status acquisition unit.
6. The brain-sensory cognitive system according to claim 1, characterized in that, The dynamic rule learning and optimization module includes: The learning data acquisition unit collects the adaptation channel records of the source system adaptation module, the preprocessing logs of the data perception preprocessing module, the quality assessment results of the quality assessment and status perception module, and the control data of the task control module in real time, extracts key data, and forms a learning dataset. The rule analysis and optimization unit calls machine learning algorithms to perform in-depth analysis on the learning dataset of the learning data collection unit, explore the potential for rule optimization, and output the optimized rules. The rule verification and update unit generates candidate solutions from the optimized rules of the rule analysis and optimization unit, verifies them in a simulated environment, compares the data status before and after optimization, and automatically synchronizes them to the rule base of the corresponding module after verification.
