Method and device for identifying network state of train levitation control system and related product

Abstract

This application discloses a method, apparatus, and related products for identifying the network status of a train levitation control system. By real-time collection of environmental data such as altitude, temperature, air pressure, electromagnetic interference intensity, and geographical location, the operating parameters of the monitoring probes are dynamically adjusted to adapt to complex and diverse operating conditions. Based on the adapted operating parameter response baseline, the operating data of the maglev train is analyzed to obtain the network status of the maglev train. This effectively avoids the problems of misjudgment and missed judgment caused by environmental fluctuations due to a fixed baseline. Simultaneously, it can accurately distinguish between environmental interference and malicious network attacks, improving the accuracy and reliability of network status identification and ensuring the continuity and stability of network security monitoring of the maglev train levitation control system under extreme conditions such as high altitudes, tunnels, and strong electromagnetic interference.

Description

Technical Field

[0001] This application relates to the field of network security technology, and in particular to a method, apparatus and related products for identifying the network status of a train suspension control system. Background Technology

[0002] With the rapid development of rail transit technology, the requirements for network security in the levitation control system of plateau maglev trains are constantly increasing. Traditional rail transit network security monitoring solutions are mostly designed for conventional wheel-rail maglev trains, using fixed operating parameter response baselines, which are difficult to adapt to the complex operating conditions of plateau maglev trains.

[0003] In existing technologies, rail transit network security monitoring solutions rely on static thresholds, preset network attack feature databases, and fixed baselines for monitoring. These solutions are prone to monitoring failure when location is lost or wireless communication is interrupted, making it difficult to cope with complex operating conditions and thus affecting the safe and stable operation of the maglev train suspension control system. Summary of the Invention

[0004] Based on the above problems, this application provides a method, device and related products for identifying the network status of a train suspension control system, so as to introduce environmental data, accurately distinguish between malicious network attacks and environmental interference, and solve the problems of poor adaptability and poor accuracy of existing technologies.

[0005] This application discloses a method for identifying the network status of a train suspension control system, the method comprising: The operating parameter response baseline of the monitoring probe is adjusted in real time based on the environmental data collected in real time. The environmental data includes one or more of the following: altitude, temperature, air pressure, electromagnetic interference intensity, and geographical location of the environment in which the maglev train suspension control system is located. The monitoring probe is set on the suspension control system to collect the operating data of the maglev train in real time. The operating parameter response baseline includes probe sensing sensitivity, sampling frequency, data filtering rules, and signal preprocessing threshold. At the same time, based on the adjusted operating parameter response baseline, the operating data of the suspension control system is collected; the operating data includes suspension gap, electromagnet current, vehicle acceleration, network latency, and packet loss rate. The network status of the suspension control system is obtained by analyzing the operational data.

[0006] Optionally, the real-time adjustment of the baseline response of the monitoring probe's operating parameters based on real-time acquired environmental data includes: Based on the environmental data, the operating route of the maglev train is divided into different sections; The operating parameter response baseline is adjusted differently depending on the segment.

[0007] Optionally, the step of differentially adjusting the operating parameter response baseline according to the different segments includes: In the case that the section is a tunnel section, the operating parameter response baseline is first adjusted; the first adjustment includes increasing network latency tolerance and / or reducing remote command trust weight; When the section is a high-altitude section, a second adjustment is made to the operating parameter response baseline; the second adjustment includes one or more of the following: reducing the sensing sensitivity in the operating parameter response baseline, adjusting the relationship between electromagnetic force and current, and adjusting the relationship between suspension gap and acceleration; In the case that the section is a strong electromagnetic interference section, a third adjustment is made to the baseline of the operating parameter response; the third adjustment includes reducing the sensitivity of network anomaly detection; When the section is an open area, adjust the operating parameters to the baseline to the initial state.

[0008] Optionally, analyzing the operational data to obtain the network status of the suspension control system includes: The operational data is identified using the adjusted operating parameters and response baseline to obtain the anomaly confidence level of the suspension control system; The network state is determined based on the value of the anomaly confidence level.

[0009] Optionally, determining the network state based on the value of the anomaly confidence score includes: If the anomaly confidence level is lower than a first threshold, it is determined that the network state is subject to environmental interference. If the abnormal confidence level is between the first threshold and the second threshold, the network state is determined to be suspected of being under network attack; the second threshold is greater than the first threshold. If the anomaly confidence level is higher than the second threshold, the network state is determined to be under network attack.

[0010] Optionally, after obtaining the network status of the suspension control system, the method further includes: If the network is experiencing environmental interference, record the abnormal information of the network status. If the network status indicates a suspected network attack, limit the rate of the affected ports and increase the sampling frequency of the monitoring probes; If the network status indicates a network attack, the suspension control system will be set to local security mode, the speed of the maglev train will be limited, and the train will be guided to a stop.

[0011] Optionally, the environmental data includes the geographical location. After adjusting the baseline of the monitoring probe's operating parameters in real time based on the real-time collected environmental data, the method further includes: Obtain the geolocation tags from each frame of instructions received by the suspension control system; If the geographic location tag matches the geographic location, the instruction is determined to be a trusted instruction; If the geographic location tag does not match the geographic location, the instruction is determined to be a suspicious instruction.

