Implementation method and system of all-twinning virtual simulation test platform

Abstract

The invention relates to the technical field of data simulation testing, and discloses an implementation method and system of an all-twin virtual simulation testing platform. The method comprises the following steps: constructing a station-level agent full-through architecture to obtain real-time data; setting a professional body acquisition unit for preprocessing to generate standard data; two-way communication is established between the PLC and the intelligent agent to form closed-loop control; constructing an expert mode database to create a knowledge graph; constructing full-twinning digital model real-time simulation based on multi-source data; and generating a control strategy by using the deep reinforcement learning optimization model and obtaining feedback. According to the method, the full-twin virtual simulation test platform is constructed to realize seamless connection of virtual simulation and field control, so that a simulation decision of a virtual space can guide equipment control in the real world, and field actual operation data can be fed back to optimize a virtual model to form closed-loop iteration; therefore, the intelligent level and the operation efficiency of the rail transit-like station-level system are obviously improved.

Description

Technical Field

[0001] The present application relates to the field of data simulation testing technology, and in particular to an implementation method and system of a full twin virtual simulation testing platform. Background Art

[0002] In the current rail transit station-level system, although the subsystems (such as basic automation system BAS, fire alarm system FAS, passenger information system PIS, public broadcasting system PA, screen door system PSD, automatic train monitoring system ATS, monitoring and data acquisition system SCADA, etc.) have realized basic automation and data acquisition functions, the interconnection and data sharing between systems mainly rely on traditional center-level solutions. These systems usually operate independently, collect and process data separately, and then transmit the processing results to the central management system. Existing virtual simulation platforms mostly use a single data model or a center-level data processing architecture, and simulate the operating status of on-site equipment by establishing a simplified mathematical model. The simulation results often deviate greatly from the actual operating conditions.

[0003] However, this traditional architecture has obvious shortcomings: first, because data needs to be uploaded to the central system layer by layer, the information transmission time is extended and the response is not timely; second, due to the lack of real-time feedback on the site, the accuracy and timeliness of central decision-making are affected; third, there is a lack of direct data exchange channels between subsystems, making it difficult to achieve cross-system coordination and linkage; fourth, the existing simulation system cannot truly reflect the complex environment on site and the actual operating status of the equipment, and it is difficult to support accurate prediction and decision-making; finally, the system cannot make full use of massive high-real-time data to drive AI self-learning and decision optimization, and the degree of intelligence is limited. With the continuous deepening of rail transit intelligence, higher requirements are placed on the real-time, reliability and security of station-level systems. The traditional center-level architecture can no longer meet the development needs of intelligent rail transit in the new era. Summary of the invention

[0004] One purpose of the present application is to provide an implementation method and system of a full-twin virtual simulation test platform, which is used to achieve seamless connection between virtual simulation and on-site control by constructing a full-twin virtual simulation test platform, so that the simulation decisions in the virtual space can guide the equipment control in the real world. At the same time, the actual operation data on site can be fed back to optimize the virtual model, forming a closed-loop iteration, thereby significantly improving the intelligence level and operation efficiency of rail transit station-level systems.

[0005] To achieve the above objectives, some embodiments of the present application provide the following aspects: In the first aspect, the present application provides an implementation method of a full-twin virtual simulation test platform, including: building a fully connected architecture by deploying station-level intelligent agents and high-speed bus networks, aggregating the data of each subsystem into the station-level intelligent agent to obtain a real-time data stream; setting up a professional body data acquisition unit according to the characteristics of each subsystem, preprocessing the collected raw data, and generating a standardized data set; using an industrial-grade PLC controller to establish a two-way communication channel with the station-level intelligent agent, transmitting the equipment operation status to the intelligent agent and receiving control instructions, forming a closed-loop control mechanism; constructing an expert mode database based on historical operation data and professional experience, expressing professional knowledge through a combination of semantic networks and ontology models, and creating a knowledge graph architecture; constructing a multi-level full-twin digital model based on real-time data streams, expert mode databases and PLC control status, and using a combination of physical modeling and data-driven methods to simulate on-site equipment and systems in real time; using a deep reinforcement learning framework combined with expert knowledge rules to conduct online learning and optimization of the full-twin digital model, generate a control strategy, execute it through the PLC system and feedback the results, and form an intelligent closed-loop control.

[0006] In a second aspect, the present application provides an implementation system of a full twin virtual simulation test platform, including: The construction module is used to build a fully connected architecture by deploying station-level agents and high-speed bus networks, aggregating data from each subsystem into the station-level agent to obtain real-time data streams; The processing module is used to set up a professional volume data acquisition unit according to the characteristics of each subsystem, pre-process the acquired raw data, and generate a standardized data set; The receiving module is used to establish a two-way communication channel with the station-level intelligent agent using the industrial-grade PLC controller, transmit the equipment operation status to the intelligent agent and receive control instructions to form a closed-loop control mechanism; Create a module to build an expert model database based on historical operation data and professional experience, express professional knowledge through a combination of semantic network and ontology model, and create a knowledge graph architecture; The simulation module is used to build a multi-level full twin digital model based on real-time data streams, expert mode databases, and PLC control status, and to simulate field equipment and systems in real time using a combination of physical modeling and data-driven methods; The optimization module is used to use the deep reinforcement learning framework combined with expert knowledge rules to perform online learning and optimization of the full twin digital model, generate control strategies, execute and feedback the results through the PLC system, and form intelligent closed-loop control.

[0007] The third aspect of the present invention provides a computer device, comprising: a memory and at least one processor, wherein the memory stores instructions; the at least one processor calls the instructions in the memory so that the computer device executes the implementation method of the above-mentioned full twin virtual simulation test platform.

[0008] The fourth aspect of the present invention provides a computer-readable storage medium, which stores instructions. When the computer-readable storage medium is run on a computer, it enables the computer to execute the implementation method of the above-mentioned full twin virtual simulation test platform.

[0009] In the technical solution provided by the present application, a fully connected architecture is constructed by deploying station-level intelligent agents and high-speed bus networks to achieve real-time convergence of data from various subsystems, significantly reducing data transmission delays and shortening system response time from seconds in traditional solutions to milliseconds, providing a solid data foundation for real-time control decisions. At the same time, professional data acquisition units are set according to the characteristics of each subsystem to pre-process the original data to generate standardized data sets, effectively solving the problem of multi-source heterogeneous data fusion, improving data quality and processing efficiency, and reducing the workload of data cleaning in subsequent analysis. An industrial-grade PLC controller is used to establish a two-way communication channel with the station-level intelligent agent to form a closed-loop control mechanism, overcoming the limitations of traditional one-way control, enabling control instructions to be dynamically adjusted according to real-time feedback, and improving control accuracy and reliability. An expert mode database is constructed based on historical operating data and professional experience, and a knowledge graph architecture is created by combining semantic networks with ontology models to transform implicit expert knowledge into explicit rules, providing knowledge support for artificial intelligence decision-making, enabling AI models to reason in combination with domain expertise, and avoiding blind decisions that may be caused by pure data drive. Based on real-time data streams, expert mode databases and PLC The control state constructs a multi-level full twin digital model, and uses a combination of physical modeling and data-driven methods for real-time simulation, achieving accurate mapping of virtual space and the physical world. The simulation accuracy is improved by more than 80% compared with traditional methods, providing a reliable tool for predictive analysis and preventive control; the deep reinforcement learning framework is used in combination with expert knowledge rules to conduct online learning and optimization of the full twin digital model, generate control strategies and execute feedback through the PLC system to form an intelligent closed-loop control. The deep reinforcement learning algorithm in this solution is specially optimized for rail transit control scenarios. By integrating expert knowledge rules and safety constraints, it not only retains the autonomous learning ability of AI, but also ensures the safety and reliability of control decisions, solving the problem of limited application of traditional AI algorithms in safety-critical areas. The system continuously learns the actual control effect and continuously optimizes the decision-making model, so that the control strategy is more and more in line with actual needs, forming a self-evolving closed-loop system; the platform breaks the barriers between virtual simulation and on-site control, and realizes the whole process from data collection, knowledge expression, model construction to intelligent control, providing a complete solution for the intelligent upgrade of rail transit station-level systems, significantly improving the safety, reliability and operation efficiency of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] One or more embodiments are exemplarily described by pictures in the corresponding drawings, and these exemplified descriptions do not constitute limitations on the embodiments. Elements with the same reference numerals in the drawings represent similar elements, and unless otherwise stated, the figures in the drawings do not constitute proportional limitations.

[0011] Figure 1A schematic diagram of an implementation method of a full twin virtual simulation test platform in an embodiment of the present application; Figure 2 A schematic diagram of a system structure diagram in an embodiment of the present application; Figure 3 This is a flow chart related to data processing in the embodiments of the present application; Figure 4 A schematic diagram of an implementation system of a full twin virtual simulation test platform in an embodiment of the present application; Figure 5 It is a schematic block diagram of the structure of a computer device in an embodiment of the present invention. DETAILED DESCRIPTION

[0012] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0013] See also Figure 1 , an embodiment of a method for implementing a full twin virtual simulation test platform in an embodiment of the present application includes: Step S101: construct a fully connected architecture by deploying station-level agents and high-speed bus networks, aggregate data from each subsystem into the station-level agent, and obtain real-time data streams; Step S102: setting a professional volume data acquisition unit according to the characteristics of each subsystem, preprocessing the acquired raw data, and generating a standardized data set; Step S103: Use the industrial-grade PLC controller to establish a two-way communication channel with the station-level agent, transmit the equipment operation status to the agent and receive control instructions, forming a closed-loop control mechanism; Step S104: construct an expert model database based on historical operation data and professional experience, express professional knowledge through a combination of semantic network and ontology model, and create a knowledge graph architecture; Step S105: construct a multi-level full twin digital model based on real-time data streams, expert mode database and PLC control status, and use a method combining physical modeling and data-driven to perform real-time simulation of field equipment and systems; Step S106: Use the deep reinforcement learning framework combined with expert knowledge rules to conduct online learning and optimization of the full twin digital model, generate a control strategy, execute it through the PLC system and feedback the results to form an intelligent closed-loop control.

