Switching station control method and system of subway station

Abstract

The invention provides a switching station control method and system of a subway station. The system comprises a subsystem control unit, a main control unit, a communication module and a data caching module, the subsystem control unit transmits an equipment on-off state, a fault signal and an environment parameter to the main control unit in a hybrid networking mode, and a first cooperative control instruction set is generated in combination with a linkage rule base; a target priority is distributed for the operation instruction to obtain a second cooperative control instruction set, and the communication module adopts a multi-channel parallel distribution mechanism for pushing; in the process of executing the switching station operation, the data caching module stores running state data, identifies a time sequence mode by using a machine learning model, pre-judges a switching station linkage scene and loads a control strategy; and the main control unit detects the response delay of the subsystem control unit and generates a third cooperative control instruction set according to the control strategy so as to optimize the switching station control time sequence. According to the invention, intelligent cooperative control of multiple electromechanical devices can be realized.

Description

Technical Field

[0001] This application relates to the technical field of urban rail transit, and particularly to a method and system for controlling a switch station of a subway station. Background Art

[0002] The management of subway station opening and closing involves the coordinated control of multiple devices, such as lighting, ventilation, platform doors, power supply systems, etc. It is necessary to complete equipment status detection, instruction issuance, and execution feedback within a short period of time, and at the same time, it is necessary to cope with dynamic scenarios such as passenger flow fluctuations and equipment failures. The acquisition of equipment status and the transmission of control instructions need to be completed within milliseconds to avoid equipment response lags or operation conflicts caused by delays. Dynamically allocate operation priorities according to preset rules to ensure that key devices, such as power supply systems and signal devices, are started or shut down first, to avoid failures caused by disorderly operations. Adjust the control strategy according to real-time environmental parameters, such as temperature, humidity, and passenger flow density. For example, start the ventilation system in advance during peak hours, or quickly switch to the backup power supply in case of a failure.

[0003] In response to such requirements, a typical solution provided by existing solutions is the timing scheduling technology based on a centralized automation control system. This solution uniformly collects equipment status through a central control unit, generates an operation instruction sequence based on preset timing logic, and distributes it to each subsystem through a single-channel communication network. For example, in the scenario of distribution network automation, the main control unit receives signals from substations and switch stations through subsystem control units, centrally determines the fault location, and then issues tripping or closing instructions. Such systems usually adopt fixed priority rules, such as giving priority to processing power supply loop protection actions and then executing environmental equipment regulation.

[0004] The core defect of the existing solution lies in the insufficient adaptability of the static timing logic to dynamic scenarios. Since the control strategy depends on preset fixed rules, it is unable to dynamically adjust the operation priority according to the real-time equipment status or environmental changes. For example, when a certain device responds with a delay, the system lacks the ability of adaptive timing optimization, and the overall operation process may stagnate due to waiting timeouts. In addition, the single-channel communication mechanism is prone to instruction distribution delays due to network congestion, and lacks learning ability, making it difficult to predict complex linkage scenarios, resulting in a rigid control strategy. Summary of the Invention

[0005] This application provides a method and system for controlling a switch station of a subway station to solve the problems of insufficient adaptability of static timing logic to dynamic scenarios, delays in the single-channel communication mechanism, and lack of learning ability in the prior art.

[0006] In a first aspect, the present application provides a switch station control system for a subway station, including: a subsystem control unit distributedly deployed within the switch station control system, configured to collect device switch states, fault signals, and environmental parameters, and transmit the device switch states, fault signals, and environmental parameters to a main control unit through a hybrid networking method; The main control unit is configured to generate a first collaborative control instruction set that matches the switch station control requirements according to the device switch states, fault signals, and environmental parameters, in combination with a linkage rule library. The first collaborative control instruction set includes operation instructions for each electromechanical device, and corresponding target priorities are assigned to the operation instructions for each electromechanical device to obtain a second collaborative control instruction set; A communication module is configured to encapsulate the second collaborative control instruction set and push the second collaborative control instruction set to each subsystem control unit by using a multi-channel parallel distribution mechanism; A data cache module is configured to store the operation state data of each electromechanical device during the process of the subsystem control unit performing switch station operations based on the second collaborative control instruction set, identify a timing pattern by using a machine learning model, predict a switch station linkage scenario based on the timing pattern, and load a corresponding control strategy; The main control unit is configured to construct a digital twin model based on the operation state data and execution feedback data of each electromechanical device, dynamically evaluate the digital twin model, and then detect the response delay of the subsystem control unit. When the response delay exceeds a set threshold, a third collaborative control instruction set is generated according to the control strategy to optimize the switch station control timing.

[0007] Optionally, the data cache module, configured to store the operation state data of each electromechanical device during the process of the subsystem control unit performing switch station operations based on the second collaborative control instruction set, identify a timing pattern by using a machine learning model, predict a switch station linkage scenario based on the timing pattern, and load a corresponding control strategy, includes: The data cache module is configured to, when the subsystem control unit executes the second collaborative control instruction set, obtain the operation state data of each electromechanical device in real time. The operation state data is recorded in the form of an operation event stream, and the operation event stream includes a device operation timestamp, a device state code, and environmental parameters. The operation state data is stored in a buffer in ascending order of the timestamp, and a sliding window elimination operation is performed on old data that exceeds a preset time threshold to form an operation event sequence pool; The data caching module is further configured to extract overlapping data segments of adjacent time windows from the operation event sequence pool, convert the device status codes in each data segment into state transition segments; after feature splicing the state transition segments with the environmental parameters, input them into a machine learning model for frequency statistics and correlation analysis of cross-device state transition paths, and output a timing pattern representing the causal relationship between devices; The data caching module is further configured to match the timing pattern with a preset topological structure of the switchyard linkage scenario, generate a candidate scenario queue sorted by priority according to the matching result, and select the control strategy corresponding to the candidate scenario with the highest priority.

[0008] Optionally, the data caching module is further configured to extract overlapping data segments of adjacent time windows from the operation event sequence pool, convert the device status codes in each data segment into state transition segments; after feature splicing the state transition segments with the environmental parameters, input them into a machine learning model for frequency statistics and correlation analysis of cross-device state transition paths, and output a timing pattern representing the causal relationship between devices, including: A window intercepting unit, configured to intercept overlapping data segments of adjacent time windows from the operation event sequence pool according to a preset sliding step length, where the overlapping data segments include a device status code sequence, construct state transition pairs according to the device status code sequence in ascending order of time stamps, and each state transition pair is composed of a previous state code and a subsequent state code, and associate the corresponding operation time interval to form a state transition segment set; A feature embedding unit, configured to perform feature splicing on the state transition segments and the environmental parameters within the same time window, and embed the environmental parameters into a preset bit field of the state transition segments during the feature splicing process to form a state transition vector; An analysis unit, configured to input the state transition vector into a machine learning model, count the co-occurrence frequencies of the state transition vectors of different devices, analyze the time window overlap degree and interval correlation of the state transition segments of different devices, generate a set of tightness indicators for cross-device state transitions, construct a weighted directed acyclic graph, where the nodes are the device status codes, and the weight value of the edge is determined by the product of the co-occurrence frequency and the time window overlap degree, and output a timing pattern.

[0009] Optionally, the analysis unit is configured to input the state transition vector into a machine learning model, count the co-occurrence frequencies of the state transition vectors of different devices, analyze the time window overlap degree and interval correlation of the state transition segments of different devices, generate a set of tightness indicators for cross-device state transitions, including: A co-occurrence detection subunit, which is configured to input the state transition vector into a machine learning model, traverse the state transition vectors of different devices, identify the co-occurrence frequency of the state transition vectors of different devices within the same time window, and generate a co-occurrence frequency matrix of device pairs; An overlap degree calculation subunit, which is configured to calculate the time window overlap length of the state transition segments of two devices in a device pair, and determine the ratio of the time window overlap length to the total time window length as the time window overlap degree; A correlation analysis subunit, which is configured to extract the time interval sequence of the state transition between two devices in a device pair, and calculate the interval correlation according to the time interval sequence through the Pearson correlation coefficient; A tightness index set construction subunit, which is configured to combine the co-occurrence frequency matrix, the time window overlap degree, and the interval correlation to obtain a tightness index set.

[0010] Optionally, the main control unit is configured to generate a first collaborative control instruction set that matches the switchyard control requirements according to the device switch state, fault signal, and environmental parameters, in combination with the linkage rule library. The first collaborative control instruction set includes operation instructions for each electromechanical device, and assigns corresponding target priorities to the operation instructions for each electromechanical device to obtain a second collaborative control instruction set, including: A rule activation subunit, which is configured to match the device switch state, fault signal, and environmental parameters with the rule trigger conditions in the linkage rule library, filter out an activation rule set that meets the consistency of the device state transition direction and the environmental parameters meet the threshold constraints, and extract the first collaborative control instruction set bound to the activation rule set. The rule trigger conditions are quantified switchyard control requirements; A priority weighting subunit, which is configured to obtain the basic priority of each operation instruction in the first collaborative control instruction set from a preset priority table according to the device type, and perform dynamic weighting processing on the basic priority value in combination with the fault signal level to obtain a transition instruction set with target priority labels; A conflict adjudication subunit, which is configured to detect mutually exclusive operation instructions acting on the same device in the transition instruction set, and retain the highest priority instruction to generate a conflict-free instruction set; A timing arrangement subunit, which is configured to perform device-level timing arrangement on the conflict-free instruction set, and merge consecutive instructions according to the minimum operation interval time of the device to obtain a second collaborative control instruction set.