[0012] Based on the above-mentioned method for identifying the network status of a train suspension control system, this application also discloses a device for identifying the network status of a train suspension control system, comprising: an adjustment unit, a monitoring unit, and an analysis unit. The adjustment unit is used to adjust the working parameter response baseline of the monitoring probe in real time based on the environmental data collected in real time. The environmental data includes one or more of the following: altitude, temperature, air pressure, electromagnetic interference intensity, and geographical location of the environment in which the maglev train suspension control system is located. The monitoring probe is set on the suspension control system and is used to collect the operating data of the maglev train in real time. The working parameter response baseline includes probe sensing sensitivity, sampling frequency, data filtering rules, and signal preprocessing threshold. The monitoring unit is used to collect the operating data of the suspension control system at the same time based on the adjusted operating parameters response baseline; the operating data includes suspension gap, electromagnet current, vehicle acceleration, network latency, and packet loss rate. The analysis unit is used to analyze the operating data to obtain the network status of the suspension control system.

[0013] Optionally, the adjustment unit includes: The sub-units are used to divide the operating route of the maglev train into different sections based on the environmental data. The adjustment subunit is used to differentiate the operating parameter response baseline according to the different sections.

[0014] Optionally, the adjustment subunit includes: A first adjustment subunit is configured to perform a first adjustment on the operating parameter response baseline when the segment is a tunnel segment; the first adjustment includes increasing network latency tolerance and / or reducing remote command trust weight. The second adjustment subunit is used to make a second adjustment to the operating parameter response baseline when the section is a high-altitude section; the second adjustment includes one or more of the following: reducing the sensing sensitivity in the operating parameter response baseline, adjusting the relationship between electromagnetic force and current, and adjusting the relationship between suspension gap and acceleration. The third adjustment subunit is used to make a third adjustment to the operating parameter response baseline when the section is a strong electromagnetic interference section; the third adjustment includes reducing the sensitivity of network anomaly detection; The recovery subunit is used to adjust the operating parameter response baseline to its initial state when the section is an open area section.

[0015] Optionally, the analysis unit includes: An identification subunit is used to identify the operating data based on the adjusted operating parameters response baseline to obtain the anomaly confidence level of the suspension control system. A sub-unit is defined to determine the network state based on the value of the anomaly confidence level.

[0016] Optionally, the determining subunit includes: An interference determination subunit is used to determine that environmental interference exists in the network state when the anomaly confidence level is lower than a first threshold. An attack suspicion subunit is used to determine that the network state is suspected of being under a network attack when the abnormal confidence level is between the first threshold and the second threshold; the second threshold is greater than the first threshold. An attack determination subunit is used to determine that the network state is under attack when the anomaly confidence level is higher than the second threshold.

[0017] Optionally, the device further includes: A recording unit is used to record abnormal information about the network state when the network state is subject to environmental interference. A limiting unit is used to limit the rate of the affected ports and increase the sampling frequency of the monitoring probe when the network status is suspected of being under network attack. The setting unit is used to set the suspension control system to a local security mode, limit the speed of the maglev train, and guide it to stop when the network status indicates a network attack.

[0018] Optionally, the environmental data includes the geographical location, and the device further includes: The tag acquisition unit is used to acquire the geographic location tags in each frame of instructions received by the suspension control system; The instruction credibility determination unit is used to determine that the instruction is a credible instruction when the geographic location tag matches the geographic location; The instruction suspicion determination unit is used to determine that the instruction is a suspicious instruction when the geographic location tag does not match the geographic location.

[0019] Based on the above-described method for identifying the network status of a train suspension control system, this application also discloses a computer program product, which includes a computer program that, when run by a processor, is used to implement the above-described method.

[0020] Based on the above-described method for identifying the network status of a train suspension control system, this application also discloses a storage medium for storing computer program instructions, which, when executed by a central processing unit, are used to implement the above-described method.

[0021] Based on the above-mentioned method for identifying the network status of a train suspension control system, this application also discloses a maglev train, including an environmental sensing device, a processor, a monitoring probe, and a storage device, for implementing the above-mentioned method.

[0022] This application discloses a method, apparatus, and related products for network status identification of a train levitation control system. Lightweight monitoring probes are deployed at the perception, network, and application layers of the levitation control system, achieving physical integration of monitoring nodes and the control closed loop. By collecting environmental data such as altitude, temperature, air pressure, electromagnetic interference intensity, and geographical location in real time, the operating parameter response baseline of the monitoring probes is dynamically adjusted, allowing the monitoring standards to adaptively adapt to complex and diverse operating conditions. Based on the adapted operating parameter response baseline, the operating data of the maglev train is analyzed to obtain the network status of the maglev train. This effectively avoids the problems of misjudgment and missed judgment caused by environmental fluctuations due to fixed baselines. Simultaneously, it can accurately distinguish between environmental interference and malicious network attacks, improving the accuracy and reliability of network status identification and ensuring the continuity and stability of network security monitoring of the maglev train levitation control system under extreme conditions such as high altitudes, tunnels, and strong electromagnetic interference. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of this application. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0024] Figure 1 This is a flowchart illustrating a method for identifying the network status of a train suspension control system disclosed in an embodiment of this application. Figure 2 This is an architecture diagram of a train suspension control system network status identification disclosed in an embodiment of this application; Figure 3 This is a flowchart illustrating another method for identifying the network status of a train suspension control system disclosed in an embodiment of this application. Figure 4 This is a logic diagram of network status identification for a train suspension control system disclosed in an embodiment of this application; Figure 5 This is a timing diagram of the hierarchical response disclosed in an embodiment of this application; Figure 6 This is a schematic diagram of the working parameter response baseline update process disclosed in the embodiments of this application; Figure 7 This is a schematic diagram of the structure of a device for identifying the network status of a train suspension control system disclosed in an embodiment of this application. Detailed Implementation

[0025] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0026] Example 1: This application discloses a method for identifying the network status of a train suspension control system.