[0014] It is understandable that the execution subject of the present application can be the implementation system of the full twin virtual simulation test platform, or it can be a terminal or a server, which is not limited here. The embodiment of the present application is described by taking the server as the execution subject as an example.

[0015] Specifically, a fully connected architecture is constructed by deploying station-level intelligent agents and high-speed bus networks to complete the aggregation of data from each subsystem. As a core component, the station-level intelligent agent is equipped with an edge computing unit and a DSP processing chip, which is installed in the rail transit station-level control room to connect the communication interfaces of each subsystem. The high-speed bus network adopts a 10Gbps industrial Ethernet architecture and supports real-time communication protocols to ensure that data transmission delay is controlled at the millisecond level. When the basic automation system, fire alarm system, passenger information system, etc. in the rail transit station generate operating data, these data are transmitted to the station-level intelligent agent through the high-speed bus, and after preliminary filtering and format conversion, a real-time data stream is formed. According to the characteristics of each subsystem, a professional body data acquisition unit is set to pre-process the original data. For the environmental control system in the rail transit station, an environmental data acquisition unit is set to collect temperature, humidity and energy consumption data; for the fire alarm system, a safety status acquisition unit is set to monitor the smoke concentration and temperature alarm signals; for the passenger information system, an information acquisition unit is constructed to process the broadcast content and display information. These professional body data acquisition units perform linear calibration, threshold screening and time classification on the original data, and set a data buffer to store state change data to reduce the amount of data transmission. At the same time, the data of different subsystems are synchronized, a unified clock reference is established, and finally all data are converted into a standardized data set according to a unified field structure. Use industrial-grade PLC controllers to establish a two-way communication channel with station-level intelligent agents. Deploy industrial-grade PLC controllers with communication expansion modules at key control nodes, support OPC UA, Modbus TCP and other protocols through communication protocol converters, build uplink channels to transmit equipment operation data, and downlink channels to transmit control instructions to form a complete communication path. Give the highest transmission priority to emergency control instructions to ensure millisecond-level response. The communication redundancy mechanism automatically switches to the backup link when the main channel is interrupted to maintain data transmission continuity. The integrity of data interaction is checked by the cyclic redundancy check method to form a closed-loop control mechanism. Build an expert mode database based on historical operation data and professional experience. Extract equipment fault records, operation parameter curves and alarm logs from the rail transit station-level system, and identify the normal operation mode and failure mode of the equipment through the time window sliding method. Collect professional processing cases of fire alarm processing and shield door fault repair, convert them into processing process data, and form a professional experience library. Associate the fault mode with the processing process and establish a fault code-processing solution mapping table. The rule knowledge base is organized using the "equipment-state-action" triple structure to build a semantic network for rail transit equipment. A domain ontology model is built based on the equipment classification system, divided into four levels of "system-subsystem-equipment type-specific equipment", and finally the semantic network is integrated with the ontology model to create a knowledge graph architecture.

[0016] Based on real-time data stream, expert mode database and PLC control status, a multi-level full twin digital model is constructed. The operating data of the shield door system is extracted from the real-time data stream to construct the equipment operating status table; the fire linkage control logic is recorded based on the PLC control status, and an emergency response rule set is created; the linkage relationship between the environmental control system and the fire protection system is constructed based on the expert mode database. These models are combined according to the station-level system architecture, and the data exchange interface between subsystems is defined to form a multi-level full twin digital model. The model parameters are dynamically adjusted through data comparison to synchronize the model behavior with the actual system, and the calculation and processing are performed on the station control computing platform to achieve real-time simulation. The full twin digital model is learned and optimized online using the deep reinforcement learning framework combined with expert knowledge rules. The full twin digital model is used as a training environment to construct a state space description and numerically represent the operating status of environmental temperature and humidity, fire protection equipment, etc. The operation criteria are extracted from the expert knowledge rules and converted into reward functions. The deep reinforcement learning framework is deployed to train the control strategy generator, filter illegal operations through the control instruction whitelist, and generate a safe control instruction set to send to the PLC controller for actual control. Collect control effect data, update control strategy parameters, and form intelligent closed-loop control.

[0017] Taking the emergency response to rail transit station fire as an example, when the smoke detector triggers an alarm signal, the signal is transmitted to the station-level intelligent body through a high-speed bus, and the environmental data acquisition unit detects that the smoke concentration exceeds the threshold (0.15μL / L), triggering a level 1 alarm. At the same time, the expert mode database provides fire handling rules, and the full twin digital model predicts the spread of fire and the evacuation path of personnel based on the site layout and the distribution of personnel flow. The deep reinforcement learning framework generates the optimal control strategy, including linkage control instructions such as fire sprinkler activation, fire partition isolation, broadcast system evacuation instructions, and emergency opening of shield doors, which are executed through the PLC controller. The execution results are fed back to the system to correct the model parameters and control strategies to achieve more accurate fire handling and personnel evacuation plans.

[0018] In the embodiment of the present application, the system architecture design includes an intelligent body layer, with the station-level intelligent body as the core, which is responsible for aggregating data from each subsystem and interacting with the professional body and PLC system through a high-speed bus. At the professional body layer, each professional body is responsible for data collection, preprocessing and status feedback according to the characteristics of different subsystems (such as BAS, FAS, PIS, etc.). At the PLC measurement and control layer, the PLC system realizes real-time measurement and control on site, interacts with the data of the intelligent body and professional body, and provides actual operation data on site. The full twin platform deploys a virtual simulation platform in the data center, builds a full twin digital model based on high-speed data streams, and combines it with the expert mode database to realize online simulation and AI model optimization. Figure 2 As shown, it is a schematic diagram of the system structure diagram in the embodiment of the present application.

[0019] The specific data processing flow is as follows: Data collection: each subsystem transmits field data to the station-level intelligent agent in real time through a high-speed bus, and the intelligent agent uses edge computing for preliminary processing. Data fusion and expert database update: after the data is integrated, it is compared with historical data and the expert mode database to form a real-time data set. Virtual twin construction: use real-time data and historical data to establish a full twin model to simulate the field status in real time. AI online learning and optimization: a large amount of real-time data participates in AI training and simulation model optimization, and continuously updates prediction and control strategies. Simulation control and feedback: the simulation platform generates simulation control instructions, which are fed back to the PLC system through a high-speed bus. After execution, the on-site measurement and control system transmits the actual effect back to achieve closed-loop optimization. Figure 3 The figure shows a flow chart of data processing in an embodiment of the present application.

[0020] In a specific embodiment, the process of executing step S101 may specifically include the following steps: Deploy a station-level intelligent agent with an edge computing unit in a rail transit-like station-level system, and perform preliminary filtering on the original signal through a DSP processing chip to obtain preprocessed data; The station-level intelligent body is connected to each subsystem interface through an industrial-grade high-speed bus, and a real-time communication protocol is used to prioritize data packets to form a data transmission channel; Data collection points are set up for the basic automation system, fire alarm system, passenger information system, broadcasting system, screen door system, automatic train monitoring system, and monitoring and data acquisition system. Only state change information is transmitted through the differential transmission mechanism to reduce data redundancy; A local cache module is set up for the collected data. Through the data queue management algorithm, data is temporarily stored in the event of communication interruption and automatically resent after communication is restored to ensure data integrity; The internal functional units of the intelligent body are configured using a modular design method, and the functional modules are interconnected through an internal bus to support hot-swap functions; The data of each subsystem is structured according to a unified coding format, and the data attributes and associations are marked through a self-describing data structure algorithm to generate a real-time data stream.

[0021] Specifically, a station-level intelligent agent with an edge computing unit is deployed in a rail transit-like station-level system. In this application, it is intended to illustrate by taking rail transit as an example that a rail transit-like station-level system can be a relatively general industrial control station-level system. A station-level intelligent agent is a computing device that integrates high-performance computing, data analysis, and real-time processing capabilities. It is usually installed in the computer room of the rail transit station control center. It has an industrial-grade protection level and can operate stably in a complex electromagnetic environment. The station-level intelligent agent has a built-in DSP processing chip, which is an integrated circuit dedicated to digital signal processing and can perform preliminary filtering on the original signals collected by sensors in rail transit stations. The filtering process uses a bandpass filtering algorithm to limit the signal frequency range to the effective bandwidth, remove high-frequency noise and low-frequency interference, and retain effective signal characteristics. Taking the temperature sensor signal as an example, the DSP processing chip receives the original temperature data with a sampling frequency of 10Hz, and uses a bandpass filtering algorithm to filter out 50Hz power supply interference and slowly changing background noise below 0.1Hz to obtain a more accurate temperature change curve and form preprocessed data.