[0011] Optionally, the rule activation subunit is configured to match the device switch state, fault signal, and environmental parameters with the rule trigger conditions in the linkage rule library, filter out an activation rule set that meets the consistency of the device state transition direction and the environmental parameters meet the threshold constraints, and extract the first collaborative control instruction set bound to the activation rule set, including: An event coupling component, configured to receive the jump direction identifier of the device switch state, the encoded stream of the fault signal, and the sampled stream of the environmental parameters, and generate a composite event data packet with environmental constraints according to the timestamp alignment method; A rule penetration component, configured to perform two-way verification on the composite event data packet and the rule trigger conditions in the linkage rule library, and filter out an activation rule set that simultaneously satisfies the consistency of the device state jump direction and the environmental parameter compliance with the threshold constraint. The two-way verification includes forward verification and reverse verification; An instruction synthesis component, configured to extract the original operation instruction fragments bound by the activation rule set, eliminate mutually exclusive instructions according to the device operation conflict pre-check result, and merge and generate a first collaborative control instruction set.

[0012] Optionally, the communication module is configured to encapsulate the second collaborative control instruction set and push the second collaborative control instruction set to each subsystem control unit by using a multi-channel parallel distribution mechanism, including: An instruction sharding unit, configured to shard the second collaborative control instruction set according to the distribution of the physical locations of the devices into regionalized instruction subsets, and each subset is encapsulated into an independent data packet carrying the device group code; A channel binding unit, configured to query the preset communication topology relationship according to the device group code, and allocate a primary transmission channel identifier and a backup transmission channel identifier to each independent data packet to form a dual-channel bound data packet; A redundant transmission unit, configured to synchronously send the dual-channel bound data packet through the primary and backup dual channels, and monitor the channel health status in real time during the primary channel transmission. When it is detected that the data packet has not been delivered within the device response timeout window, it automatically switches to the backup channel to continue the transmission; A delivery confirmation unit, configured to receive the data packet reception status code returned by each subsystem control unit, and trigger the channel switching and retransmission of the corresponding block instruction subset for the device group code that has not been completely received until all subsystem control units return the reception success status code to complete the push of the second collaborative control instruction set.

[0013] In a second aspect, the present application provides a method for controlling a switch station in a subway station, including: Collecting the device switch state, fault signal, and environmental parameters through the subsystem control units distributedly deployed in the switch station control system, and transmitting the device switch state, fault signal, and environmental parameters to the main control unit through a hybrid networking method; The main control unit generates a first collaborative control instruction set that matches the control requirements of the switchyard according to the device switch state, fault signals, and environmental parameters, in combination with the linkage rule library. The first collaborative control instruction set includes operation instructions for each electromechanical device, and corresponding target priorities are assigned to the operation instructions for each electromechanical device to obtain a second collaborative control instruction set; The communication module encapsulates the second collaborative control instruction set, and the second collaborative control instruction set is pushed to each subsystem control unit by using a multi-channel parallel distribution mechanism; During the process of the subsystem control unit performing switchyard operations based on the second collaborative control instruction set, the data cache module stores the operation status data of each electromechanical device, identifies the timing pattern in combination with the machine learning model, anticipates the switchyard linkage scenario based on the timing pattern, and loads the corresponding control strategy; The main control unit constructs a digital twin model based on the operation status data and execution feedback data of each electromechanical device, dynamically evaluates the digital twin model, and then detects the response delay of the subsystem control unit. When the response delay exceeds the set threshold, the switchyard control timing is optimized.

[0014] In a third aspect, the present application provides a computing device, including a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a switchyard control method for a subway station as described in the second aspect.

[0015] In a fourth aspect, the present application provides a computer storage medium storing a computer program, and when the computer program is executed by a computer, it implements a switchyard control method for a subway station as described in the second aspect.

[0016] In an embodiment of the present application, a switch station control system for a subway station is provided. The system includes: a subsystem control unit distributedly deployed within the switch station control system, configured to collect device switch states, fault signals, and environmental parameters, and transmit the device switch states, fault signals, and environmental parameters to a main control unit through a hybrid networking method; the main control unit, configured to generate a first collaborative control instruction set matching the switch station control requirements according to the device switch states, fault signals, and environmental parameters, in combination with a linkage rule library, where the first collaborative control instruction set includes operation instructions for each electromechanical device, and assign corresponding target priorities to the operation instructions for each electromechanical device to obtain a second collaborative control instruction set; a communication module, configured to encapsulate the second collaborative control instruction set and push the second collaborative control instruction set to each subsystem control unit by using a multi-channel parallel distribution mechanism; a data cache module, configured to store the operation state data of each electromechanical device during the process of the subsystem control unit performing switch station operations based on the second collaborative control instruction set, identify a timing pattern by using a machine learning model, predict a switch station linkage scenario based on the timing pattern, and load a corresponding control strategy; the main control unit, configured to construct a digital twin model based on the operation state data and execution feedback data of each electromechanical device, dynamically evaluate the digital twin model, and then detect the response delay of the subsystem control unit. When the response delay exceeds a set threshold, generate a third collaborative control instruction set according to the control strategy to optimize the switch station control timing.

[0017] The present application has the following advantages: Through distributed acquisition and hybrid networking transmission, real-time aggregation of device state data is achieved; in combination with a linkage rule library, a control instruction set with clear priorities is dynamically generated to avoid operation conflicts, ensure that key devices respond first, and reduce the risk of human misjudgment. Based on the analysis of the timing pattern of operation data by a machine learning model, a linkage scenario is predicted and a matching strategy is loaded; through a digital twin model, the response delay is evaluated in real time, and an optimized control instruction set is dynamically generated to solve the problem of insufficient adaptability between traditional static timing logic and dynamic scenarios. The multi-channel parallel distribution mechanism is adopted to avoid the congestion risk of single-channel communication and ensure the fast and reliable transmission of the instruction set to each subsystem; the data cache module stores the operation state data in real time, supports fast recovery after an abnormal interruption, and improves the robustness of the system. By using machine learning to predict complex linkage scenarios and preloading control strategies in advance; combined with the dynamic evaluation of the digital twin model, closed-loop optimization of the control timing is achieved, and the response time under abnormal conditions is shortened.

[0018] Furthermore, the data cache module constructs an operation event sequence pool in the order of timestamps by collecting the operation status data of electromechanical devices in real time, and maintains dynamic updates by using a sliding window to eliminate old data; extracts device state transition segments based on the overlapping data segments of adjacent time windows, splices them with environmental parameters into feature vectors, and then inputs them into a machine learning model to analyze the co-occurrence frequency, time window overlap degree, and interval correlation of cross-device state transitions, constructs a weighted directed acyclic graph to represent the causal relationship between devices; finally, matches the generated time series pattern with the preset linkage scenario topology, generates a candidate control strategy queue according to the priority, and realizes the adaptive loading of the scenario. Through the sliding window and state transition fragmentation processing, this solution accurately captures the temporal correlation of device state changes and the coupling effect of environmental parameters, breaking through the dependence of traditional rule bases on static causal relationships; quantifies the linkage strength between devices based on the tightness index of the directed acyclic graph, combines the machine learning model to dynamically mine potential fault propagation paths, and improves the prediction accuracy of complex linkage scenarios; through the candidate scenario queue sorted by priority, ensures that the control strategy is highly compatible with the real-time operation mode, reduces the need for manual intervention, and reduces the risk of control delay or misoperation caused by scenario misjudgment.

[0019] These aspects or other aspects of the present application will be more clearly understood in the following description of the embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained according to these drawings.

[0021] Figure 1 It is a schematic structural diagram of a switch station control system for a subway station provided by an embodiment of the present application; Figure 2 It is a flowchart of a switch station control method for a subway station provided by an embodiment of the present application; Figure 3 It is a schematic structural diagram of a computing device provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0022] In order to enable those skilled in the art to better understand the solution of the present application, the following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application.

[0023] In some processes described in the specification, claims, and the above-mentioned drawings of this application, a plurality of operations appear in a specific order. However, it should be clearly understood that these operations may not be executed in the order in which they appear herein or may be executed in parallel. The serial numbers of the operations, such as 11, 12, etc., are only used to distinguish different operations, and the serial numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions such as "first" and "second" in this article are used to distinguish different messages, devices, modules, etc., do not represent a sequence, and do not limit that "first" and "second" are of different types.

[0024] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of protection of the present application.

[0025] To solve the problems of insufficient adaptability between static timing logic and dynamic scenarios, delay in single-channel communication mechanisms, and lack of timing pattern learning ability in the prior art, the embodiments of the present application provide a switch station control system for a subway station. The method adopts the following concept: by collecting device data distributively and transmitting it through a hybrid network, the main control unit dynamically generates a priority instruction set in combination with a rule library and distributes it to the subsystems through multiple channels; the data cache module records the operating status in real time and uses machine learning to predict the linkage scenario, while the digital twin model monitors the delay and dynamically optimizes the instruction timing, constructing a "collection - decision - feedback - optimization" closed loop to achieve dynamic adaptive control of the switch station.