[0027] For details, please refer to Figure 1 The method for identifying the network status of a train suspension control system disclosed in this embodiment includes the following steps: Step 101: Adjust the baseline of the monitoring probe's operating parameters in real time based on the environmental data collected in real time.

[0028] Figure 2 This is an architecture diagram of a network state recognition system for a maglev train suspension control system disclosed in an embodiment of this application, as shown below. Figure 2 As shown, the method in this embodiment deploys lightweight monitoring probes in the sensing layer, network layer, and application layer of the maglev train suspension control system, respectively, realizing the physical integration of monitoring nodes and control closed loop.

[0029] As a feasible solution, the physical layer comprises the levitation control system itself, which may include physical sensing objects such as levitation electromagnets and levitation gaps. The sensing layer deploys monitoring probes to collect key physical quantities such as levitation gaps, vehicle acceleration, and electromagnet current in real time, obtaining the maglev train's operational data. These monitoring probes may include one or more types of gap sensor probes, acceleration sensor probes, and current sensor probes. The network layer deploys multi-source data fusion and traffic monitoring probes, as well as an industrial real-time Ethernet protocol parsing module, used to collect communication traffic, monitor network status (latency, packet loss rate), and perform field-level parsing of the protocol to establish a statistical distribution model of communication delay under different operating conditions and dynamically adjust the timing deviation tolerance. The application layer deploys a command-response verification module, a dynamic threshold / baseline correction module, and an anomaly detection unit. It constructs physical consistency verification rules that integrate electromagnetic force response characteristics and gap-current coupling relationships, and adjusts the dynamic threshold / baseline based on environmental data to achieve control command validity verification, anomaly detection, and safety decision output.

[0030] In the method of this embodiment, the environmental perception module can integrate an altitude / barometric pressure sensor and an electromagnetic interference sensor to collect environmental data in real time. The geographic location fusion unit can use "wheel speed sensor + inertial navigation + map" fusion positioning, which can provide continuous geographic location tags. It also supports dividing the maglev train's operating route into different sections and preset dynamic operating parameter response baselines. Furthermore, it uses a lightweight anomaly detection algorithm to perform local multi-source data fusion analysis, providing a basis for dynamic adjustment of the baseline.

[0031] In the method of this embodiment, the environmental perception module can collect environmental data of the maglev train suspension control system in real time, and the environmental data may specifically include one or more of altitude, temperature, air pressure, electromagnetic interference intensity, and geographical location.

[0032] In this embodiment, the monitoring probe is deployed inside the maglev train's suspension control system, enabling it to collect various operational data in real time during the maglev train's operation. This includes key parameters such as suspension gap, electromagnet current, vehicle acceleration, network latency, and packet loss rate, providing fundamental data support for network security monitoring. Based on the collected environmental data, such as altitude, temperature, air pressure, electromagnetic interference intensity, and geographical location, the maglev train's current operating route can be divided into sections. The overall operating route is then divided into different types of sections according to environmental characteristics. Furthermore, based on the environmental characteristics of different sections, the baseline response parameters of the monitoring probe are adjusted differentiated and specifically, allowing the monitoring standards to dynamically adapt to changes in the external environment, thus improving the environmental adaptability and accuracy of the monitoring.

[0033] For example, when the current segment is identified as a tunnel segment, a first type of baseline adjustment can be performed, including increasing network latency tolerance and reducing remote command trust weight, to adapt to the conditions of wireless signal attenuation and increased communication latency within the tunnel. When the current segment is identified as a high-altitude segment, a second type of baseline adjustment can be performed, including reducing sensing sensitivity, adjusting the electromagnetic force-current mapping relationship, and adjusting one or more of the suspension gap-acceleration coupling relationship, to match the physical characteristics differences caused by low air pressure and air density changes at high altitudes. When the current segment is identified as a strong electromagnetic interference segment, a third type of baseline adjustment can be performed, namely, reducing the network anomaly detection sensitivity to avoid normal signal fluctuations caused by electromagnetic interference being misjudged as network anomalies. When the current segment is identified as an open area segment, the operating parameter response baseline can be restored to the initial reference state to ensure the accuracy of monitoring under normal operating conditions.

[0034] In the method of this embodiment, it should be noted that the above-mentioned segment division method and corresponding baseline adjustment strategy are illustrative examples and not the only limitation. Those skilled in the art can reasonably extend, replace or combine them according to the actual route environment and monitoring needs, all of which fall within the protection scope of this invention.

[0035] Step 102: At the same time, based on the adjusted operating parameter response baseline, collect the operating data of the suspension control system.

[0036] In this embodiment, after real-time dynamic adjustment of the operating parameter response baseline, the system simultaneously collects the maglev train's operating data based on the adjusted response baseline and performs comprehensive, multi-dimensional analysis and identification of the operating data collected in real time by the monitoring probes. The operating data covers multiple key parameters such as network latency, packet loss rate, command frame integrity, electromagnet current, suspension gap, and vehicle acceleration, comprehensively reflecting the operating characteristics of the suspension control system and the quality of network communication. By comparing and modeling the data with the adjusted operating parameter response baseline, the system outputs an anomaly confidence level to characterize the degree of network anomaly suspicion, thus providing a quantitative basis for determining the network status.