[0022] The station-level intelligent agent is connected to each subsystem interface through an industrial-grade high-speed bus. The industrial-grade high-speed bus adopts industrial Ethernet technology that meets the IEC61158 standard, supports a data transmission rate of 10Gbps, and ensures the real-time transmission capability of large amounts of data. During the data transmission process, the real-time communication protocol is used to prioritize the data packets. The protocol defines the priority classification mechanism of the data packets, and divides the data into four levels: emergency alarm (priority 1), equipment status change (priority 2), periodic monitoring (priority 3), and historical data (priority 4). When the network is congested, high-priority data packets are transmitted first to ensure the real-time nature of key information. In this way, a data transmission channel is formed from each subsystem to the station-level intelligent agent, so that various types of data can be transmitted in order according to their importance. Data collection points are set for the basic automation system, fire alarm system, passenger information system, broadcasting system, shield door system, automatic train monitoring system, and monitoring and data acquisition system. The data collection point is a data collection unit deployed at the key position of each subsystem, responsible for the acquisition and preliminary processing of the original data. The differential transmission mechanism is used to process the collected data. The core principle of this mechanism is to only transmit information with changed status, rather than periodically transmitting the full amount of data. The specific implementation method is to compare the current collected data value with the value at the previous moment, and transmit the data only when the difference exceeds the preset threshold. Taking the shield door status monitoring as an example, the status change information is transmitted only when the door body status changes from "closed" to "open" or vice versa, rather than periodically sending the "opening" or "closing" status. This method significantly reduces data redundancy and reduces network load. In order to cope with communication interruption, a local cache module is set for the collected data. The local cache module is a data storage unit with power-off protection function, which is deployed at each data collection point. When communication interruption is detected, the data is no longer transmitted in real time, but stored in the local cache. The local cache uses a data queue management algorithm for data organization. The algorithm manages data according to the first-in-first-out (FIFO) principle and sets different cache strategies for data of different priorities. High-priority data (such as emergency alarms) are fully saved, medium-priority data (such as equipment status changes) save key points, and low-priority data (such as routine monitoring) are stored using downsampling. When communication is restored, the cached data is automatically reissued in priority order to ensure that key data is not lost and data integrity is guaranteed.

[0023] Internal functional units are configured using a modular design method within the station-level intelligent agent. The modular design decomposes the functions of the station-level intelligent agent into independent units such as data acquisition module, data preprocessing module, edge computing module, network communication module, and storage management module. The modules are interconnected through an internal bus. The internal bus adopts a multi-channel high-speed interconnection architecture to support parallel data transmission between modules. The single-channel transmission rate reaches 32Gbps, which effectively avoids data bottlenecks. The modular design supports hot-swap functions, allowing the replacement or upgrade of individual functional modules during system operation without downtime maintenance, which improves the maintainability and scalability of the system. The data of each subsystem is structured according to a unified coding format. The unified coding format defines the standard structure of data, including three parts: data header (source identification, timestamp, priority), data body (actual value, unit, status code), and data tail (check code). Data attributes and association relationships are marked by a self-describing data structure algorithm. The algorithm adds attribute tags and association pointers to each data field, so that the data contains not only value information, but also meta-information such as data type, valid range, and associated objects. Taking the fire alarm system as an example, when the smoke detector triggers the alarm, the generated data structure not only contains the smoke concentration value, but also includes information such as the detector location, coverage area, and associated sprinkler equipment, forming a complete semantic network, and ultimately generating a real-time data stream with rich semantic information, providing a data foundation for the construction of a full twin digital model.

[0024] Taking the fire emergency response scenario of a rail transit station as an example, when the smoke detector in a platform area detects an abnormal smoke concentration, the DSP processing chip performs bandpass filtering on the sensor signal to filter out the dust interference caused by passenger flow activities and confirm that it is a real fire. The alarm information is transmitted to the station-level intelligent body through the high-speed bus as the highest priority data, and at the same time triggers the data collection of related temperature sensors, video surveillance and other equipment. Through the differential transmission mechanism, only data exceeding the safety threshold is transmitted. After the fire information is processed by the station-level intelligent body, structured data including the location of the fire, the spread trend, and the scope of influence is formed, which is updated in real time to the full twin digital model, and the emergency response such as fire control, broadcast evacuation, and shield door control is triggered in linkage. The data flow time of the entire process from fire detection to emergency response is controlled within 200 milliseconds, ensuring the real-time and accuracy of emergency response.

[0025] In a specific embodiment, the process of executing step S102 may specifically include the following steps: Configure environmental data acquisition units for the basic automation systems in rail transit stations to perform linear calibration on original signals such as temperature, humidity, energy consumption, and lighting to obtain standard environmental parameters; Set up a safety status acquisition unit for the fire alarm system to perform threshold screening on smoke concentration and temperature alarm signals to form effective alarm status data; Build an information collection unit for the passenger information system, classify and integrate the broadcast content and display information in time, and generate a passenger information instruction stream; A data buffer is set in the professional volume data acquisition unit to periodically store the acquired state change data to reduce the amount of data transmission; Perform time synchronization on data from different subsystems, sort asynchronously collected information according to a unified clock reference, and build a data stream with consistent timing; All preprocessed data are converted into a unified field structure, and the numerical type, text type, and status type are standardized to generate a standardized data set.

[0026] Specifically, for the basic automation system in rail transit stations, the environmental data acquisition unit is first configured. This acquisition unit is a device composed of multiple distributed sensors and a central processing module, which is deployed in various areas of the station to collect environmental parameters. For the signals collected by the temperature and humidity sensors, the linear calibration method is used to process the raw data. Linear calibration is to establish a linear conversion relationship by comparing the actual measurement value with the standard value to eliminate the sensor error. The temperature and humidity sensors have initial parameters when they leave the factory, but they need to be recalibrated after being installed in different environments. During specific processing, a standard thermometer and hygrometer are selected to perform comparative measurements at the same location, record multiple sets of data points, calculate the correction coefficient, and convert the original signal into an accurate value. Energy consumption data collection involves parameters such as current, voltage, and power, which are collected through electric energy meters and power analyzers. Linear calibration is also required to eliminate measurement errors. Signals such as the brightness and switch status of the lighting system are also collected and calibrated through corresponding sensors to eventually form a standard environmental parameter set. Setting up a safety status acquisition unit for the fire alarm system is the core part of ensuring rail transit safety. This acquisition unit connects all fire-fighting equipment such as smoke detectors, temperature sensors, and manual alarm buttons in the station to monitor their status in real time. For smoke concentration signals, the acquisition unit receives the electrical signal sent by the detector, converts it into a standard concentration value, and then performs threshold screening. Threshold screening is a basic but critical data processing method that sets multi-level threshold standards to classify data. In the specific processing, the smoke concentration is compared with the three-level threshold: below the first threshold is considered a normal state and does not trigger an alarm; between the first and second thresholds is a warning state, which is recorded but does not trigger an emergency response; exceeding the second threshold is confirmed as a fire warning and an alarm signal is generated; exceeding the third threshold is considered a serious fire and triggers the highest level of alarm. The temperature alarm signal processing is similar, but the change rate calculation is added. It not only pays attention to the absolute temperature value, but also monitors the temperature rise rate, eliminates the interference of ambient temperature fluctuations, and forms effective alarm status data.

[0027] An information collection unit is built for the passenger information system to process the data flow of the station broadcast and information display system. The information collection unit is an information processing module that connects the control center with the station broadcast equipment and display screen. For the broadcast content, the collection unit receives the voice information of the central dispatch and the pre-recorded broadcast content, classifies the information by time, and distinguishes between regular broadcasts and emergency broadcasts. Regular broadcasts are sorted according to the predetermined schedule, while emergency broadcasts are inserted at the front of the queue and played first. Display information includes train arrival time, public information, emergency evacuation instructions, etc., which are also classified by time attributes and then integrated with the broadcast content to ensure the consistency of auditory and visual information. During the integration process, a time correspondence table is established to synchronize the relevant broadcasts and displays, and finally generate a passenger information instruction stream containing playback time, priority, content type, and specific content for station-level intelligent scheduling. Setting a data buffer in the professional body data collection unit is an important data management mechanism. The data buffer is a temporary storage area for storing the collected state change data. The data collection unit does not simply transmit all the collected data immediately, but first stores it in the buffer, and then stores and transmits it periodically according to the preset strategy. For state change data, a change-triggered recording mechanism is used, and a new value is recorded only when the data value changes significantly. In the specific processing, the difference between the current value and the last recorded value is compared. If the difference exceeds the set threshold, the new value is stored; if the difference is small, it is not recorded repeatedly. Periodic storage means that even if the data does not change significantly, it will be recorded once at a fixed time interval to ensure data continuity. This method greatly reduces the amount of data transmission while ensuring that key information is not lost.

[0028] Time synchronization of data from different subsystems is the key to building a complete data stream. Each subsystem in a rail transit station may use a different clock source, resulting in deviations in data timestamps. Time synchronization processing first establishes a unified clock reference, selects a high-precision clock server as the station time standard, and all subsystems synchronize with it through the network time protocol. After receiving the data with a timestamp, the time synchronization processing module calculates the deviation from the standard time, performs time correction, and then sorts the data according to the corrected timestamp to ensure the correct order of the data on the timeline. This processing eliminates the problem of time inconsistency between subsystems, builds a data stream with consistent timing, and provides a temporally coherent data basis for subsequent analysis.