[0026] Figure 1 The structural schematic diagram of a switch station control system provided by the embodiments of the present application is as Figure 1 shown. The system includes: The subsystem control unit 10 distributedly deployed in the switch station control system, which is used to collect the device switch status, fault signals, and environmental parameters, and transmit the device switch status, fault signals, and environmental parameters to the main control unit 11 through a hybrid networking method.

[0027] Among them, the hybrid networking method refers to the composite networking technology that simultaneously uses industrial Ethernet, 5G network, and fiber optic private network. The electromechanical equipment includes lighting, platform screen doors, escalators, ticket gates, ticket vending machines, broadcasting equipment, environmental control equipment, etc. The execution feedback data refers to the real-time result data generated when each electromechanical equipment executes operations in response to the collaborative control instruction, and is captured in real time by the sensor networks built in each subsystem control unit 10, such as the lighting control unit and the escalator control unit.

[0028] In the embodiment of the present application, the subsystem control unit 10 distributedly deployed in the switch station control system collects the switch state signals, fault signals and environmental parameters of electromechanical devices such as lighting devices, escalators, and rolling shutters in real time through the Internet of Things protocol. After the collection, a hybrid networking method is used to transmit the multi-source heterogeneous data. Specifically: for fixed devices, industrial Ethernet is used for connection, for mobile inspection devices, 5G slice network is used for transmission, and key control signals are transmitted through a dedicated optical fiber network. After each subsystem control unit 10 preprocesses the collected data, data fusion is performed through the edge computing node, and a device status message in a unified format is generated and transmitted to the main control unit 11.

[0029] The main control unit 11 is used to generate a first collaborative control instruction set that matches the switch station control requirements according to the device switch state, fault signal and environmental parameters, in combination with the linkage rule library. The first collaborative control instruction set includes operation instructions for each electromechanical device, and corresponding target priorities are assigned to the operation instructions for each electromechanical device to obtain a second collaborative control instruction set.

[0030] Among them, the linkage rule library refers to a database that stores the device linkage logic, including the mapping relationship between device state conditions and corresponding control instructions. The target priority refers to the instruction execution priority level calculated according to the device importance, operation timeliness, and system load rate. The target priority can be low priority, medium priority, medium-high priority, high priority, etc. The first collaborative control instruction set refers to a set of original operation instructions that are initially screened and not sorted by priority. The second collaborative control instruction set refers to an executable instruction sequence after conflict elimination and timing optimization.

[0031] In the embodiment of the present application, after receiving the device status message, the main control unit 11 can traverse the linkage rule library based on the depth-first search algorithm; perform pattern matching through the rule engine to generate a first collaborative control instruction set containing device operation instructions, and then use the weighted priority algorithm to sort the instructions. Among them, the device importance coefficient can be determined by the analytic hierarchy process, and finally a second collaborative control instruction set sorted by the target priority is generated.

[0032] The communication module 12 is used to encapsulate the second collaborative control instruction set and push the second collaborative control instruction set to each subsystem control unit 10 by using a multi-channel parallel distribution mechanism.

[0033] Among them, multi-channel parallel distribution refers to a technology in which instructions of different service types are transmitted simultaneously through dedicated communication channels.

[0034] In the embodiment of the present application, the communication module 12 performs protocol encapsulation on the second collaborative control instruction set, and then adopts a multi-channel parallel distribution mechanism. Exemplarily, key control instructions are transmitted through a real-time control bus, batch status instructions are distributed through a message queue, and video monitoring instructions are transmitted through a streaming media private network. An independent quality of service policy is set for each channel. Among them, the real-time control bus adopts a token ring mechanism to ensure that the transmission delay is <50 ms, and the message queue sets priority sharding to ensure that high-priority instructions are dequeued first.

[0035] The data cache module 13 is used to store the operation status data of each electromechanical device during the process of the subsystem control unit 10 performing switch station operations based on the second collaborative control instruction set, identify the timing pattern using a machine learning model, predict the switch station linkage scenario based on the timing pattern, and load the corresponding control strategy.

[0036] Among them, the timing pattern refers to a combination of regular feature changes in the operation status of the device over time.

[0037] In the embodiment of the present application, the data cache module 13 uses a distributed database cluster to store the timing data of the device operation status, constructs a long short-term memory neural network model, and then analyzes the device action timing pattern. Feature vectors are extracted through a sliding time window to predict the linkage scenario that may be triggered within the next 5 minutes. When the prediction confidence level > 85%, the corresponding control strategy is automatically loaded into the memory pre-execution queue, shortening the actual response time.

[0038] The main control unit 11 is used to construct a digital twin model based on the operation status data and execution feedback data of each electromechanical device, dynamically evaluate the digital twin model, and then detect the response delay of the subsystem control unit 10. When the response delay exceeds the set threshold, a third collaborative control instruction set is generated according to the control strategy to optimize the switch station control timing.

[0039] Among them, the digital twin model refers to a virtual simulation model of a physical device that can map the device status in real time. The response delay refers to the time interval from the instruction being issued to the device's feedback. The control strategy refers to a set of device operation plans preset for specific scenarios. In the embodiment of the present application, the specific value of the threshold can be set. For example, the instruction delay for the rolling shutter door to open and close is 150 ms, and the emergency stop response delay for the escalator is 80 ms. The third collaborative control instruction set is a set of device operation instructions generated by the main control unit 11 according to the dynamically optimized control strategy after detecting that the subsystem response delay exceeds the threshold.

[0040] In the embodiment of the present application, the main control unit 11 constructs a digital twin model based on the three-dimensional point cloud data of the device, and uses the Kalman filter algorithm to fuse real-time operation data and historical data. The subsystem response delay is detected through time series analysis. When the delay exceeds the threshold for 3 consecutive sampling periods, the dynamic rescheduling algorithm is triggered to generate a third collaborative control instruction set considering the device load balance, which can reduce the multi-device collaborative error after optimization.

[0041] The following is a specific example: When the subway closes at the evening rush hour, the escalator system detects abnormal vibrations and uploads them to the main control unit in real time through the 5G network. The rule engine matches the linkage rule of "closing the adjacent turnstile when the escalator fails" and generates an instruction to give priority to closing the turnstile at Exit B. The communication module 12 preferentially transmits the turnstile control instruction through the controller area network bus, and at the same time distributes the lighting adjustment instruction through the distributed publish queue. The data cache module 13 monitors that the escalator is abnormal 5 times in a row, anticipates that the passenger flow diversion scenario will be triggered, and pre-loads the broadcast control strategy in advance. When the digital twin model detects that the response delay at Exit C reaches 250 ms, it dynamically adjusts the instruction timing, and finally completes the collaborative subway closing operation including multiple subsystems within 175 seconds.

[0042] The embodiment of the present application jointly realizes the efficient acquisition of multi-source heterogeneous data through distributed data collection and hybrid networking; generates accurate control instructions based on the rule engine and weighted algorithm; the multi-channel distribution mechanism ensures the reliability of instruction transmission; the timing pattern prediction improves the system response efficiency; the digital twin dynamic optimization ensures the system robustness. Therefore, the embodiment of the present application can overall shorten the switch station operation time, effectively reduce the device collaborative error, greatly improve the emergency response ability, and at the same time enhance the stability and adaptability of multi-system collaborative control.

[0043] In a possible embodiment, the data cache module 13 is used to store the operation state data of each electromechanical device during the switch station operation process when the subsystem control unit 10 executes the second collaborative control instruction set, identify the timing pattern using a machine learning model, anticipate the switch station linkage scenario based on the timing pattern and load the corresponding control strategy, including: The data cache module 13 is used to obtain the operation state data of each electromechanical device in real time when the subsystem control unit 10 executes the second collaborative control instruction set. The operation state data is recorded in the form of an operation event stream, and the operation event stream includes the device operation timestamp, device status code, and environmental parameters. The operation state data is stored in the buffer in ascending order of the timestamp, and the sliding window elimination operation is performed on the old data that exceeds the preset time threshold to form an operation event sequence pool.

[0044] Among them, the operation event stream refers to the device operation log recorded in chronological order, including triplets such as device operation timestamp, device status code, and environmental parameters. The sliding window elimination operation refers to a dynamic data elimination mechanism based on a time threshold, which retains the data in the latest time window.

[0045] In the embodiment of the present application, the data cache module 13 collects the operating status data of each electromechanical device in real time through the Internet of Things protocol, and writes the device operation timestamp, device status code, and environmental parameters into the circular buffer in ascending order of timestamp. A sliding window elimination algorithm is adopted, and the time threshold is set to 30 minutes. When new data is written, old data that exceeds the time range of 30 minutes is automatically eliminated. A mapping relationship between timestamps and storage locations is established through hash indexes to ensure a write throughput of 100,000 operation event streams per second.

[0046] The data cache module 13 is also used to extract overlapping data segments of adjacent time windows from the operation event sequence pool, and convert the device state encoding in each data segment into a state transition fragment; after feature splicing of the state transition fragment and the environmental parameters, the state transition fragment is input into the machine learning model to perform frequency statistics and correlation analysis of the cross-device state transition path, and output a time series pattern that represents the causal relationship between devices.