[0037] Step 103: Analyze the operating data to obtain the network status of the suspension control system.

[0038] In this embodiment, the operational data is analyzed to obtain an anomaly confidence level characterizing the network state of the suspension control system. Further, based on the numerical range of the anomaly confidence level, the current network state of the maglev train can be accurately identified and determined. For example, a first threshold and a second threshold are preset, with the second threshold being greater than the first threshold. When the anomaly confidence level is lower than the first threshold, it indicates that the current anomaly characteristics highly match environmental interference, and the network state is determined to have environmental interference. When the anomaly confidence level is between the first and second thresholds, it indicates that the anomaly characteristics are suspicious but not yet clear, and the network state is determined to be suspected of having a network attack. When the anomaly confidence level is higher than the second threshold, it confirms the existence of obvious malicious behavior characteristics, and the network state is determined to have a network attack.

[0039] In this embodiment, the system implements differentiated and tiered security handling strategies for different network states to balance operational continuity and security. Specifically, when environmental interference is identified, the system only logs relevant environmental data, operational data, and interference events, without interfering with the normal operation of the maglev train, thus avoiding unnecessary security degradation. When a suspected network attack is identified, the system proactively limits the transmission rate of the affected communication ports to reduce the scope of the anomaly, while increasing the monitoring frequency, encrypting the data collection and analysis cycle, and continuously tracking the trend of abnormal changes. When a network attack is confirmed, the system immediately takes the highest level of security protection measures, switching the maglev train's levitation control system to local security mode, blocking suspicious external commands, implementing speed limit control on the maglev train, and guiding the maglev train to stop at the nearest safe section, fundamentally eliminating the safety risk.

[0040] In the method of this embodiment, the above-mentioned abnormal confidence interval division, threshold setting method, and corresponding hierarchical handling strategy are all illustrative examples and not the only limitations. Those skilled in the art can reasonably adjust, replace, or combine the threshold values, interval division logic, and handling procedures according to actual operating conditions, security protection levels, system performance requirements, etc., all of which fall within the protection scope of this invention.

[0041] The method described in this embodiment addresses the network security monitoring needs of maglev train suspension control systems in complex environments. It achieves several technological breakthroughs and significant performance improvements by constructing an environmental perception and hierarchical fusion monitoring architecture, dynamically adjusting the operating parameter response baseline, and establishing a multi-source fusion verification and hierarchical response mechanism. Dynamically adjusting the operating parameter response baseline using real-time acquired environmental data enables monitoring standards to adapt to complex operating conditions such as high altitudes, tunnels, and strong electromagnetic interference, avoiding misjudgments and missed judgments caused by environmental fluctuations due to fixed baselines. This significantly reduces the false alarm rate and greatly improves environmental adaptability. Secondly, the sunken, lightweight monitoring probe achieves physical integration of the monitoring node and the control closed loop. Combined with deep parsing of industrial real-time Ethernet protocols and multi-source data fusion technology, it can accurately capture key operational data such as network latency, packet loss rate, and sensor signals at the millisecond level, balancing real-time monitoring with fine-grained identification capabilities to meet the millisecond-level response requirements of the suspension control system.

[0042] Simultaneously, a dynamic consistency verification model integrating electromagnet current, suspension gap, and vehicle acceleration is constructed. By adjusting the electromagnetic force-current and gap-acceleration coupling relationships based on environmental data, it can accurately distinguish between natural environmental interference and malicious network attacks, significantly improving attack identification accuracy and effectively preventing security threats such as command tampering and injection. Furthermore, a wheel speed, inertia, and map-fusion positioning architecture is adopted to ensure position continuity under weak / no signal conditions. Combined with a tiered response and local autonomous protection mechanism, it can maintain complete monitoring and protection capabilities under extreme conditions such as communication interruption and location loss, significantly improving protection continuity and operational safety.

[0043] Finally, through the vehicle-ground collaborative incremental learning mechanism, the operating parameter response baseline and detection model are continuously optimized, enabling the system to have scenario-based adaptive iteration capabilities. This can continuously improve monitoring accuracy and generalization capabilities, providing long-term, stable, and reliable network security protection for the maglev train suspension control system in complex environments.

[0044] Example 2: Figure 3 This is a flowchart illustrating another method for identifying the network status of a train levitation control system disclosed in this application. The method in this embodiment revolves around geographical location, using the real-time geographical location of the maglev train, satellite signal status, and route segment type as inputs. It completes location source switching, segment identification, baseline dynamic adjustment, verification of the legality of command geographical location tags, and attack discrimination. This achieves accurate differentiation between environmental interference and malicious network attacks, as well as determination of command credibility, providing a basic decision-making basis for subsequent graded response steps. The specific implementation is as follows: First, the onboard sensing unit, positioning unit, and communication unit simultaneously collect real-time geographical location, operating speed, levitation status, communication link status, and geographical information of the maglev train, forming the original data base for dynamic evaluation and providing comprehensive and reliable data support for subsequent analysis.

[0045] Subsequently, the validity of the satellite positioning signal is determined in real time, and a dual-mode positioning switch is executed to ensure the continuity of position data under conditions of weak or no signal, such as in high-altitude areas and tunnels. When satellite signals are available, satellite positioning data is directly used as the position reference source for the maglev train, ensuring positioning accuracy and real-time performance. When satellite signals are unavailable due to tunnel obstruction, high-altitude terrain shielding, or strong electromagnetic interference, the system automatically switches to a fusion positioning mode combining wheel speed sensors, inertial navigation, and electronic maps. Through mileage calculation, attitude displacement calculation, and route topology matching, continuous positioning is achieved in the absence of satellite signals, effectively avoiding monitoring failures caused by positioning interruptions.