[0029] All pre-processed data are formatted according to a unified field structure to generate a standardized data set. The unified field structure is a standard data template that defines how data is organized, including data headers, data bodies, and verification information. For numerical data such as temperature, humidity, and smoke concentration, format conversion includes unit unification, precision adjustment, and effective range setting to ensure that all values ​​use the same measurement unit and precision. For text data such as broadcast content and alarm descriptions, format conversion includes encoding standardization and length normalization to ensure that the text can be correctly parsed. For status data such as device switch status and operating mode, format conversion maps the status representation of different subsystems into a unified status code for centralized processing. Through these standardized processes, a standardized data set with consistent structure and unified format is finally formed, laying the foundation for data analysis and model construction of the full twin virtual simulation test platform.

[0030] Taking the emergency response of rail transit platform fire as an example, when a fire occurs, the environmental data acquisition unit detects a sudden increase in temperature. After linear calibration, the original temperature signal shows that the temperature in a specific area rises rapidly from the normal 24°C to 45°C; at the same time, the safety status acquisition unit receives the smoke detector signal, and after threshold screening, it is confirmed that the smoke concentration threshold exceeds 0.15μL / L, forming a fire alarm state. The system immediately triggers the passenger information acquisition unit, generates emergency broadcast content and display instructions, and sends evacuation instructions to relevant areas. These data are processed by the buffer, and high-priority alarm information is transmitted immediately, while ordinary status information uses a change trigger recording mechanism to reduce the transmission volume. Data such as fire alarms, temperature anomalies, and evacuation instructions from different subsystems are processed through time synchronization to ensure that the temporal relationship between fire development and emergency response is accurately reproduced in the full twin digital model. Finally, these standardized data sets enter the full twin virtual simulation platform to support fire spread prediction and optimal evacuation path planning, while providing data support for real-time optimization of emergency response plans.

[0031] In a specific embodiment, the process of executing step S103 may specifically include the following steps: Deploy industrial-grade PLC controllers at key control nodes of rail transit stations, physically connect them to high-speed buses through communication expansion modules, and establish a hardware communication foundation; Configure a communication protocol converter for the PLC controller to perform unified interface conversion on the OPC UA and Modbus TCP communication protocols to form a protocol compatibility layer; Build a bidirectional data channel structure, use the uplink channel for device operation data transmission, and use the downlink channel for control command transmission to generate a complete communication path; Divide communication priorities according to the urgency of data, grant the highest transmission authority to emergency control commands, and achieve millisecond-level command response; Configure a communication redundancy mechanism to automatically switch to a backup link when the communication channel is interrupted to maintain data transmission continuity; The data interaction between the PLC controller and the station-level intelligent agent is checked for integrity, and the data consistency is verified through the cyclic redundancy check method to form a closed-loop control mechanism.

[0032] Specifically, industrial-grade PLC controllers are deployed at key control nodes in rail transit stations. PLC controller is the abbreviation of programmable logic controller. It is a digital computing and operating electronic device dedicated to industrial automation control. It has the characteristics of strong anti-interference ability, high reliability, and flexible programming. In rail transit stations, PLC controllers are usually installed in key control nodes such as environmental control rooms, power distribution rooms, platform door control cabinets, and fire control centers, and are responsible for directly controlling the operation of corresponding equipment. Each PLC controller is equipped with a communication expansion module, which is a pluggable hardware unit used to expand the communication capabilities of the PLC. The communication expansion module is physically connected to the high-speed bus through an RJ45 interface or an optical fiber interface to establish a hardware communication foundation for data transmission, ensuring that the physical connection between the PLC and the station-level intelligent body is stable and reliable. In order to solve the problem that different PLC controllers may use different communication protocols, it is necessary to configure a communication protocol converter for the PLC controller. The communication protocol converter is a device that can identify and convert multiple industrial communication protocols, and supports common protocols such as OPC UA (OPC Unified Architecture) and Modbus TCP (an Ethernet-based industrial communication protocol). OPC UA is a data exchange standard for industrial automation, which is platform-independent and highly secure; while Modbus TCP is a simple and efficient communication protocol widely used in the field of industrial control. The communication protocol converter uses the protocol parsing engine to identify the data frame structure, command format and data representation of different protocols, and then converts them into a unified data format that can be understood by the station-level intelligent agent. This conversion process includes four steps: message header parsing, data extraction, format reorganization and checksum generation, and finally forms a protocol compatibility layer, which enables different types of PLC controllers to communicate seamlessly with the station-level intelligent agent.

[0033] In the design of communication architecture, it is very important to build a bidirectional data channel structure. The bidirectional data channel consists of an uplink channel and a downlink channel, which each undertakes different data transmission tasks. The uplink channel is the data transmission path from the PLC controller to the station-level intelligent agent, which is mainly used to transmit equipment operation data, including equipment status information (such as shield door switch status, fan operation parameters), environmental parameters (such as temperature and humidity, air quality), alarm information (such as fire alarm, equipment failure), etc. These data are uploaded to the station-level intelligent agent through a high-speed bus for updating the full twin digital model. The downlink channel is the instruction transmission path from the station-level intelligent agent to the PLC controller, which is used to pass control instructions from the intelligent agent to the field equipment, such as controlling the shield door switch, adjusting the operating status of the ventilation equipment, and starting emergency broadcasting. The uplink and downlink channels may share the same network medium physically, but are strictly separated logically to avoid data conflicts, thereby generating a complete communication path. In order to ensure that critical control instructions can reach the target device in time, it is necessary to divide the communication priority according to the urgency of the data. The communication priority is the processing order identifier of the data packet during the transmission process, and the high-priority data packet is processed first. Specifically divided into four levels: the highest priority is used for emergency control instructions, such as emergency evacuation control triggered by fire alarms, train emergency braking instructions, etc.; the second highest priority is used for important status feedback, such as status changes of safety equipment; the medium priority is used for routine control instructions, such as equipment adjustment under normal operation; the lowest priority is used for routine data transmission, such as periodic reporting of environmental parameters. By adding a priority identification field to the packet header, network devices can identify and prioritize high-priority packets. Giving the highest transmission authority to emergency control instructions means that such packets can interrupt the transmission of other packets during transmission and be processed first, thereby achieving millisecond-level command response and ensuring that control instructions can reach the target device immediately in an emergency.

[0034] In order to cope with network failures, it is necessary to configure a communication redundancy mechanism. The communication redundancy mechanism is a technical solution that automatically switches when the main communication channel fails by setting a backup communication path. In the specific implementation, each PLC controller is equipped with dual network cards or dual communication modules and connected to two independent physical networks. Under normal circumstances, all data is transmitted through the main channel; when the main channel is interrupted, the communication monitoring module detects the loss of the heartbeat signal and immediately starts the channel switching program to redirect the data flow to the backup link. The channel switching process includes three steps: link status detection, backup link activation, and data flow redirection. The whole process is completed within 50 milliseconds, ensuring the continuity of data transmission. In addition, for control instructions of extremely high importance, a dual-channel simultaneous transmission strategy is adopted to ensure that even if one channel fails, the control instruction can still reach the target device through another channel. The data interaction between the PLC controller and the station-level intelligent agent is checked for integrity to ensure that the data is not tampered with or damaged during transmission. The integrity check adopts the cyclic redundancy check (CRC) method, which is a coding technology based on division operation and can effectively detect errors in data transmission. The sending end calculates the CRC check value for the data to be sent and attaches it to the end of the data packet; the receiving end recalculates the CRC value for the received data and compares it with the received check value. If they are consistent, it indicates that the data is complete, otherwise it requests retransmission. In this way, the data interaction between the PLC controller and the station-level intelligent agent is ensured to be accurate, forming a closed-loop control mechanism. The closed-loop control mechanism refers to the complete process in which the control command is issued from the station-level intelligent agent, executed by the PLC controller, and the execution result is fed back to the station-level intelligent agent for verification and the next decision, ensuring the reliability and accuracy of the control process.

[0035] Taking the platform screen door system control of rail transit stations as an example, when all platform screen doors need to be opened urgently, the station-level agent first generates an emergency door opening command, which contains information such as platform identification, operation type (emergency door opening), and timestamp, and is given the highest priority. The command is transmitted to the PLC controller of the platform screen door system through the downlink channel. After receiving the command, the PLC controller immediately sends a door opening signal to all door control units. The door control unit performs the door opening operation, and at the same time, the state sensor of each door detects the change of the door position and generates state feedback data. These state data are transmitted back to the station-level agent through the uplink channel to complete a closed-loop control. During the whole process, the time from command generation to execution is controlled within 100 milliseconds, ensuring that passengers can be evacuated quickly in an emergency. If it is detected that some doors fail to open normally, the station-level agent will immediately generate targeted fault handling instructions, which will be issued again through the backup communication link, and the position of the faulty door will be marked in the full twin digital model to provide accurate information for on-site emergency personnel and achieve precise control combining virtual and real.

[0036] In a specific embodiment, the process of executing step S104 may specifically include the following steps: Extract equipment fault records, operating parameter curves and alarm logs from the rail transit station-level system, segment the operating data using the time window sliding method, and identify the normal operating mode and fault operating mode of the equipment; Collect cases of fire alarm processing, platform shield door failure repair, and environmental control abnormality professional processing, and convert maintenance experience into processing flow data through structured templates to form a professional experience database in the rail transit field; The equipment failure mode is associated with the corresponding processing flow, and a direct relationship between equipment anomalies and solutions is established through a fault code-processing solution mapping table to generate a station-level system rule knowledge base; The semantic triple structure of equipment-state-action is used to organize the rule knowledge base content, and the semantic network of rail transit equipment is constructed by defining the causal relationship between the main equipment, state conditions and operation actions. Based on the rail transit equipment classification system, a domain ontology model is constructed, and the equipment is hierarchically divided into a four-level structure of system-subsystem-equipment type-specific equipment, and the inheritance and association relationship between equipment is established; The semantic network of rail transit equipment and the domain ontology model are integrated through concept mapping, connecting the fault handling rules with the equipment hierarchical relationship to create a station-level system knowledge graph architecture.