[0047] Among them, the state transition fragment refers to the matrix representation of the device state code sequence conversion, which describes the state change path. The state transition fragment is determined based on the state transition pair. Each state transition pair consists of the previous state code and the next state code, and is associated with the corresponding operation time interval to form a state transition fragment set. Frequency statistics refer to the quantitative statistics of the number of occurrences of the device state change path, which is used to identify high-frequency device linkage modes. Correlation analysis is to mine the hidden causal relationship between device state transitions and environmental parameters through machine learning models.

[0048] In the embodiment of the present application, the data cache module 13 extracts overlapping data segments of adjacent time windows at intervals of 5 minutes, and performs device state encoding conversion on each data segment: converting the device state encoding sequence into a state transfer matrix. The state transfer matrix and environmental parameters are spliced ​​into a 128-dimensional feature vector, and after the dimension is reduced to 32 dimensions using principal component analysis, it is input into a time series pattern mining model based on long short-term memory to calculate the correlation between state transfers between devices. The formula for the correlation is: correlation = co-occurrence frequency ÷ total number of events in the time window × transfer path weight.

[0049] The data cache module 13 is also used to match the timing pattern with the preset switch station linkage scenario topology structure, generate a candidate scenario queue sorted by priority according to the matching result, and select the control strategy corresponding to the candidate scenario with the highest priority.

[0050] Among them, the linkage scenario topology structure can refer to a pre-defined network diagram of device linkage relationships, stored in the form of an adjacency matrix. The candidate scenario queue can be a set of potential linkage scenarios sorted by priority for policy selection. The control policy refers to a set of device operation rules preset for a specific scenario.

[0051] In the embodiment of the present application, the data cache module 13 performs similarity matching between the time series pattern output by the long short-term memory neural network model and the pre-set switch station linkage scenario topology structure, and uses the cosine similarity algorithm to calculate the matching degree. The matching degree = the dot product of the time series pattern vector and the scenario vector divided by the product of the norms of the two. For scenarios with a matching degree exceeding 0.7, the weight = the device importance coefficient × the scenario matching degree. According to the device weights, a candidate scenario queue is generated, the control policy corresponding to the scenario with the highest weight value in the queue is selected, and the policy is pre-loaded into the memory execution queue.

[0052] The following is a specific example: When a certain subway station closes, the data cache module 13 receives the status codes and passenger flow data of the escalators at exits A - F in real time. After the sliding window eliminates the old data 10 minutes ago, the overlapping data segments within the past 30 minutes are extracted to generate a state transition matrix. The long short-term memory neural network model analyzes and discovers a time series pattern of "the passenger flow at exit C increases by 70% within 5 minutes after the escalator at exit B fails", and the matching degree with the pre-set "large passenger flow diversion" scenario topology reaches 0.85. The system generates a control policy to preferentially close exit B and open the standby channel at exit C, which is loaded and executed after weight sorting, shortening the station closing time.

[0053] The implementation of the present application realizes the efficient storage and dynamic update of massive time series data through the sliding window mechanism, accurately mines the implicit associations between devices by using state transition analysis and feature fusion technologies, generates optimized control policies through intelligent matching with the pre-set scenario topology, improves the real-time performance and accuracy of multi-device collaborative control, and enhances the system's adaptability to complex working conditions.

[0054] In a possible embodiment, the data cache module 13 is further configured to extract overlapping data segments of adjacent time windows from the operation event sequence pool, convert the device status codes in each data segment into state transition segments; after feature splicing the state transition segments with environmental parameters, input them into a machine learning model for frequency statistics and correlation analysis of cross-device state transition paths, and output a time series pattern representing the causal relationship between devices, including: The window intercepting unit is configured to intercept overlapping data segments of adjacent time windows from the operation event sequence pool according to a preset sliding step size. The overlapping data segments contain a sequence of device status codes. State transition pairs are constructed according to the sequence of device status codes in ascending order of time stamps. Each state transition pair consists of a previous state code and a subsequent state code, and is associated with the corresponding operation time interval to form a set of state transition segments.

[0055] Among them, the sliding step is the fixed time interval for each window slide. The overlapping data segment is the repeated time region between adjacent windows. The device status coding sequence is the structured device status identifier. The state transition pair is the transition relationship composed of two consecutive status codings. The operation time interval is the time difference between two state changes. The state transition segment set is the data set containing multiple state transition pairs and their time attributes.

[0056] In the specific example in the subway switch station scenario, the state transition pair can refer to the assumption that in a certain station opening operation, the device status changes in the following time sequence: lighting on at 10:00:00 → escalator start at 10:00:30 → rolling shutter door open at 10:01:15. The state transition pairs are lighting on → escalator start, with a time interval of 30 seconds, and escalator start → rolling shutter door open, with a time interval of 45 seconds.

[0057] The feature embedding unit is used to splice the features of the state transition segment and the environmental parameters within the same time window, and embed the environmental parameters into the preset bit field of the state transition segment during the feature splicing process to form a state transition vector.

[0058] Among them, the environmental parameters refer to environmental monitoring data such as temperature, humidity, and passenger flow. The preset bit field refers to the positions specifically reserved for environmental parameters in the feature vector, such as the 100th - 128th dimensions. The state transition vector is the structured data representation formed by fusing the device state transition segment with real-time environmental parameters, and is used to comprehensively depict the correlation features between the device state changes and environmental factors within a certain time period.

[0059] The analysis unit is used to input the state transition vector into a machine learning model, count the co-occurrence frequencies of the state transition vectors of different devices, analyze the time window overlap degree and interval correlation of the state transition segments of different devices, generate a tightness index set for cross-device state transitions, construct a weighted directed acyclic graph, where the nodes are device status codings, and the weight value of the edge is determined by the product of the co-occurrence frequency and the time window overlap degree, and output the time sequence pattern.

[0060] Among them, the time window overlap degree is the proportion of the overlapping part of two window times in the total window duration. The interval correlation is the statistical correlation of the time intervals of different device state changes. The tightness index set is the numerical value quantifying the degree of device state association, such as in the range of 0 - 1.

[0061] The following is a specific example: In the scenario of the subway station closing, the system intercepts the device status data during the time period from 22:00 to 22:05, and sets the sliding step length to 60 seconds. The window intercepting unit captures the state transition chain of "lighting off - escalator out of service - rolling shutter door failure", generates 3 groups of state transition pairs and records the time intervals. The feature embedding unit splices the environmental parameters into the state coding sequence to form the feature vector "110_205_308_0.3_25". The analysis unit, through the long short-term memory network model, identifies that the probability of the escalator going out of service within 5 minutes after the lighting is turned off reaches 82%, and it is strongly correlated with the low passenger flow environment in the evening. Based on this, the system automatically optimizes the station closing process: after the lighting is turned off, a forced additional escalator status review link is added, successfully improving the detection rate of device anomalies during the station closing operation.

[0062] In the implementation of this application, the temporal evolution characteristics of device status are dynamically captured through the sliding window mechanism, combined with multi-dimensional environmental parameters to construct a fused feature vector, and the potential correlation rules between devices are deeply mined using machine learning models. Compared with the traditional single-device independent monitoring mode, it can predict the risk of cross-device chain failures in advance and improve the accuracy of fault warning. The constructed weighted directed topological network can clearly present the influence conduction path of the core device status, providing a basis for priority decision-making for operation personnel, enabling them to focus on the collaborative handling of key device combinations, thereby systematically improving the device operation and maintenance efficiency and risk prevention and control capabilities.

[0063] In a possible embodiment, the analysis unit is used to input the state transition vector into a machine learning model, count the co-occurrence frequencies of the state transition vectors of different devices, and analyze the time window overlap degree and interval correlation of the state transition segments of different devices to generate a tightness index set for cross-device state transitions, including: The co-occurrence detection subunit is used to input the state transition vector into a machine learning model, traverse the state transition vectors of different devices, and identify the co-occurrence frequencies of the state transition vectors of different devices within the same time window to generate a co-occurrence frequency matrix of device pairs.

[0064] Among them, the co-occurrence frequency matrix refers to an N×N matrix that records the number of times the state transitions of each pair of devices occur simultaneously within the same time window. N is any integer.

[0065] In the embodiment of this application, the state transition vectors of devices such as lighting, escalators, and rolling shutter doors are extracted from the subway station device database, scanned through each time window, and the co-occurrence times of the state transitions of different devices within the same time window are counted. For example, within the station closing time window from 22:00 to 22:05, if "lighting off → escalator out of service" and "rolling shutter door closing failure" occur simultaneously, the co-occurrence count of this device pair is incremented by 1. Finally, an N×N co-occurrence frequency matrix is generated.

[0066] An overlap degree calculation subunit is used to calculate the time window overlap length of the state transition segments of two devices in equipment alignment, and determine the ratio of the time window overlap length to the total time window length as the time window overlap degree. Herein, the total time window length refers to the maximum duration of a single device state transition segment.

[0067] In an embodiment of the present application, for each device pair in the co-occurrence frequency matrix, the overlap degree calculation subunit extracts the time window information of the state transition segments of the two devices. Calculate the time window overlap length of the two devices, and divide the overlap length by the maximum value of the total window lengths of the two devices to obtain the time window overlap degree. For example, 3-minute overlap / 5-minute total length = 0.6, indicating that the time overlap degree between device A and B in this operation is 60%.