[0046] In this embodiment, the real-time positioning data after switching and calibration is matched with the onboard pre-stored electronic map with high precision. Based on geographical and environmental characteristics, the type of section the maglev train is currently in is identified, including tunnel sections, high-altitude sections, sections with strong electromagnetic interference, and open areas. Subsequently, the corresponding operating parameter response baseline is dynamically loaded. For example, in tunnel sections, network latency tolerance is increased and the trust weight of remote commands is reduced to adapt to wireless signal attenuation and latency fluctuations. In high-altitude sections, perception sensitivity is reduced, and physical coupling relationships such as electromagnetic force-current and levitation gap-acceleration are adjusted to match the characteristics of low-pressure environments. In sections with strong electromagnetic interference, the sensitivity of network anomaly detection is lowered to avoid misjudgments caused by natural electromagnetic fluctuations. In open areas, the normal monitoring benchmark is restored to improve the accuracy of attack identification.

[0047] After the dynamic operating parameter response baseline is loaded, the spatiotemporal consistency verification phase begins, where each frame of network command received by the levitation controller undergoes dual verification. If the geographic location tag carried by the command matches the real-time location of the maglev train, it is determined to be a trusted command and allowed to be issued and executed normally. If the tag does not match, it is identified as malicious behavior such as a replay attack, determined to be a suspicious command, and a security risk is promptly marked.

[0048] In this embodiment, the combined results of geographic location verification, segment adjustment, and command credibility determination are output to the multi-source anomaly detection module as input for anomaly confidence calculation, tiered response triggering, and security degradation execution. Simultaneously, the evaluation data is fed back to the baseline update unit to continuously optimize the dynamic trust model for each segment, forming a closed-loop evaluation mechanism encompassing location awareness, environment adaptation, command verification, decision output, and model iteration, thereby improving adaptability and long-term stability.

[0049] In the method of this embodiment, Figure 4This application discloses a logic diagram for network state identification of a maglev train suspension control system. It introduces the overall logic based on four-dimensional heterogeneous feature acquisition (sensing layer, network layer, application layer, and environment layer). Through multi-source data fusion and a lightweight anomaly detection algorithm, it completes feature association modeling, anomaly quantification and discrimination, and attack-interference separation, ultimately outputting anomaly confidence and performing differentiated handling. Specifically, it simultaneously extracts the full-link operational features of the suspension control system from four dimensions: physical sensing, network transmission, command execution, and external environment. This constructs a complete detection feature set covering sensing, transmission, execution, and environment, comprehensively reflecting the operating status and external conditions of the suspension control system.

[0050] The sensing layer primarily collects three types of parameters: raw sensor measurements, signal noise levels, and changes in physical quantities. This directly reflects the original state and signal interference characteristics of key physical quantities such as suspension gap, electromagnet current, and vehicle acceleration, providing underlying data support for subsequent physical consistency verification. The network layer focuses on collecting four indicators: communication latency, data packet loss rate, command transmission frequency, and frame structure integrity. These indicators characterize the transmission quality, message validity, and link health of the in-vehicle industrial Ethernet, accurately identifying abnormal transmission behavior at the network layer. The application layer collects three types of parameters: command response deviation, control response time, and electromagnetic force-current mapping deviation. Based on the inherent physical characteristics of the suspension control system's electromagnetic force response and gap-current coupling, it verifies the command execution effect and the compliance of dynamic logic. The environment layer simultaneously collects parameters such as altitude, air pressure, ambient temperature, tunnel markings, electromagnetic interference intensity, and maglev train speed, comprehensively covering complex environmental interference factors and providing a benchmark for distinguishing between environmental fluctuations and malicious network attacks.

[0051] Subsequently, the aforementioned four layers of heterogeneous feature data are fed into a multi-source data fusion unit, where data normalization, spatiotemporal alignment, feature association, and redundancy removal are performed sequentially to eliminate dimensional differences, temporal deviations, and invalid information, forming a unified comprehensive detection input vector. The fused data is then input into a lightweight anomaly detection algorithm for real-time computation. This algorithm is optimized for the limited computing power of automotive embedded platforms and the millisecond-level response requirements of the suspension control system. Using environmental data as a dynamic adjustment benchmark, it integrates features from perception, network, and application layers to construct an adaptive detection model that considers physical characteristics, network status, and environmental interference, achieving anomaly feature extraction and anomaly degree quantification.

[0052] The lightweight anomaly detection algorithm relies on a fusion model to classify anomalies into two categories: environmental interference and malicious network attacks, accurately locating the root cause of the anomaly. If the abnormal changes are highly correlated with environmental factors such as altitude, air pressure, tunnels, and electromagnetic interference, and conform to natural fluctuation patterns and do not violate dynamic constraints, then it is determined to be environmental interference, belonging to normal operating condition fluctuations. If it manifests as characteristics such as command tampering, frame corruption, abnormal electromagnetic force-current mapping, and severe command response mismatch, and cannot be explained by environmental factors, then it is determined to be a malicious network attack, constituting an illegal act that threatens operational security.