[0037] Specifically, in the process of implementing the full twin virtual simulation test platform, building an expert mode database based on historical operation data and professional experience is a key link in achieving high-precision simulation. First, it is necessary to extract equipment fault records, operating parameter curves and alarm logs from the rail transit station-level system. Equipment fault records refer to detailed records of each subsystem equipment failure, including fault equipment identification, fault occurrence time, fault type, fault duration and other information; operating parameter curves record the data of various operating parameters of the equipment under normal and abnormal conditions over time, such as the air volume and current change curves of ventilation equipment, and mechanical parameter changes during the opening and closing of shield doors; alarm logs contain various warning and alarm information automatically generated by the system. These historical data are processed in sections using the time window sliding method. This method is a data analysis technology that sets a fixed length time window (such as 30 minutes) and gradually slides on the time axis (such as 5 minutes each time), and performs feature extraction and pattern recognition on the data in each window. By comparing the differences in data features during normal operation and before and after the fault occurs, the normal operation mode and fault operation mode of the equipment are identified. For example, when the wind turbine is operating normally, the current fluctuation range is stable at ±5% of the rated value, while abnormal patterns such as periodic spikes or continuous increases often appear before a fault occurs.

[0038] Next, professional processing cases such as fire alarm processing, platform shield door fault maintenance, and environmental control anomalies are collected. These cases come from various channels such as maintenance records, operation manuals, and expert interviews, and contain rich professional processing experience. These unstructured experience knowledge are converted through structured templates. Structured templates are a predefined data format framework that contains fixed fields such as problem description, symptom characteristics, diagnostic methods, processing steps, and verification methods. The technicians fill in the maintenance experience according to the template requirements, and the system automatically extracts key information and converts the unstructured text description into structured data that can be processed by the computer. For example, the processing experience of "platform shield door cannot be closed" is converted into structured data containing fields such as symptoms (door body stagnated in a half-open state), possible causes (limit switch failure, drive motor abnormality, control signal interruption), and processing steps (check limit switch, measure motor current, verify control signal). In this way, a large number of professional processing cases are converted into processing flow data in a standard format to form a professional experience library in the field of rail transit.

[0039] Subsequently, the equipment failure mode and the corresponding processing flow are associated and marked. Association marking is the process of establishing a mapping relationship between failures and solutions. By analyzing the matching degree between failure characteristics and processing cases, one or more applicable processing flows are matched for each failure mode. In the specific implementation, a fault code-processing solution mapping table is used to record this correspondence. The fault code is a coded representation of the abnormal state of the equipment, such as "E-PSD-001" indicates that the shielding door cannot be opened, and "E-FAS-002" indicates a false alarm of the smoke detector. Each fault code is associated with a set of applicable processing solution IDs to form a many-to-many mapping relationship. When the system detects that a certain device is abnormal, the corresponding fault code can be matched according to the fault characteristics, and then the corresponding processing suggestions can be directly obtained through the mapping table, a direct relationship between the equipment abnormality and the solution is established, and a station-level system rule knowledge base is generated.

[0040] In order to effectively organize and express the content of the rule knowledge base, a semantic triple structure in the form of "equipment-state-action" is adopted. The semantic triple is the basic unit of knowledge representation, which consists of three parts: subject (equipment), predicate (state) and object (action), and is formally represented as (equipment, state, action). For example, (shield door, cannot be closed, check limit switch) means that when the shield door cannot be closed, the limit switch should be checked. By defining the causal relationship between the main equipment (such as fire sprinkler system, shield door controller, platform broadcasting equipment, etc.), state conditions (such as fault, alarm, abnormality, etc.) and operation actions (such as inspection, maintenance, replacement, etc.), the content in the rule knowledge base is represented as a series of semantic triples, and the logical relationship between the triples is established to construct the semantic network of rail transit equipment. The nodes in the semantic network represent equipment or actions, and the edges represent the relationship between them, making the correlation between knowledge clearly visible.

[0041] Building a domain ontology model based on the rail transit equipment classification system is an important part of knowledge expression. The domain ontology is a formal description of specific domain concepts and their relationships, providing a shared vocabulary and semantic framework for domain knowledge. The rail transit equipment classification system adopts a four-level hierarchical structure: the first level is the system level, such as environmental control systems, fire protection systems, shield door systems, etc.; the second level is the subsystem level, such as ventilation subsystems and air conditioning subsystems under environmental control systems; the third level is the equipment type level, such as axial flow fans and centrifugal fans under ventilation subsystems; the fourth level is the specific equipment level, which identifies each actual equipment in the station, such as "Axial flow fan No. 3 on platform 1". Through this hierarchical division, the relationship between equipment at all levels is clearly defined, including inheritance relationships (subclasses inherit the properties of parent classes) and association relationships (functional connections between different equipment). For example, all fans inherit the common characteristics of "rotating equipment" and have a monitoring association relationship with a specific "temperature sensor".

[0042] Finally, the semantic network of rail transit equipment and the domain ontology model are integrated through concept mapping. Concept mapping is a technology that establishes the correspondence between two forms of knowledge representation. By defining the correspondence rules between the nodes in the semantic network and the concepts in the ontology model, the two knowledge structures are integrated. During the fusion process, the device nodes in the semantic network are mapped to the specific device instances in the ontology model, the state nodes are mapped to the device attributes, and the action nodes are mapped to the operation methods. Through this mapping, the scattered fault handling rules are connected with the systematic equipment hierarchical relationship to create a station-level system knowledge graph architecture. The knowledge graph architecture is a multi-level knowledge representation structure, with the bottom layer being basic facts (equipment attributes, status data), the middle layer being entity relationships (functional connections between devices), and the top layer being reasoning rules (fault diagnosis and processing logic).

[0043] Taking the emergency response of rail transit station fire as an example, when a smoke detector in a platform area triggers an alarm, the system first identifies the characteristic pattern of this type of alarm from historical data. The system analyzes the recent temperature change trend, passenger flow density change and other related data through the time window sliding method to determine whether it is a real fire or a possible false alarm. The system queries the processing flow of similar cases in the professional experience library and matches the most suitable emergency plan through the fault code-processing solution mapping table. Based on the "equipment-state-action" semantic triple, the system generates a series of control instructions: (fire sprinkler, regional confirmation of fire, start water spray), (shield door, emergency evacuation, all open), (platform broadcast, fire level 1 alarm, play evacuation instructions), etc. These instructions are accurately sent to the corresponding control devices according to the equipment hierarchical relationship defined in the domain ontology model. Throughout the process, the knowledge graph architecture provides complete knowledge support from alarm signal analysis, situation judgment to instruction generation to ensure the accuracy and timeliness of emergency response.

[0044] In a specific embodiment, the process of executing step S105 may specifically include the following steps: Extract rail transit screen door system operation data from real-time data streams, build a device operation status table based on the door opening and closing status, locking mechanism position and drive motor current value, and form a physical layer model of the device; Based on the PLC control status, the fire linkage control logic is recorded, and the emergency response rule set is created through the correspondence between the fire alarm signal and the shield door control command, and the control layer model is established; Based on the fault handling experience in the expert mode database, the linkage relationship between the environmental control system and the fire protection system is established, and the emergency handling model is generated through the collaborative workflow diagram of the ventilation equipment and the fire protection equipment; Combine the equipment physical layer model, control layer model and emergency processing model according to the station-level system architecture, establish inter-system communication relationships by defining data exchange interfaces between subsystems, and form a multi-level full twin digital model; Compare the data of the multi-level full twin digital model with the actual operating status on site, and dynamically adjust the model parameters according to the deviation value to synchronize the model behavior with the actual system; The multi-level full twin digital model is calculated and processed on the rail transit station control computing platform, and the simulation results are fed back to the operator through the real-time computing engine, so as to perform real-time simulation of on-site equipment and systems.

[0045] Specifically, in the process of realizing the full twin virtual simulation test platform, building a multi-level full twin digital model based on real-time data stream, expert mode database and PLC control state is the core link. First, the operation data of the rail transit platform shield door system is extracted from the real-time data stream. The real-time data stream is a continuous data stream collected from each subsystem by the station-level intelligent agent, which contains various equipment status information. For the platform shield door system, three key data are extracted: the door body opening and closing state, the locking mechanism position and the drive motor current value. The door body opening and closing state is collected by the position sensor and recorded as fully open, fully closed or intermediate position (the degree of opening is expressed in percentage); the locking mechanism position reflects whether the door lock is correctly engaged, which is obtained by the locking switch signal; the drive motor current value is measured in real time by the current sensor, reflecting the load condition of the door body during operation. These three types of data are integrated into an equipment operation status table, which is a multidimensional data structure that uses timestamps as indexes to record the values ​​of various parameters at different time points and mark abnormal conditions. By processing the state table data, a mathematical model that reflects the physical characteristics and operating rules of the shielding door can be established to form a physical layer model of the equipment. This model can accurately describe the operating behavior and state changes of the shielding door under different control signals. The fire linkage control logic is recorded based on the PLC control state. The PLC control state refers to the various control parameters and program logic stored in the PLC controller, which reflects the linkage relationship between the equipment. For the fire linkage control logic, the correspondence between the fire alarm signal and the shielding door control instruction is mainly analyzed. The fire alarm signal comes from fire fighting equipment such as smoke detectors and temperature detectors. After being processed by the fire control PLC, a series of linkage controls will be triggered, including the control of the shielding door. By analyzing the conditional judgment statements and control flow in the PLC program, the rules on how to control the shielding door under different fire levels and fire conditions in different areas are extracted. These rules are organized into a standard format emergency response rule set, which contains information such as trigger conditions, response actions, and priorities. For example, when the fire alarm in the platform area reaches level 2 or above, all shielding doors in the area should be opened automatically; when a fire occurs between equipment, the shielding doors in adjacent areas should be closed to isolate the fire. These rules constitute the control layer model, which describes the control decision-making process of the system in different situations.