[0068] A correlation analysis subunit is used to extract the time interval sequence of the state transition between two devices in a device pair, and calculate the interval correlation according to the time interval sequence through the Pearson correlation coefficient.

[0069] Herein, the time interval sequence refers to the sequence composed of the time differences of the front and rear state changes in the device state transition pair. The Pearson correlation coefficient is a statistic that measures the degree of linear correlation between two variables, and its range is [-1, 1].

[0070] In an embodiment of the present application, for two devices in a device pair, the correlation analysis subunit extracts the time interval sequences of their respective state transitions. For example, the state transition time interval sequence of the lighting system is [30s, 45s, 60s], and that of the escalator system is [45s, 50s, 70s], and calculate the Pearson correlation coefficient between the two. The specific process is as follows: First, calculate the means μ_A and μ_B of the two sequences, and then calculate the covariance divided by the product of their standard deviations. If the product of their standard deviations = 0.8, it indicates that the state transition time intervals of the two devices are highly positively correlated.

[0071] A tightness index set construction subunit is used to combine the co-occurrence frequency matrix, the time window overlap degree, and the interval correlation to obtain a tightness index set.

[0072] Herein, in an embodiment of the present application, the tightness index set construction subunit combines the co-occurrence frequency matrix C_ij value, the time window overlap degree O_ij, and the interval correlation r_ij. The calculation formula is: tightness index T_ij = C_ij × O_ij × r_ij. For example, if the co-occurrence frequency C between device A and B is 5 times, the overlap degree O = 0.6, and the correlation r = 0.8, then T = 5 × 0.6 × 0.8 = 2.4. Finally, a set containing the tightness indexes of all device pairs is generated and sorted in descending order of the index values to form a priority processing list.

[0073] The following is a specific example: In the scenario of subway station closure, in the historical station closure data, the system discovers that "lighting off → escalator shutdown" and "failed rolling shutter door closure" co-occur in 80% of the station closure operations, and C_light - rolling shutter door = 32 times in the co-occurrence frequency matrix. Analyzing a certain station closure, the overlapping duration between the lighting off operation window from 22:00 to 22:05 and the rolling shutter door closure window from 22:02 to 22:07 is 3 minutes, and the overlap degree O = 3 / 5 = 0.6. Extracting the time interval sequences of the two, the average lighting operation interval is 40s, and the sequence is [35s, 45s, 40s], the average rolling shutter door interval is 75s, and the sequence is [60s, 80s, 85s]. Calculating gives r = 0.78. Calculating T = 32×0.6×0.78 = 15.0. This value is higher than the threshold of 10, so the system automatically inserts a forced inspection in the station closure process: after the escalator is shut down, the status of the rolling shutter door needs to be rechecked within 60 seconds, reducing the failure rate of the rolling shutter door.

[0074] The implementation of this application breaks through the limitations of traditional single-device monitoring by quantifying the spatio-temporal correlation characteristics of the state transfer between devices. Combining three-dimensional indicators of co-occurrence frequency, time overlap degree, and operation interval correlation, it accurately identifies device combinations prone to cascading failures. The constructed tightness index set can dynamically guide the allocation of operation and maintenance resources, improve the cross-device collaborative fault warning ability, and avoid operation accidents caused by device association failures.

[0075] In a possible embodiment, the main control unit 11 is configured to generate a first collaborative control instruction set that matches the switch station control requirements according to the device switch state, fault signal, and environmental parameters, in combination with the linkage rule library. The first collaborative control instruction set includes operation instructions for each electromechanical device. Corresponding target priorities are assigned to the operation instructions for each electromechanical device to obtain a second collaborative control instruction set, including: The rule activation subunit is configured to match the device switch state, fault signal, and environmental parameters with the rule trigger conditions in the linkage rule library, screen out an activation rule set that satisfies the consistency of the device state transition direction and the environmental parameters meet the threshold constraints, and extract the first collaborative control instruction set bound to the activation rule set. The rule trigger conditions are quantified switch station control requirements.

[0076] Among them, the consistency of the device state transition direction means that multiple devices need to perform the same type of operation synchronously, such as all closing or all starting. The environmental parameter threshold constraint refers to the condition limit triggered when environmental indicators such as temperature and humidity exceed the preset critical value. The activation rule set is a set of linkage rules triggered under specific device states and environmental conditions.

[0077] A priority weighting subunit is used to obtain the basic priorities of the operation instructions in the first collaborative control instruction set from a preset priority table according to the device type, and perform dynamic weighting on the basic priority values in combination with the fault signal level to generate a transition instruction set with target priority tags.

[0078] Among them, the preset priority table is an instruction execution priority level mapping table predefined according to the importance of the device. The dynamic weighting process is a calculation process for adjusting the priority according to the real-time fault level. The transition instruction set is an intermediate instruction set generated by the priority weighting subunit, and its characteristic is that it carries the dynamically calculated target priority tags, but the conflict adjudication and timing arrangement have not been completed. The transition instruction set is the intermediate state from the first collaborative control instruction set to the second collaborative control instruction set. The fault signal level is a quantitative grading of the severity of the device fault, which directly affects the dynamic weighting coefficient of the priority weighting subunit.

[0079] A conflict adjudication subunit is used to detect mutually exclusive operation instructions acting on the same device in the transition instruction set and retain the instruction with the highest priority to generate a conflict-free instruction set.

[0080] Among them, the mutually exclusive operation instructions are instructions with contradictory operation requirements for the same device. The conflict-free instruction set is an instruction set generated by the conflict adjudication subunit after eliminating conflicts in the transition instruction set. Its core feature is to eliminate the mutually exclusive operation instructions acting on the same device, ensure that each device only retains a single executable instruction, and this instruction has the highest execution priority in the current scenario.

[0081] A timing arrangement subunit is used to perform device-level timing arrangement on the conflict-free instruction set, and merge consecutive instructions according to the minimum operation interval time of the device to obtain the second collaborative control instruction set.

[0082] Among them, the device-level timing arrangement refers to the operation instructions for each subway device, such as escalators, rolling shutters, and lighting systems.

[0083] The following is a specific example: In the station shutdown process, when the system detects that the rolling door device status in a certain area is "operating", the fault signal level is secondary, and the temperature sensor in the environmental parameters shows that 35°C exceeds the threshold, the rule activation subunit traverses the linkage rule library and matches two rules: "Rolling door closing condition: the device status jump direction is the closing trend and the temperature < 30°C" and "Ventilation start condition: the temperature ≥ 30°C for 5 minutes". Since the current temperature parameter exceeds the threshold and the device status conforms to the closing direction, the activation rule set is screened out, including two first collaborative control instructions: "Forcibly close the rolling door" and "Start the exhaust system". The priority weighting subunit obtains the basic priority of 90 points for "rolling door control" and 70 points for "environmental control equipment" from the preset table, and combines the weighted coefficient of 1.2 for the secondary fault signal to generate a transition instruction set including a rolling door closing instruction with a priority of 108 points and a ventilation start instruction with a priority of 84 points. The conflict adjudication subunit discovers that there is a power load conflict between the "ventilation start" in the transition instruction set and the existing "lighting off" instruction for the same environmental control power distribution cabinet, and retains the rolling door closing instruction with a priority of 108 points according to the priority. The timing arrangement subunit combines the ventilation start instruction originally with an interval of 8 seconds and the subsequent drainage pump instruction into a second collaborative control instruction set executed at an interval of 10 seconds according to the requirement of the minimum operation interval of 10 seconds for environmental control equipment, and finally forms an ordered operation sequence of "Immediately close the rolling door and then start the exhaust and close the drainage after 10 seconds".

[0084] The implementation of this application realizes precise collaborative control in complex scenarios through a multi-level instruction processing mechanism. The rule activation module ensures that the operation logic strictly follows the preset safety policy, the priority weighting mechanism dynamically adapts to the severity of different faults, the conflict adjudication module effectively eliminates the logical contradictions between operation instructions, and the timing arrangement module systematically optimizes the instruction execution efficiency. Compared with the traditional single instruction trigger mode, it can reduce the risk of misoperation, comprehensively improve the response timeliness and collaborative control accuracy of multi-device linkage, and provide a higher level of automated safety guarantee for subway operation.

[0085] In a possible embodiment, the rule activation subunit is used to match the device switch status, fault signal, and environmental parameters with the rule trigger conditions in the linkage rule library, screen out the activation rule set that meets the consistency of the device status jump direction and the environmental parameters conform to the threshold constraint, and extract the first collaborative control instruction set bound to the activation rule set, including: The event coupling component is used to receive the jump direction identifier of the device switch status, the encoded stream of the fault signal, and the sampled stream of the environmental parameters, and generate a composite event data packet with environmental constraints in accordance with the timestamp alignment method.

[0086] Among them, the jump direction identifier is a binary identifier representing the mutation direction of the device state. For example, 0 indicates from off to on, and 1 indicates from on to off. The encoded stream is a sequence of fault codes encapsulated in the Type-Length-Value (TLV) format. The sampling stream refers to the time series data of environmental parameters periodically collected by the sensor. The composite event data packet is a structured event entity generated through multi-source data fusion and spatio-temporal alignment.