[0053] After completing the judgment, the algorithm outputs the anomaly confidence level, which serves as the basis for subsequent handling and executes differentiated strategies. For environmental interference events, a lightweight approach is adopted, such as reducing the alarm level or only logging, to avoid false alarms and ensure the continuous and stable operation of the suspension control system. For malicious network attack events, a tiered response is triggered based on the confidence level, linking command interception, security degradation, local protection, and other measures to quickly block the attack and ensure the safety of suspension control in complex environments.

[0054] Example 3: Figure 5 This is a hierarchical response timing diagram disclosed in an embodiment of this application. The method in this embodiment uses the judgment result of the anomaly detection module as the trigger source, linking four functional units: safety audit, suspension control, in-vehicle human-machine interaction, and ground remote collaboration, to construct a multi-level gradient response mechanism. Specifically, it can be as follows: In this embodiment, each unit works collaboratively according to a fixed timing sequence and command link to form a complete security handling closed loop. The anomaly detection module is responsible for the decision-making and triggering of graded responses, determining the response level based on the anomaly confidence and anomaly type output by multi-source anomaly detection, and issuing response commands to downstream units. The security audit module records anomaly events, response actions, and suspension control system status data throughout the process for attack tracing, model optimization, and security compliance tracing. The suspension controller, as the terminal execution unit, receives response commands and performs control layer protection operations such as port restriction, command interception, and mode switching. The driver's cab display unit pushes prompts, warnings, or alarms to the driver in real time, enabling human-machine collaborative handling. The ground security center receives vehicle-mounted anomaly information, participates in low-to-medium risk response decisions, and automatically degrades to local autonomous protection when wireless communication is interrupted.

[0055] As an example, when the anomaly confidence level is below 0.3 and environmental interference is determined to be the primary cause, a Level 1 response is initiated, implementing minimal intervention measures. Specifically, the anomaly detection module records event characteristics, environmental data, and preliminary classification results to the safety audit module, while simultaneously sending a notification to the driver's cab, informing the driver that there is a minor environmental fluctuation and no adjustment to driving operations is necessary. During this phase, there is no intervention in the operation of the suspension controller, nor is there communication with the ground safety center, ensuring that the normal operation of the maglev train remains unaffected.

[0056] When the anomaly confidence level is between 0.3 and 0.7, indicating a suspected network attack, a Level 2 response is initiated. Specifically, the anomaly detection module sends a command to the suspension controller to limit the affected port's rate to 50% of the normal rate, suppressing the spread of the anomaly. Simultaneously, a warning message is sent to the driver's cab, reminding them to pay attention to the suspension control system's status, and an enhanced monitoring mode is activated, shortening the detection cycle from 100 milliseconds to 20 milliseconds and encrypting data collection. All actions at this stage are completed autonomously on the vehicle, without relying on ground communication.

[0057] When the anomaly confidence level exceeds 0.7, indicating a network attack, a Level 3 response is initiated. Specifically, the anomaly detection module instructs the levitation controller to switch to local security mode and reject untrusted external commands. The levitation controller executes local security policies, maintaining the preset levitation gap and current curve, while simultaneously limiting the maglev train's speed to 80 km / h and guiding it to a stop. Alarm information is simultaneously displayed and sent to the driver's cab, and attack details are encrypted and uploaded to the ground security center for coordinated defense and model updates.

[0058] Furthermore, in the event of wireless communication interruption, the ground reporting stage of the Level 3 response can be downgraded to local storage. When satellite positioning is lost, geographic location verification can be suspended, and protection can be maintained using other methods. Under conditions of strong electromagnetic interference, the Level 3 trigger threshold can be raised to 0.8 to prevent false escalation.

[0059] In the method of this embodiment, the hierarchical response timing fully covers the progressive link from anomaly identification to safety handling. By grading confidence and adaptively adjusting the environment, it takes into account both safety and operational continuity, ensuring the continuous and stable operation of the maglev train levitation control system in complex environments.

[0060] Example 4: Figure 6 This is a schematic diagram of the working parameter response baseline update process disclosed in an embodiment of this application. This embodiment describes the adaptive update of the working parameter response baseline for model optimization, and the specific implementation method is as follows: In this embodiment, the vehicle-mounted system, acting as both the data source and local execution unit, is responsible for comprehensive data acquisition, local storage, and basic maintenance, providing continuous and reliable raw data support for model iteration. During the data accumulation phase, the vehicle-mounted system records in real-time and automatically the input features, detection results, environmental data, and manual or automatic confirmation results generated by the anomaly detection module. The data covers physical features at the perception layer, transmission features at the network layer, execution features at the application layer, interference parameters at the environmental layer, and anomaly judgment conclusions, forming a complete and comprehensive labeled training sample library. This achieves the systematic accumulation of multi-dimensional operational data, providing a high-quality data foundation for subsequent model learning.

[0061] After data accumulation, the vehicle-mounted system enters the scenario-based classification and storage phase. The collected raw data undergoes cleaning, deduplication, standardization, and labeling to remove invalid and redundant information, ensuring data quality. Subsequently, the processed data is finely classified and stored according to segment type, environmental characteristics, and anomaly type, ensuring that the local data structure is standardized, traceable, and easy to retrieve and call later, providing orderly data support for scenario-based model training. The system then performs incremental learning and model fine-tuning, employing incremental clustering or online gradient descent algorithms to dynamically fine-tune the vehicle-mounted offline basic model according to operating mileage or time periods (e.g., every 1000 kilometers or every 24 hours).