[0046] The linkage relationship between the environmental control system and the fire protection system is established based on the fault handling experience in the expert mode database. The expert mode database stores a large number of historical fault handling cases and expert handling experience, which is very valuable for building the linkage relationship between systems. By analyzing the historical linkage cases between the fire protection system and the environmental control system, the collaborative working mode of the ventilation equipment and the fire protection equipment is extracted. For example, in a fire scenario, the ventilation system needs to perform the smoke exhaust function to guide the smoke to a specific area or discharge it outside the station; at the same time, it needs to adjust the airflow direction to avoid introducing fresh air into the fire area to intensify the combustion. These collaborative working rules are organized into a collaborative workflow diagram of ventilation equipment and fire protection equipment. The flowchart describes the complete process from fire detection to smoke exhaust operation, including equipment start-stop sequence, operation parameter adjustment, fault switching and other operation steps. Based on these flowcharts, an emergency handling model is constructed, which can guide the collaborative operation of the environmental control system and the fire protection system in an emergency and ensure the effective implementation of emergency measures. The equipment physical layer model, control layer model and emergency handling model are combined according to the station-level system architecture to form a multi-level full twin digital model. The station-level system architecture is the overall framework of the rail transit station control system, which defines the hierarchical relationship and functional division of each subsystem. In this architecture, three models are vertically integrated: the equipment physical layer model is located at the bottom layer, describing the physical characteristics and operating rules of the equipment; the control layer model is located in the middle layer, describing the control logic and decision-making process of the system; the emergency response model is located in the upper layer, describing the collaborative working mode across the system. In order to achieve effective communication between these three layers of models, it is necessary to define the data exchange interface between subsystems. The data exchange interface is a standardized channel for transmitting information between systems, including data format specifications, communication protocols, and interface functions. Through these interfaces, the upper-level model can obtain the status information of the lower-level model, and the lower-level model can receive the control instructions of the upper-level model, forming a complete information closed loop. This multi-level architecture enables the full twin digital model to accurately simulate the physical characteristics of a single device and correctly reflect the complex interactions between systems, thereby achieving a comprehensive simulation of the entire station-level system.

[0047] Data comparison is performed between the multi-level full twin digital model and the actual operating status on site to ensure the accuracy of the model. Data comparison refers to comparing the system behavior predicted by the model with the operating data of the actual system and calculating the deviation between the two. For the shield door system, the comparison content includes the door opening and closing time, motor current waveform, lock state change, etc.; for the environmental control system, the comparison content includes fan start and stop delay, air flow organization change, temperature response curve, etc. Through the comparison, the deviation value between the model parameters and the actual system is calculated. The deviation value is the basis for model adjustment and reflects the direction and magnitude of the model improvement. According to the deviation value, the model parameters are dynamically adjusted using the parameter adaptive algorithm. The parameter adaptive algorithm is a method that can automatically adjust the model parameters according to system feedback, which can gradually make the model behavior close to the actual system. Through this continuous comparison and adjustment, the synchronization between the full twin digital model and the actual system is ensured, and the simulation accuracy of the model is improved.

[0048] The multi-level full twin digital model is processed on the rail transit station control computing platform to achieve real-time simulation. The station control computing platform is a high-performance computing device deployed in the station control center with large-scale parallel computing capabilities. The full twin digital model runs on this platform, receives real-time sensor data, updates the model status, and calculates the future behavior of the system. The calculation process is managed by the real-time computing engine, which is a software system specially designed to handle time-sensitive computing tasks and can complete complex model calculations in milliseconds. The calculation results include the current state of the system, the predicted future state, potential abnormal conditions, etc. This information is fed back to the operator through the human-machine interface to help them monitor the system operation status, predict possible problems, and formulate response strategies. At the same time, the simulation results can also be directly used as a reference for control decisions, assist or automatically generate control instructions in emergency situations, control on-site equipment, and realize closed-loop operation from simulation to control.

[0049] Taking the emergency response of rail transit station fire as an example, when a smoke detector in a platform area triggers an alarm signal, the signal is first received by the fire protection system PLC, and then enters the full twin digital model through the station-level intelligent agent. The model extracts other related sensor data from the real-time data stream, including the temperature sensor data, passenger flow density data, and screen door status data of the area. The physical layer model judges the severity and development trend of the fire based on these data; the control layer model selects the appropriate emergency plan from the emergency response rule set according to the fire level and generates a control instruction sequence; the emergency response model coordinates the linkage operation of the environmental control system and the fire protection system, plans the smoke exhaust channel, and adjusts the operating parameters of the ventilation equipment. These model calculations are executed in parallel on the station control computing platform, and the results show that the fire has a tendency to spread to the station hall. The system immediately displays the fire analysis results and recommended disposal plans to the operator, and sends control instructions to the on-site equipment: open all the screen doors in the platform area to facilitate passenger evacuation, start the smoke exhaust fan to create a positive pressure zone to prevent smoke from spreading to the station hall, and adjust the station hall ventilation system to ensure fresh air in the evacuation channel. As the situation on site changes, the model is continuously updated and the control strategy is dynamically adjusted to ensure the accuracy and effectiveness of emergency response.

[0050] In a specific embodiment, the process of executing step S106 may specifically include the following steps: Using the full twin digital model as the training environment, the state space description of the rail transit station-level system is constructed, and the operating status of the environmental temperature and humidity, fire-fighting equipment, shielding doors, and broadcasting systems is numerically represented to form a system state vector. Extract station-level system operation rules from expert knowledge rules, transform emergency evacuation rules, equipment linkage strategies, and early warning disposal methods into numerical reward functions, and establish control objective functions; Deploy a deep reinforcement learning training framework on the station-level computing platform, train the control strategy generator based on the operation efficiency, energy consumption level, and safety indicators of rail transit stations as optimization directions; Perform security checks on the instruction schemes generated by the control strategy generator, filter illegal operations through the control instruction whitelist, and generate a safe control instruction set; The safety control instruction set is sent to the industrial-grade PLC controller through a two-way communication channel, and the PLC system actually controls the environmental equipment, access control system, and broadcasting equipment; The actual control effect data sent back by the PLC system is collected, the difference data between the theoretical expectation and the actual result is input into the reinforcement learning framework, the control strategy generator parameters are updated, and an intelligent closed-loop control is formed.

[0051] Specifically, the state space description is a comprehensive numerical expression of the current state of the system, covering all key parameters of the system. In the specific implementation, the ambient temperature and humidity data are collected, and the temperature range (such as 15-30℃) and humidity range (such as 30%-70%) are linearly mapped and converted into normalized values ​​between 0-1; the fire equipment status is encoded, and the equipment working status (standby, warning, alarm, fault, etc.) is converted into discrete numerical identifiers; for the shielding door system, parameters such as door position (opening percentage), locking status (locked / unlocked) and operating mode (automatic / manual) are recorded; for the broadcasting system, information such as the current playback status, volume level and content type is recorded. Through feature engineering technology, these heterogeneous data are integrated into a fixed-dimensional system state vector, each dimension represents a specific attribute of the system, forming a complete description of the entire station-level system status.

[0052] Extract station-level system operation rules from expert knowledge rules. Expert knowledge rules are operation specifications and experience summaries stored in the expert mode database, which contain the practical wisdom accumulated by human experts over many years. The operation rule extraction process includes three steps: rule parsing, key decision point identification and numerical conversion. For emergency evacuation rules, extract key points such as trigger conditions, evacuation path selection, and crowd control strategy; for equipment linkage strategy, extract equipment start and stop sequence, parameter adjustment method, fault replacement plan and other contents; for early warning disposal method, extract early warning level classification, response time requirements, upgrade processing flow and other regulations. Then, these qualitative operation rules are converted into quantitative reward functions. The reward function is a mathematical expression for evaluating the quality of actions in reinforcement learning, which is formed by weighted summation of various system indicators. For example, for emergency evacuation scenarios, the shorter the evacuation completion time and the lower the congestion, the higher the reward value; for equipment linkage, the lower the energy consumption, the faster the response, and the better the stability, the higher the reward value. These reward functions together constitute the control objective function, which provides optimization direction for reinforcement learning algorithms. Deploying a deep reinforcement learning training framework on a station-level computing platform is the technical basis for realizing intelligent control. Deep reinforcement learning is an artificial intelligence method that combines deep learning and reinforcement learning, and can continuously optimize decision-making strategies through interaction with the environment. The training framework is implemented using algorithms such as deep Q network (DQN) or advantage actor-critic (A2C). The former is suitable for discrete action space, and the latter is suitable for continuous action space. The framework construction includes three parts: environment interface definition, neural network structure design, and training strategy configuration. The environment interface connects the full twin digital model to achieve state acquisition and action execution; the neural network uses a multi-layer perceptron or convolutional network structure to map the state vector to action value or strategy distribution; the training strategy sets hyperparameters such as learning rate, discount factor, and exploration parameter to control the learning process. The training process takes the operation efficiency of rail transit stations (such as passenger flow throughput), energy consumption level (such as power consumption per unit time), and safety indicators (such as the frequency of safety incidents) as optimization directions. Through repeated interaction with the environment, the network parameters are continuously adjusted to eventually form a control strategy generator with decision-making capabilities.