[0087] The rule penetration component is used to perform two-way verification on the composite event data packet and the rule trigger conditions in the linkage rule library, and filter out the activation rule set that simultaneously meets the consistency of the device state jump direction and the environmental parameters conforming to the threshold constraint. The two-way verification includes forward verification and reverse verification.

[0088] Among them, the two-way verification includes a dual verification mechanism of forward rule matching and reverse parameter verification. Timestamp alignment is achieved through a clock synchronization protocol to align the time dimensions of multi-source data. Forward verification focuses on the structural matching between the rule trigger conditions and the device state. Reverse verification emphasizes the adaptability verification of environmental parameters and dynamic thresholds.

[0089] The instruction synthesis component is used to extract the original operation instruction fragments bound to the activation rule set, eliminate mutually exclusive instructions according to the pre-check result of device operation conflicts, and merge and generate the first collaborative control instruction set.

[0090] Among them, the original operation instruction fragments are the raw materials for generating the first collaborative control instruction set. The original operation instruction fragments are extracted from the instruction fragment binding relationship in the rule library. The pre-check result of device operation conflicts is an intermediate product generated through dependency relationship modeling and conflict detection algorithms.

[0091] The following is a specific example: When a subway station starts one-key station shutdown, the event coupling component receives the escalator stop signal, the rolling shutter door closing signal, and the passenger flow data of the platform infrared sensor in real time, and generates a "station shutdown ready" composite event packet through time window alignment; the rule penetration component matches the "low passenger flow station shutdown rule", verifies that the passenger flow density < 10 people / m 2 and the device state is normal, and activates the "close device group A" rule set; the instruction synthesis component extracts the instruction chain of "close escalator - close rolling shutter door - turn off lighting", and pre-checks that the lighting should be turned off after the escalator stops completely, and generates a control instruction with optimized timing. Finally, the system completes the collaborative shutdown of 20 devices within 3 seconds.

[0092] The implementation of this application realizes the spatio-temporal consistency alignment of multi-source data through an event coupling mechanism to ensure the accurate correlation analysis of multi-dimensional information; uses a rule penetration component to perform forward and reverse two-way verification on composite events to enhance the logical completeness of rule trigger conditions and the matching degree of environmental constraints; combines a conflict pre-check algorithm in the instruction synthesis stage to systematically eliminate logical contradictions between multi-device operation instructions. Compared with the traditional single-level processing mode, this solution has higher response timeliness and operation reliability in complex scenarios and can effectively support the collaborative control requirements of a large-scale device group. In a possible embodiment, the communication module 12 is configured to encapsulate a jump direction identifier and push the second collaborative control instruction set to each subsystem control unit 10 by using a multi-channel parallel distribution mechanism, including: An instruction sharding unit is configured to shard the second collaborative control instruction set into regionalized instruction subsets according to the physical location distribution of devices, and each subset is encapsulated into an independent data packet carrying a device group code.

[0093] Among them, the physical location distribution of devices refers to the actual spatial coordinates and the affiliated area division of devices in the subway station. The regionalized instruction subset refers to a set of device control instructions grouped by physical regions. The device group code is a unique code composed of a region identifier and a device type.

[0094] A channel binding unit is configured to query a preset communication topology relationship according to the device group code and allocate a primary transmission channel identifier and a backup transmission channel identifier to each independent data packet to form a dual-channel bound data packet.

[0095] Among them, the primary transmission channel identifier refers to the communication path identifier preferentially used. The backup transmission channel identifier refers to the alternative communication path identifier enabled when the primary channel fails.

[0096] A redundant transmission unit is configured to synchronously send the dual-channel bound data packet through the primary and backup dual channels, and monitor the channel health status in real time during primary channel transmission. When it is detected that the data packet fails to be delivered within the device response timeout window, it automatically switches to the backup channel to continue transmission.

[0097] Among them, the channel health status refers to the communication quality status evaluated by indicators such as packet loss rate and latency. The device response timeout window refers to the preset maximum waiting time for the instruction to be delivered.

[0098] A delivery confirmation unit is configured to receive the data packet reception status code returned by each subsystem control unit 10, trigger the channel switching and retransmission of the corresponding block instruction subset for the device group code that is not completely received until all subsystem control units 10 return a reception success status code to complete the push of the second collaborative control instruction set.

[0099] Among them, the block instruction subset refers to an instruction data block unit divided by device groups.

[0100] The following is a specific example: When the subway executes the one-key station shutdown, the instruction slicing unit slices the instruction set of "turn off the lighting, lower the rolling shutter door, and stop the escalator" into 6 regionalized instruction subsets such as the "West Hall Lighting Group" and the "East Exit Rolling Shutter Door Group" according to the equipment location. The channel binding unit assigns the main channel and the backup channel to the "West Hall Lighting Group". When the redundant transmission unit synchronously sends instructions and detects that the packet loss rate of fiber optic channel A01 suddenly increases to 15%, it immediately switches to 5G channel G07 for continued transmission. After receiving the timeout status code of the East Exit Rolling Shutter Door Group, the delivery confirmation unit only retransmits the instructions of this group, and finally completes the shutdown operation of 436 devices in the whole station within 1.8 seconds.

[0101] The implementation of this application optimizes the transmission efficiency of large-scale device control instructions through instruction slicing, realizes high-reliability transmission guarantee based on dual-channel binding and health monitoring mechanisms, and the block retransmission mechanism greatly reduces redundant data transmission in local failure scenarios. In practical applications, in the scenario of concurrent control of multiple devices in the subway station, the delivery efficiency of the complete instruction set is improved, the channel switching response is rapid, the device control operation is stable and reliable, ensuring the safety and integrity of the full-process execution.

[0102] Figure 2 The flowchart of a switch station control method for a subway station provided by an embodiment of this application is as Figure 2 shown, and the method includes: S21. Collect the device switch status, fault signals, and environmental parameters through the subsystem control units distributedly deployed in the switch station control system, and transmit the device switch status, fault signals, and environmental parameters to the main control unit through a hybrid networking method.

[0103] S22. Through the main control unit, according to the device switch status, fault signals, and environmental parameters, combined with the linkage rule library, generate a first collaborative control instruction set that matches the switch station control requirements. The first collaborative control instruction set contains the operation instructions of each mechanical and electrical device, and assigns corresponding target priorities to the operation instructions of each mechanical and electrical device to obtain a second collaborative control instruction set.

[0104] S23. Encapsulate the second collaborative control instruction set through the communication module, and push the second collaborative control instruction set to each subsystem control unit by using a multi-channel parallel distribution mechanism.

[0105] S24. During the process of the subsystem control unit executing the switch station operation based on the second collaborative control instruction set through the data cache module, store the operation status data of each mechanical and electrical device, identify the timing pattern in combination with the machine learning model, predict the switch station linkage scenario based on the timing pattern, and load the corresponding control strategy.

[0106] S25. Based on the operation status data and execution feedback data of each electromechanical device, the main control unit constructs a digital twin model, dynamically evaluates the digital twin model, detects the response delay of the subsystem control unit, and when the response delay exceeds the set threshold, optimizes the control timing of the switchyard.

[0107] Furthermore, this embodiment can provide a one-key switchyard according to the above method by setting the switchyard inspection and execution processes and strategies, and setting the switchyard time, so as to complete the relevant operations of starting / stopping the station service at a single point and in one stop. Through the process configuration of the one-key switchyard, an automated management strategy for starting and stopping the station is realized, and the operation personnel cost is reduced. To ensure operation safety, the one-key switchyard operation has a secondary confirmation function, effectively solving the following existing problems: the subway opening and closing stations have long response times, low operation efficiency, and inevitable human errors or omissions. The existing manual operations have not effectively and accurately integrated the effective management of opening and closing stations, resulting in inflexible responses and daily operations of the subway operation, which is not conducive to the safety of subway operation.

[0108] Specifically, the embodiment of the present application can configure the one-key switchyard control method in a process. Before the opening operation, the content of the one-key opening operation instructions is popped up first. And display settings can also be performed, such as supporting the setting of whether the operation instructions are not displayed within a week. If this item is checked, this step will be skipped when the opening operation is performed next time this week.

[0109] The equipment self-inspection includes the following three aspects: self-inspection equipment configuration, equipment self-inspection, and abnormal situation handling. Among them, self-inspection equipment configuration: supports configuring the equipment that needs to be self-inspected during the opening of the station at the back-end management console, including lighting, escalators, rolling shutters, etc. Management can be understood as displaying / viewing / adding / deleting / modifying the information of the equipment. Equipment self-inspection can refer to supporting the user to check the equipment information and abnormal items on the large screen when performing the one-key opening of the station. Abnormal situation handling can refer to when an abnormality is found during the equipment self-inspection, it is necessary to support controlling the user's operation according to the abnormal level, that is, for events in an emergency state, the event needs to be processed first before performing the switchyard operation, otherwise the user is prompted that there is a non-emergency event and whether to perform the next operation.

[0110] Manual monitoring includes manual monitoring point configuration and manual monitoring. Among them, manual monitoring point configuration can refer to supporting the configuration of video points that need to be manually monitored at the back-end management console.