[0062] As a feasible solution, for false alarm events, the system learns the feature distribution patterns in the corresponding environmental scenarios, updates the segment baseline model, reduces monitoring sensitivity in similar scenarios, and decreases false alarms. For missed alarm events, the system strengthens the weights of relevant feature dimensions to improve the ability to identify similar anomalies. For new environmental scenarios, the system automatically expands the baseline model library to achieve zero-sample adaptation and ensure the model has good generalization ability. Finally, the system enters the model update and deployment phase, loading the fine-tuned local baseline model parameters into the running monitoring engine to complete the online model update, forming a continuously iterating and optimizing in-vehicle self-learning closed loop.

[0063] In this embodiment, the ground safety analysis platform, acting as the central hub for global data aggregation and model optimization, receives sample data uploaded by multiple maglev trains. It then meticulously categorizes and stores the samples according to segment types such as tunnels, high altitudes, areas with strong electromagnetic interference, and open areas, constructing a standardized sample system categorized by scenario, environment, and anomaly type. This provides diverse and high-quality data support for global model training. During maintenance at the depot, each maglev train encrypts and uploads its onboard safety logs (which may include sample libraries, model parameters, and anomaly records) to the ground platform to ensure data security. After aggregating data from multiple trains, the platform utilizes enhanced computing power to conduct global model training, generating superior model parameters. Finally, the optimized parameters are distributed to each onboard system via wireless communication or maintenance interfaces, completing the global model iteration.

[0064] After the model is updated, the vehicle system continues to collect data, detect anomalies, and accumulate samples using the new baseline and model. It cyclically executes the processes of data accumulation, uploading, global training, parameter distribution, and local updates, forming a closed-loop adaptive evolution mechanism to achieve continuous model optimization.

[0065] The method in this embodiment breaks through the technical bottlenecks of traditional static baselines, fixed models, and lack of iteration. Relying on a vehicle-to-ground collaborative distributed learning architecture, it takes into account both the real-time performance of the vehicle-mounted end and the generalization capability of the ground end, and realizes the scenario-based, adaptive, and continuous iteration of the monitoring model and dynamic baseline. This provides long-term, stable, and reliable technical support for the network security monitoring of the suspension control system in complex environments.

[0066] Based on the method for identifying the network status of a train suspension control system disclosed in the above embodiments, this embodiment correspondingly discloses a device for identifying the network status of a train suspension control system. Please refer to... Figure 7 The device for identifying the network status of a train suspension control system includes: an adjustment unit 201, a monitoring unit 202, and an analysis unit 203; The adjustment unit 201 is used to adjust the working parameter response baseline of the monitoring probe in real time based on the environmental data collected in real time; the environmental data includes one or more of the following: altitude, temperature, air pressure, electromagnetic interference intensity, and geographical location of the environment in which the maglev train suspension control system is located; the monitoring probe is set on the suspension control system and is used to collect the operating data of the maglev train in real time; the working parameter response baseline includes probe sensing sensitivity, sampling frequency, data filtering rules, and signal preprocessing threshold. The monitoring unit 202 is used to collect the operating data of the suspension control system at the same time based on the adjusted operating parameter response baseline; the operating data includes suspension gap, electromagnet current, vehicle acceleration, network latency, and packet loss rate. The analysis unit 203 is used to analyze the operating data to obtain the network status of the suspension control system.

[0067] Optionally, the adjustment unit 201 includes: The sub-units are used to divide the operating route of the maglev train into different sections based on the environmental data. The adjustment subunit is used to differentiate the operating parameter response baseline according to the different sections.

[0068] Optionally, the adjustment subunit includes: A first adjustment subunit is configured to perform a first adjustment on the operating parameter response baseline when the segment is a tunnel segment; the first adjustment includes increasing network latency tolerance and / or reducing remote command trust weight. The second adjustment subunit is used to make a second adjustment to the operating parameter response baseline when the section is a high-altitude section; the second adjustment includes one or more of the following: reducing the sensing sensitivity in the operating parameter response baseline, adjusting the relationship between electromagnetic force and current, and adjusting the relationship between suspension gap and acceleration. The third adjustment subunit is used to make a third adjustment to the operating parameter response baseline when the section is a strong electromagnetic interference section; the third adjustment includes reducing the sensitivity of network anomaly detection; The recovery subunit is used to adjust the operating parameter response baseline to its initial state when the section is an open area section.

[0069] Optionally, the analysis unit 203 includes: An identification subunit is used to identify the operating data based on the adjusted operating parameters response baseline to obtain the anomaly confidence level of the suspension control system. A sub-unit is defined to determine the network state based on the value of the anomaly confidence level.

[0070] Optionally, the determining subunit includes: An interference determination subunit is used to determine that environmental interference exists in the network state when the anomaly confidence level is lower than a first threshold. An attack suspicion subunit is used to determine that the network state is suspected of being under a network attack when the abnormal confidence level is between the first threshold and the second threshold; the second threshold is greater than the first threshold. An attack determination subunit is used to determine that the network state is under attack when the anomaly confidence level is higher than the second threshold.

[0071] Optionally, the device further includes: A recording unit is used to record abnormal information about the network state when the network state is subject to environmental interference. A limiting unit is used to limit the rate of the affected ports and increase the sampling frequency of the monitoring probe when the network status is suspected of being under network attack. The setting unit is used to set the suspension control system to a local security mode, limit the speed of the maglev train, and guide it to stop when the network status indicates a network attack.