[0053] It is crucial to perform security verification on the instruction scheme generated by the control strategy generator. This step ensures the safety and reliability of the control strategy generated by AI before actual execution. The security verification adopts a multi-layer filtering mechanism. First, a syntax check is performed to ensure that the instruction format conforms to the system specification; then a semantic check is performed to verify the logical rationality of the instruction content; and finally a security domain check is performed to ensure that the instruction will not cause the system to enter a dangerous state. Among them, the control instruction whitelist is the most critical security guarantee mechanism. The whitelist is a pre-defined set of operations that are allowed to be executed, which contains all verified safe operations. The instruction filtering process first parses the instruction scheme generated by the control strategy generator into basic operation units, and then matches it with the whitelist. Only operations that exist in the whitelist are allowed to be executed. If a potential dangerous operation is found, the system will automatically replace it with a safe alternative or issue a warning to the human operator. The instruction set that has been security-verified forms a safe control instruction set, which ensures the safety and reliability of all automatic control operations. The safe control instruction set is sent to the industrial-grade PLC controller through a two-way communication channel to achieve the connection between virtual simulation and actual control. The two-way communication channel is a data transmission path connecting the station-level computing platform and the field control equipment, which is implemented using technologies such as industrial Ethernet and fieldbus. The communication process includes three links: instruction encoding, transmission encryption, and reception confirmation. Instruction encoding converts abstract control instructions into specific command formats that can be recognized by PLC; transmission encryption ensures the security of instructions during transmission; and the reception confirmation mechanism ensures that instructions are received correctly and retransmitted when necessary. After receiving the control instructions, the industrial-grade PLC controller performs actual control operations on environmental equipment (such as air conditioning, ventilation systems), access control systems (such as shielded doors, ticket gates), broadcasting equipment (such as emergency broadcasts, information display screens), etc. according to the preset program logic, executes the operation requirements in the security control instruction set, and changes the working status or operating parameters of the equipment.

[0054] Collecting the actual control effect data sent back by the PLC system to form an intelligent closed-loop control is the final link of the full twin virtual simulation test platform. After executing the control command, the PLC system collects the execution results through the built-in sensors and status monitoring modules, including execution time, execution status, equipment response and other data, and transmits them back to the station-level computing platform through the uplink channel. The computing platform compares these actual results with the theoretical expectations and calculates the difference data. The difference data reflects the deviation between the model prediction and the actual execution, and is an important basis for evaluating the control effect and improving the control strategy. These difference data are input into the reinforcement learning framework as a training signal to drive the further optimization of the control strategy. The learning framework adjusts the neural network parameters according to the difference data and updates the internal model of the control strategy generator so that the control decision it generates is more in line with the response characteristics of the actual system, thereby forming a complete closed loop from environmental perception, decision generation, instruction execution to feedback learning, and realizing intelligent closed-loop control. Taking the large passenger flow in the rail transit station as an example, the full twin virtual simulation test platform can intelligently manage the passenger flow diversion process. The system first obtains passenger flow density data, station temperature and humidity data, and equipment status data from the real-time data stream to construct the current system state vector. Based on this state, the deep reinforcement learning framework combines the experience of handling large passenger flows in the expert knowledge base to generate a preliminary control plan, including opening additional ticket inspection channels, adjusting the opening time of the platform shield doors, optimizing the ventilation parameters in the station, and playing guidance broadcasts. The safety verification mechanism checks these operations to ensure that they do not violate the system safety constraints, such as preventing the platform shield doors from opening before the train arrives. The verified safety instructions are sent to each PLC controller through the communication channel for execution, and the system continuously monitors the control effect. If it is found that the passenger flow in a certain channel is still congested, the system will adjust the strategy in real time, increase the guidance broadcast frequency of the adjacent channel, and adjust the opening time of the platform shield doors to guide more passengers to use other entrances and exits. These actual effect data are collected and compared with the expected effect, and the difference information is used to update the control model so that the system performs better when handling similar situations next time, thereby realizing the continuous optimization of passenger flow management strategies and improving station operation efficiency.

[0055] The above describes the implementation method of the full twin virtual simulation test platform in the embodiment of the present application. The following describes the implementation system of the full twin virtual simulation test platform in the embodiment of the present application. Please refer to Figure 4 In the embodiment of the present application, an implementation system of the full twin virtual simulation test platform includes: The construction module is used to build a fully connected architecture by deploying station-level agents and high-speed bus networks, aggregating data from each subsystem into the station-level agent to obtain real-time data streams; The processing module is used to set up a professional volume data acquisition unit according to the characteristics of each subsystem, pre-process the acquired raw data, and generate a standardized data set; The receiving module is used to establish a two-way communication channel with the station-level intelligent agent using the industrial-grade PLC controller, transmit the equipment operation status to the intelligent agent and receive control instructions to form a closed-loop control mechanism; Create a module to build an expert model database based on historical operation data and professional experience, express professional knowledge through a combination of semantic network and ontology model, and create a knowledge graph architecture; The simulation module is used to build a multi-level full twin digital model based on real-time data streams, expert mode databases, and PLC control status, and to simulate field equipment and systems in real time using a combination of physical modeling and data-driven methods; The optimization module is used to use the deep reinforcement learning framework combined with expert knowledge rules to perform online learning and optimization of the full twin digital model, generate control strategies, execute and feedback the results through the PLC system, and form intelligent closed-loop control.

[0056] Through the collaboration of the above components, a fully connected architecture is built by deploying station-level intelligent agents and high-speed bus networks to achieve real-time convergence of data from each subsystem, significantly reducing data transmission delays and shortening system response time from seconds in traditional solutions to milliseconds, providing a solid data foundation for real-time control decisions. At the same time, professional data acquisition units are set up according to the characteristics of each subsystem to pre-process the raw data to generate standardized data sets, effectively solving the problem of multi-source heterogeneous data fusion, improving data quality and processing efficiency, and reducing the workload of data cleaning in subsequent analysis. Industrial-grade PLC controllers are used to establish a two-way communication channel with station-level intelligent agents to form a closed-loop control mechanism, overcoming the limitations of traditional one-way control, enabling control instructions to be dynamically adjusted according to real-time feedback, and improving control accuracy and reliability. An expert pattern database is built based on historical operating data and professional experience, and a knowledge graph architecture is created by combining semantic networks with ontology models to transform implicit expert knowledge into explicit rules, providing knowledge support for artificial intelligence decision-making, enabling AI models to reason in combination with domain expertise, and avoiding blind decisions that may be caused by pure data drive. Based on real-time data streams, expert pattern databases, and P The LC control state constructs a multi-level full twin digital model, and uses a combination of physical modeling and data-driven methods for real-time simulation, achieving accurate mapping of virtual space and the physical world. The simulation accuracy is improved by more than 80% compared with traditional methods, providing a reliable tool for predictive analysis and preventive control; the deep reinforcement learning framework is used in combination with expert knowledge rules to conduct online learning and optimization of the full twin digital model, generate control strategies and execute feedback through the PLC system to form an intelligent closed-loop control. The deep reinforcement learning algorithm in this solution is optimized specifically for rail transit control scenarios. By integrating expert knowledge rules and safety constraints, it not only retains the autonomous learning ability of AI, but also ensures the safety and reliability of control decisions, solving the problem of limited application of traditional AI algorithms in safety-critical areas. The system continuously learns the actual control effect and continuously optimizes the decision-making model, making the control strategy more and more in line with actual needs, forming a self-evolving closed-loop system; the platform breaks the barriers between virtual simulation and on-site control, and realizes the full process from data collection, knowledge expression, model construction to intelligent control, providing a complete solution for the intelligent upgrade of rail transit station-level systems, significantly improving the safety, reliability and operation efficiency of the system.

[0057] Reference Figure 5 In an embodiment of the present invention, a computer device is also provided. The computer device may be a server, and its internal structure may be as follows: Figure 5As shown. The computer device includes a processor, a memory, a display screen, an input device, a network interface and a database connected through a system bus. Among them, the processor designed by the computer is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the corresponding data in this embodiment. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the above method is implemented.

[0058] Those skilled in the art will understand that Figure 5 The structure shown in the figure is merely a block diagram of a portion of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied.

[0059] An embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the above method is implemented. It can be understood that the computer-readable storage medium in this embodiment can be a volatile readable storage medium or a non-volatile readable storage medium.

[0060] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0061] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or the whole or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program code.

[0062] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for implementing a full twin virtual simulation test platform, characterized in that: include: By deploying station-level agents and high-speed bus networks to build a fully connected architecture, data from each subsystem is aggregated into the station-level agent to obtain real-time data streams; According to the characteristics of each subsystem, a professional volume data acquisition unit is set up to pre-process the collected raw data and generate a standardized data set; Use industrial-grade PLC controllers to establish a two-way communication channel with station-level agents, transmit the equipment operating status to the agent and receive control instructions, forming a closed-loop control mechanism; Build an expert model database based on historical operation data and professional experience, express professional knowledge through a combination of semantic network and ontology model, and create a knowledge graph architecture; Based on real-time data streams, expert mode databases and PLC control status, a multi-level full twin digital model is constructed, and a combination of physical modeling and data-driven methods is used to perform real-time simulation of field equipment and systems. The deep reinforcement learning framework is combined with expert knowledge rules to conduct online learning and optimization of the full twin digital model, generate control strategies, execute and feedback the results through the PLC system, and form intelligent closed-loop control.