[0111] Manual monitoring can refer to after the equipment self-inspection is completed, entering the manual monitoring step, supporting the user to remotely view the real-time video of the video points on the large screen, and after confirmation, entering the next operation. After the manual monitoring is confirmed to be completed, the content of the linkage actions to be executed during the opening operation is displayed. The content of the linkage actions includes control operations of power lighting, platform screen doors, escalators, ticket gates, ticket vending machines, broadcasting equipment, environmental control equipment, etc.

[0112] Submit the site opening operation. A password can be entered. If the linked configuration actions are confirmed to be correct, enter the steps for executing the site opening action and enter the site opening password before the operation. Then execute the site opening. If the password is confirmed to be correct, the system automatically executes the device actions related to site opening and synchronously displays the execution results and log information on the large screen, including the execution results, execution actions, and execution times. If the execution fails, the specific reasons need to be displayed. The site opening execution actions include: playing the site opening broadcast information, turning on the lighting, starting the escalator, opening the rolling shutter door, and turning on the power of the large screen. Next, view the real-time video of the rolling shutter door. When the rolling shutter door opening action is executed, it supports viewing the real-time video of the corresponding video device on the large screen in real time to ensure that the rolling shutter door is successfully opened. The execution time of one-key site opening is controlled within 180s.

[0113] The specific description of one-key site closing is similar to that of one-key site opening and will not be elaborated here.

[0114] It should be noted that different subsystem control units can manage different devices such as lighting equipment, platform screen doors, escalators, turnstiles, ticket vending machines, broadcast equipment, and environmental control equipment. The main control unit is responsible for receiving and processing data from multiple subsystems, including data related to device switch states, faults, etc. It provides high-speed computing and decision-making capabilities and coordinates the control of multiple subsystems based on the set linkage rules. Each subsystem has an independent control unit for real-time data collection and task execution under the instructions of the main control unit. Data is transmitted between subsystems through a high-speed communication interface, supporting fast response and coordination between multiple systems. Communication module 12 can be a high-speed communication module that uses a low-latency communication protocol (such as fiber optic communication, 5G, industrial Ethernet, etc.) to achieve high-speed data transmission between the main control unit and each subsystem. Data processing can adopt a parallel processing mechanism to support multiple systems to receive and transmit control signals simultaneously, reducing the response latency. To reduce latency, a data caching module 13 is set in the system to cache the data of the subsystems and anticipate possible linkage requirements in advance, thereby accelerating the execution of instructions. The system analyzes historical data through machine learning algorithms to anticipate the triggering conditions for multi-system linkage and prepares control strategies in advance to shorten the actual response time.

[0115] Exemplarily, each subsystem collects data in real time and transmits it to the main control unit. The system analyzes, classifies, and judges the data. Based on pre-set linkage rules, the system identifies the linkage conditions between multiple subsystems. For example, in the subway linkage rules, when the camera detects a large passenger flow warning, the ventilation will be automatically increased and a pre-recorded broadcast will be played. When the escalator camera detects a passenger falling, the escalator operation will be automatically stopped. The main control unit generates control instructions based on the linkage rules and distributes them to each subsystem in real time through a high-speed communication module. All subsystems immediately execute the corresponding operations after receiving the instructions. The system obtains the feedback of the execution results of the subsystems through the high-speed communication module and monitors the linkage effect in real time. If any abnormality or delay is found, the system can automatically adjust the linkage strategy or take remedial measures.

[0116] Thus, the embodiments of the present application have the following advantages: (1) It can break through the previous decentralized management mode of switch station operations and integrate the operations of multiple systems such as signals, lighting, ventilation, and water pumps into one-key control. Through simple interface and button operations, the system can synchronously close or open multiple key subsystems, avoiding complex manual operations one by one, and is especially suitable for large subway stations or scenarios with multi-station linkages. This integrated control method greatly reduces the operation steps and improves the overall efficiency of subway station management.

[0117] (2) At the same time, it has high functional scalability, supports unlimited expansion of data scale, and has strong adaptability to the scale development needs of subway station expansion, line expansion, and line network.

[0118] (3) Traditional subway station emergency shutdown relies on multi-level instruction transmission, and the processing process is relatively slow, especially prone to unnecessary delays in emergencies. However, the present invention can quickly respond in case of emergencies through centralized control and one-key operation functions, greatly shortening the time for station shutdown or isolation area. At the same time, the pre-set emergency shutdown plan of the system can complete data judgment and analysis before operation to ensure automatic response in the shortest time.

[0119] (4) Especially for the large-scale and real-time data collection and processing in the station, the microservices architecture is adopted. Based on real-time communication middleware and real-time data distribution technology, high-real-time and large-scale data collection can be achieved. A collection cycle of 500 ms can complete the processing of a data scale of millions, as well as the integration of multiple systems, complete data fusion, and achieve interconnection and interoperability between systems. The above real-time communication middleware is a software architecture designed to support real-time data exchange and message passing.

[0120] (5) The present application uses the method of parallel data processing and communication to achieve the simultaneous linkage of multiple subsystems, reducing the delay caused by the sequential execution of traditional single instructions and improving the overall system response speed.

[0121] (6) The system analyzes historical data through machine learning algorithms, anticipates possible linkage requirements of multiple systems in advance, and caches and preprocesses them in the main control unit. When the conditions are met, the system can respond quickly to achieve instant operations. This prediction mechanism effectively reduces the latency generated during the linkage process.

[0122] Figure 2 The described switch station control method for a subway station can be achieved through Figure 1 the switch station control system for a subway station described in the illustrated embodiment. Its implementation principle and technical effects will not be elaborated further. For the switch station control system for a subway station in the above embodiment, the specific ways in which each module and unit perform operations have been described in detail in the embodiment related to the method, and will not be elaborated here.

[0123] In a possible design, Figure 1 the switch station control system for a subway station in the illustrated embodiment can be implemented as a computing device, such as Figure 3 shown. The computing device can include a storage component 31 and a processing component 32.

[0124] The storage component 31 stores one or more computer instructions, where the one or more computer instructions are for the processing component 32 to call and execute.

[0125] The processing component 32 is used for: The subsystem control units distributedly deployed within the switch station control system are used to collect device switch states, fault signals, and environmental parameters, and transmit the device switch states, fault signals, and environmental parameters to the main control unit through a hybrid networking method; the main control unit is used to generate a first collaborative control instruction set that matches the switch station control requirements based on the device switch states, fault signals, and environmental parameters, in combination with the linkage rule library. The first collaborative control instruction set includes operation instructions for each electromechanical device, and assigns corresponding target priorities to the operation instructions for each electromechanical device to obtain a second collaborative control instruction set; the communication module is used to encapsulate the second collaborative control instruction set and push the second collaborative control instruction set to each subsystem control unit using a multi-channel parallel distribution mechanism; the data cache module is used to store the operation state data of each electromechanical device during the process of the subsystem control unit performing switch station operations based on the second collaborative control instruction set, identify time series patterns using a machine learning model, anticipate switch station linkage scenarios based on the time series patterns, and load corresponding control strategies; the main control unit is used to construct a digital twin model based on the operation state data and execution feedback data of each electromechanical device, dynamically evaluate the digital twin model, and then detect the response delay of the subsystem control unit. When the response delay exceeds the set threshold, generate a third collaborative control instruction set according to the control strategy to optimize the switch station control timing.

[0126] Among them, the processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above methods. Of course, the processing component 32 may also be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components for executing the above methods.

[0127] The storage component 31 is configured to store various types of data to support the operation of the terminal. The storage component 31 may be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as random access memory (RAM), static random access memory (SRAM), electrically erasable programmable read only memory (EEPROM), erasable programmable read only memory (EPROM), programmable read only memory (PROM), read only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0128] Of course, the computing device may also necessarily include other components, such as input / output interfaces, display components, communication components, etc.

[0129] The input / output interface provides an interface between the processing component 32 and the peripheral interface module, and the above peripheral interface module may be an output device, an input device, etc.

[0130] The communication component is configured to facilitate communication between the computing device and other devices in a wired or wireless manner, etc.

[0131] Among them, the computing device may be a physical device or an elastic computing host provided by a cloud computing platform, etc. At this time, the computing device may refer to a cloud server, and the above processing component 32, storage component 31, etc. may be basic server resources leased or purchased from a cloud computing platform.

[0132] The embodiments of the present application also provide a computer storage medium storing a computer program, which can implement the above-mentioned Figure 2 switching station control method of a subway station shown in the embodiments.

[0133] 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, devices, and units can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0134] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative efforts.

[0135] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the essence of the above technical solution, or the part that contributes to the prior art, can be embodied in the form of a software product, which can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0136] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application and are not intended to limit them. Although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments or equivalently replace some of the technical features. These modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A switch station control system for a subway station, characterized in that, Including: Sub-system control units distributedly deployed within the switch station control system, which are used to collect device switch states, fault signals and environmental parameters, and transmit the device switch states, fault signals and environmental parameters to the main control unit through a hybrid networking method; The main control unit is used to generate a first collaborative control instruction set that matches the switch station control requirements according to the device switch states, fault signals and environmental parameters, in combination with the linkage rule library. The first collaborative control instruction set contains operation instructions for each electromechanical device, and corresponding target priorities are assigned to the operation instructions for each electromechanical device to obtain a second collaborative control instruction set; The communication module is used to encapsulate the second collaborative control instruction set and push the second collaborative control instruction set to each sub-system control unit by using a multi-channel parallel distribution mechanism; The data cache module is used to store the operation status data of each electromechanical device during the process of the sub-system control unit performing switch station operations based on the second collaborative control instruction set, identify time series patterns by using a machine learning model, predict switch station linkage scenarios based on the time series patterns, and load corresponding control strategies; The main control unit is used to construct a digital twin model based on the operation status data and execution feedback data of each electromechanical device, dynamically evaluate the digital twin model, detect the response delay of the sub-system control unit, and when the response delay exceeds the set threshold, generate a third collaborative control instruction set according to the control strategy to optimize the switch station control timing.