2. The method for implementing the full twin virtual simulation test platform according to claim 1 is characterized in that: The fully connected architecture is constructed by deploying station-level intelligent agents and high-speed bus networks, and the data of each subsystem is aggregated into the station-level intelligent agent to obtain real-time data streams, including: Deploy a station-level intelligent agent with an edge computing unit in a rail transit-like station-level system, and perform preliminary filtering on the original signal through a DSP processing chip to obtain preprocessed data; The station-level intelligent body is connected to each subsystem interface through an industrial-grade high-speed bus, and a real-time communication protocol is used to prioritize data packets to form a data transmission channel; Data collection points are set up for the basic automation system, fire alarm system, passenger information system, broadcasting system, screen door system, automatic train monitoring system, and monitoring and data acquisition system. Only state change information is transmitted through the differential transmission mechanism to reduce data redundancy; A local cache module is set up for the collected data. Through the data queue management algorithm, data is temporarily stored in the event of communication interruption and automatically resent after communication is restored to ensure data integrity; The internal functional units of the intelligent body are configured using a modular design method, and the functional modules are interconnected through an internal bus to support hot-swap functions; The data of each subsystem is structured according to a unified coding format, and the data attributes and associations are marked through a self-describing data structure algorithm to generate a real-time data stream.

3. The method for implementing the full twin virtual simulation test platform according to claim 1 is characterized in that: The step of setting up a professional volume data acquisition unit according to the characteristics of each subsystem, preprocessing the acquired raw data, and generating a standardized data set includes: Configure environmental data acquisition units for the basic automation systems in rail transit stations to perform linear calibration on original signals such as temperature, humidity, energy consumption, and lighting to obtain standard environmental parameters; Set up a safety status acquisition unit for the fire alarm system to perform threshold screening on smoke concentration and temperature alarm signals to form effective alarm status data; Build an information collection unit for the passenger information system, classify and integrate the broadcast content and display information in time, and generate a passenger information instruction stream; A data buffer is set in the professional volume data acquisition unit to periodically store the acquired state change data to reduce the amount of data transmission; Perform time synchronization on data from different subsystems, sort asynchronously collected information according to a unified clock reference, and build a data stream with consistent timing; All preprocessed data are converted into a unified field structure, and the numerical type, text type, and status type are standardized to generate a standardized data set.

4. The method for implementing the full twin virtual simulation test platform according to claim 1 is characterized in that: The industrial-grade PLC controller is used to establish a two-way communication channel with the station-level intelligent agent, transmit the equipment operation status to the intelligent agent and receive control instructions to form a closed-loop control mechanism, including: Deploy industrial-grade PLC controllers at key control nodes of rail transit stations, physically connect them to high-speed buses through communication expansion modules, and establish a hardware communication foundation; Configure a communication protocol converter for the PLC controller to perform unified interface conversion on the OPC UA and Modbus TCP communication protocols to form a protocol compatibility layer; Build a bidirectional data channel structure, use the uplink channel for device operation data transmission, and use the downlink channel for control command transmission to generate a complete communication path; Divide communication priorities according to the urgency of data, grant the highest transmission authority to emergency control commands, and achieve millisecond-level command response; Configure a communication redundancy mechanism to automatically switch to a backup link when the communication channel is interrupted to maintain data transmission continuity; The data interaction between the PLC controller and the station-level intelligent agent is checked for integrity, and the data consistency is verified through the cyclic redundancy check method to form a closed-loop control mechanism.

5. The method for implementing the full twin virtual simulation test platform according to claim 1 is characterized in that: The expert model database is constructed based on historical operation data and professional experience, and professional knowledge is expressed by combining semantic network and ontology model to create a knowledge graph architecture, including: Extract equipment fault records, operating parameter curves and alarm logs from the rail transit station-level system, segment the operating data using the time window sliding method, and identify the normal operating mode and fault operating mode of the equipment; Collect cases of fire alarm processing, platform shield door failure repair, and environmental control abnormality professional processing, and convert maintenance experience into processing flow data through structured templates to form a professional experience database in the rail transit field; The equipment failure mode is associated with the corresponding processing flow, and a direct relationship between equipment anomalies and solutions is established through a fault code-processing solution mapping table to generate a station-level system rule knowledge base; The semantic triple structure of equipment-state-action is used to organize the rule knowledge base content, and the semantic network of rail transit equipment is constructed by defining the causal relationship between the main equipment, state conditions and operation actions. Based on the rail transit equipment classification system, a domain ontology model is constructed, and the equipment is hierarchically divided into a four-level structure of system-subsystem-equipment type-specific equipment, and the inheritance and association relationship between equipment is established; The semantic network of rail transit equipment and the domain ontology model are integrated through concept mapping, connecting the fault handling rules with the equipment hierarchical relationship to create a station-level system knowledge graph architecture.

6. The method for implementing the full twin virtual simulation test platform according to claim 1 is characterized in that: The method of constructing a multi-level full twin digital model based on real-time data flow, expert mode database and PLC control status, and using a method combining physical modeling and data-driven to perform real-time simulation of field equipment and systems includes: Extract rail transit screen door system operation data from real-time data streams, build a device operation status table based on the door opening and closing status, locking mechanism position and drive motor current value, and form a physical layer model of the device; Based on the PLC control status, the fire linkage control logic is recorded, and the emergency response rule set is created through the correspondence between the fire alarm signal and the shield door control command, and the control layer model is established; Based on the fault handling experience in the expert mode database, the linkage relationship between the environmental control system and the fire protection system is established, and the emergency handling model is generated through the collaborative workflow diagram of the ventilation equipment and the fire protection equipment; Combine the equipment physical layer model, control layer model and emergency processing model according to the station-level system architecture, establish inter-system communication relationships by defining data exchange interfaces between subsystems, and form a multi-level full twin digital model; Compare the data of the multi-level full twin digital model with the actual operating status on site, and dynamically adjust the model parameters according to the deviation value to synchronize the model behavior with the actual system; The multi-level full twin digital model is calculated and processed on the rail transit station control computing platform, and the simulation results are fed back to the operator through the real-time computing engine, so as to perform real-time simulation of on-site equipment and systems.

7. The method for implementing the full twin virtual simulation test platform according to claim 1, characterized in that: The deep reinforcement learning framework is combined with expert knowledge rules to conduct online learning and optimization of the full twin digital model, generate control strategies, execute and feedback the results through the PLC system, and form intelligent closed-loop control, including: Using the full twin digital model as the training environment, the state space description of the rail transit station-level system is constructed, and the operating status of the environmental temperature and humidity, fire-fighting equipment, shielding doors, and broadcasting systems is numerically represented to form a system state vector. Extract station-level system operation rules from expert knowledge rules, transform emergency evacuation rules, equipment linkage strategies, and early warning disposal methods into numerical reward functions, and establish control objective functions; Deploy a deep reinforcement learning training framework on the station-level computing platform, train the control strategy generator based on the operation efficiency, energy consumption level, and safety indicators of rail transit stations as optimization directions; Perform security checks on the instruction schemes generated by the control strategy generator, filter illegal operations through the control instruction whitelist, and generate a safe control instruction set; The safety control instruction set is sent to the industrial-grade PLC controller through a two-way communication channel, and the PLC system actually controls the environmental equipment, access control system, and broadcasting equipment; The actual control effect data sent back by the PLC system is collected, the difference data between the theoretical expectation and the actual result is input into the reinforcement learning framework, the control strategy generator parameters are updated, and an intelligent closed-loop control is formed.

8. A system for implementing a full twin virtual simulation test platform, used to implement the method for implementing a full twin virtual simulation test platform as described in any one of claims 1 to 7, characterized in that: include: The construction module is used to build a fully connected architecture by deploying station-level agents and high-speed bus networks, aggregating data from each subsystem into the station-level agent to obtain real-time data streams; The processing module is used to set up a professional volume data acquisition unit according to the characteristics of each subsystem, pre-process the acquired raw data, and generate a standardized data set; The receiving module is used to establish a two-way communication channel with the station-level intelligent agent using the industrial-grade PLC controller, transmit the equipment operation status to the intelligent agent and receive control instructions to form a closed-loop control mechanism; Create a module to build an expert model database based on historical operation data and professional experience, express professional knowledge through a combination of semantic network and ontology model, and create a knowledge graph architecture; The simulation module is used to build a multi-level full twin digital model based on real-time data streams, expert mode databases, and PLC control status, and to simulate field equipment and systems in real time using a combination of physical modeling and data-driven methods; The optimization module is used to use the deep reinforcement learning framework combined with expert knowledge rules to perform online learning and optimization of the full twin digital model, generate control strategies, execute and feedback the results through the PLC system, and form intelligent closed-loop control.

9. A computer device, characterized in that: It includes a memory and a processor, the memory stores a computer program that can be run on the processor, and is characterized in that when the processor executes the computer program, it implements the implementation method of the full twin virtual simulation test platform described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the processor executes the implementation method of the full twin virtual simulation test platform as described in any one of claims 1 to 7.

Images