2. The system according to claim 1, wherein The data cache module is used to store the operation status data of each electromechanical device during the process of the sub-system control unit performing switch station operations based on the second collaborative control instruction set, identify time series patterns by using a machine learning model, predict switch station linkage scenarios based on the time series patterns, and load corresponding control strategies, including: The data cache module is used to obtain the operation status data of each electromechanical device in real time when the sub-system control unit executes the second collaborative control instruction set. The operation status data is recorded in the form of an operation event stream. The operation event stream contains device operation timestamps, device status codes and environmental parameters, stores the operation status data in the buffer in ascending order of timestamps, and performs a sliding window elimination operation on the old data that exceeds the preset time threshold to form an operation event sequence pool; The data cache module is also used to extract overlapping data segments of adjacent time windows from the operation event sequence pool, convert the device status codes in each data segment into state transition segments; after feature splicing the state transition segments and the environmental parameters, input them into a machine learning model for frequency statistics and correlation analysis of cross-device state transition paths, and output a time series pattern representing the causal relationship between devices; The data cache module is also used to match the time series pattern with the pre-set switch station linkage scenario topology structure, generate a candidate scenario queue sorted by priority according to the matching result, and select the control strategy corresponding to the candidate scenario with the highest priority.

3. The system according to claim 2, wherein The data caching module is further configured to extract overlapping data segments of adjacent time windows from the operation event sequence pool, convert the device state encodings in each data segment into state transition segments; after feature splicing the state transition segments with the environmental parameters, input them into a machine learning model for frequency statistics and correlation analysis of cross-device state transition paths, and output a timing pattern representing the causal relationship between devices, including: A window intercepting unit, configured to intercept overlapping data segments of adjacent time windows from the operation event sequence pool according to a preset sliding step length, where the overlapping data segments include a device state encoding sequence, construct state transition pairs according to the device state encoding sequence in ascending order of time stamps, and each state transition pair consists of a previous state encoding and a subsequent state encoding, and associate a corresponding operation time interval to form a state transition segment set; A feature embedding unit, configured to perform feature splicing on the state transition segments and the environmental parameters within the same time window, and embed the environmental parameters into a preset bit field of the state transition segments during the feature splicing process to form state transition vectors; An analysis unit, configured to input the state transition vectors into a machine learning model, count the co-occurrence frequencies of the state transition vectors of different devices, analyze the time window overlap degree and interval correlation of the state transition segments of different devices, generate a tightness index set for cross-device state transitions, construct a weighted directed acyclic graph, where the nodes are the device state encodings, and the weight value of the edge is determined by the product of the co-occurrence frequency and the time window overlap degree, and output a timing pattern.

4. The system according to claim 3, wherein The analysis unit is configured to input the state transition vectors into a machine learning model, count the co-occurrence frequencies of the state transition vectors of different devices, analyze the time window overlap degree and interval correlation of the state transition segments of different devices, generate a tightness index set for cross-device state transitions, including: A co-occurrence detection sub-unit, configured to input the state transition vectors into a machine learning model, traverse the state transition vectors of different devices, and identify the co-occurrence frequencies of the state transition vectors of different devices within the same time window to generate a co-occurrence frequency matrix of device pairs; An overlap degree calculation sub-unit, configured to calculate the time window overlap length of the state transition segments of two devices in a device pair, and determine the ratio of the time window overlap length to the total length of the time window as the time window overlap degree; A correlation analysis sub-unit, configured to extract the time interval sequence of state transitions between two devices in a device pair, and calculate the interval correlation according to the time interval sequence through the Pearson correlation coefficient; A tightness index set construction sub-unit, configured to combine the co-occurrence frequency matrix, the time window overlap degree, and the interval correlation to obtain a tightness index set.

5. The system according to claim 1, characterized in that, The main control unit is configured to generate a first collaborative control instruction set that matches the switchyard control requirements according to the device switch state, fault signal, and environmental parameters, in combination with the linkage rule library. The first collaborative control instruction set includes operation instructions for each electromechanical device, and assigns corresponding target priorities to the operation instructions for each electromechanical device to obtain a second collaborative control instruction set, including: A rule activation subunit, configured to match the device switch state, fault signal, and environmental parameters with the rule trigger conditions in the linkage rule library, filter out an activation rule set that meets the consistency of the device state jump direction and the environmental parameters meet the threshold constraints, and extract the first collaborative control instruction set bound to the activation rule set, where the rule trigger conditions are quantified switch station control requirements; A priority weighting subunit, configured to obtain the basic priorities of the operation instructions in the first collaborative control instruction set from a preset priority table according to the device type, and perform dynamic weighting processing on the basic priority values in combination with the fault signal level to obtain a transition instruction set with target priority tags; A conflict adjudication subunit, configured to detect mutually exclusive operation instructions acting on the same device in the transition instruction set, and retain the highest priority instruction to generate a conflict-free instruction set; A timing arrangement subunit, configured to perform device-level timing arrangement on the conflict-free instruction set, and merge consecutive instructions according to the minimum operation interval time of the device to obtain a second collaborative control instruction set.

6. The system according to claim 5, wherein The rule activation subunit, configured to match the device switch state, fault signal, and environmental parameters with the rule trigger conditions in the linkage rule library, filter out an activation rule set that meets the consistency of the device state jump direction and the environmental parameters meet the threshold constraints, and extract the first collaborative control instruction set bound to the activation rule set, including: An event coupling component, configured to receive the jump direction identifier of the device switch state, the encoded stream of the fault signal, and the sampled stream of the environmental parameters, and generate a composite event data packet with environmental constraints in accordance with the timestamp alignment method; A rule penetration component, configured to perform two-way verification on the composite event data packet and the rule trigger conditions in the linkage rule library, and filter out an activation rule set that simultaneously meets the consistency of the device state jump direction and the environmental parameters meet the threshold constraints, where the two-way verification includes forward verification and reverse verification; An instruction synthesis component, configured to extract the original operation instruction fragments bound to the activation rule set, eliminate mutually exclusive instructions according to the device operation conflict pre-check result, and merge them to generate a first collaborative control instruction set.

7. The system according to claim 1, wherein The communication module, configured to encapsulate the second collaborative control instruction set, and push the second collaborative control instruction set to each subsystem control unit by using a multi-channel parallel distribution mechanism, including: An instruction sharding unit, configured to slice the second collaborative control instruction set into regionalized instruction subsets according to the physical location distribution of the devices, and encapsulate each subset into an independent data packet carrying the device group code; A channel binding unit, configured to query the preset communication topology relationship according to the device group code, and assign a main transmission channel identifier and a standby transmission channel identifier to each independent data packet to form a dual-channel binding data packet; A redundant transmission unit, configured to synchronously send the dual-channel binding data packet through the main and standby dual channels, and monitor the channel health status in real time during the main channel transmission. When it is detected that the data packet has not been delivered within the device response timeout window, it automatically switches to the standby channel to continue the transmission; A delivery confirmation unit, configured to receive the data packet reception status codes returned by each subsystem control unit, trigger channel switching retransmission of the corresponding subset of blockification instructions for the device group encoding that is not completely received until all subsystem control units return a reception success status code, so as to complete the push of the second collaborative control instruction set.

8. A method for controlling a switching station of a subway station, characterized in that, It includes: Collect the device switch status, fault signals and environmental parameters through the subsystem control units distributed in the switch station control system, and transmit the device switch status, fault signals and environmental parameters to the main control unit through a hybrid networking method; Based on the device switch status, fault signals and environmental parameters, the main control unit generates a first collaborative control instruction set that matches the switch station control requirements in combination with the linkage rule library. The first collaborative control instruction set includes operation instructions for each electromechanical device, and corresponding target priorities are assigned to the operation instructions for each electromechanical device to obtain a second collaborative control instruction set; The second collaborative control instruction set is encapsulated by a communication module, and the second collaborative control instruction set is pushed to each subsystem control unit by using a multi-channel parallel distribution mechanism; During the process of the subsystem control unit performing switch station operations based on the second collaborative control instruction set, a data cache module stores the operation status data of each electromechanical device, identifies the timing pattern in combination with a machine learning model, anticipates the switch station linkage scenario based on the timing pattern, and loads the corresponding control strategy; Based on the operation status data and execution feedback data of each electromechanical device, the main control unit constructs a digital twin model, dynamically evaluates the digital twin model, and then detects the response delay of the subsystem control unit. When the response delay exceeds the set threshold, the switch station control timing is optimized.

9. A computing device, characterized in that, It includes a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a switch station control method for a subway station as described in claim 8.

10. A computer storage medium, characterized in that, A computer program is stored, and when the computer program is executed by a computer, it implements a switch station control method for a subway station as described in claim 8.

Images