[0072] Optionally, the environmental data includes the geographical location, and the device further includes: The tag acquisition unit is used to acquire the geographic location tags in each frame of instructions received by the suspension control system; The instruction credibility determination unit is used to determine that the instruction is a credible instruction when the geographic location tag matches the geographic location; The instruction suspicion determination unit is used to determine that the instruction is a suspicious instruction when the geographic location tag does not match the geographic location.

[0073] Based on the above-described method for identifying the network status of a train suspension control system, this application also discloses a computer program product, which includes a computer program that, when run by a processor, is used to implement the above-described method.

[0074] Based on the above-described method for identifying the network status of a train suspension control system, this application also discloses a storage medium for storing computer program instructions, which, when executed by a central processing unit, are used to implement the above-described method.

[0075] Based on the above-mentioned method for identifying the network status of a train suspension control system, this application also discloses a maglev train, including an environmental sensing device, a processor, a monitoring probe, and a storage device, for implementing the above-mentioned method.

[0076] The embodiments in this specification are described in a progressive manner. For the apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple; relevant details can be found in the method section.

[0077] It should also be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0078] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0079] The features described in the embodiments of this specification can be substituted for or combined with each other, so that those skilled in the art can implement or use this application.

[0080] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for network state recognition of a train levitation control system, characterized by, include: The operating parameter response baseline of the monitoring probe is adjusted in real time based on the environmental data collected in real time. The environmental data includes one or more of the following: altitude, temperature, air pressure, electromagnetic interference intensity, and geographical location of the environment in which the maglev train suspension control system is located. The monitoring probe is set on the suspension control system to collect the operating data of the maglev train in real time. The operating parameter response baseline includes probe sensing sensitivity, sampling frequency, data filtering rules, and signal preprocessing threshold. At the same time, based on the adjusted operating parameter response baseline, the operating data of the suspension control system is collected; the operating data includes suspension gap, electromagnet current, vehicle acceleration, network latency, and packet loss rate. The network status of the suspension control system is obtained by analyzing the operational data.

2. The method of claim 1, wherein, The real-time adjustment of the monitoring probe's operating parameter response baseline based on real-time acquired environmental data includes: Based on the environmental data, the operating route of the maglev train is divided into different sections; The operating parameter response baseline is adjusted differently depending on the segment.

3. The method of claim 2, wherein, The differential adjustment of the operating parameter response baseline according to the different segments includes: In the case that the section is a tunnel section, the operating parameter response baseline is first adjusted; the first adjustment includes increasing network latency tolerance and / or reducing remote command trust weight; When the section is a high-altitude section, a second adjustment is made to the operating parameter response baseline; the second adjustment includes one or more of the following: reducing the sensing sensitivity in the operating parameter response baseline, adjusting the relationship between electromagnetic force and current, and adjusting the relationship between suspension gap and acceleration; In the case that the section is a strong electromagnetic interference section, a third adjustment is made to the baseline of the operating parameter response; the third adjustment includes reducing the sensitivity of network anomaly detection; When the section is an open area, adjust the operating parameters to the baseline to the initial state.

4. The method of claim 1, wherein, The process of analyzing the operational data to obtain the network status of the suspension control system includes: The operational data is identified using the adjusted operating parameters and response baseline to obtain the anomaly confidence level of the suspension control system; The network state is determined based on the value of the anomaly confidence level.

5. The method of claim 4, wherein, Determining the network state based on the value of the abnormal confidence level includes: If the anomaly confidence level is lower than a first threshold, it is determined that the network state is subject to environmental interference. If the abnormal confidence level is between the first threshold and the second threshold, the network state is determined to be suspected of being under network attack; the second threshold is greater than the first threshold. If the anomaly confidence level is higher than the second threshold, the network state is determined to be under network attack.

6. The method of claim 5, wherein, After obtaining the network status of the suspension control system, the method further includes: If the network is experiencing environmental interference, record the abnormal information of the network status. If the network status indicates a suspected network attack, limit the rate of the affected ports and increase the sampling frequency of the monitoring probes; If the network status indicates a network attack, the suspension control system will be set to local security mode, the speed of the maglev train will be limited, and the train will be guided to a stop.

7. The method of claim 1, wherein, The environmental data includes the geographical location. After adjusting the baseline of the monitoring probe's operating parameters in real time based on the real-time collected environmental data, the method further includes: Obtain the geolocation tags from each frame of instructions received by the suspension control system; If the geographic location tag matches the geographic location, the instruction is determined to be a trusted instruction; If the geographic location tag does not match the geographic location, the instruction is determined to be a suspicious instruction.

8. A device for network state recognition of a train levitation control system, characterized by, include: Adjustment unit, monitoring unit, and analysis unit; The adjustment unit is used to adjust the working parameter response baseline of the monitoring probe in real time based on the environmental data collected in real time. The environmental data includes one or more of the following: altitude, temperature, air pressure, electromagnetic interference intensity, and geographical location of the environment in which the maglev train suspension control system is located. The monitoring probe is set on the suspension control system and is used to collect the operating data of the maglev train in real time. The working parameter response baseline includes probe sensing sensitivity, sampling frequency, data filtering rules, and signal preprocessing threshold. The monitoring unit is used to collect the operating data of the suspension control system at the same time based on the adjusted operating parameters response baseline; the operating data includes suspension gap, electromagnet current, vehicle acceleration, network latency, and packet loss rate. The analysis unit is used to analyze the operating data to obtain the network status of the suspension control system.

9. A computer program product, characterised in that, The computer program product includes a computer program that, when executed by a processor, is used to implement the method for identifying the network status of the train suspension control system as described in any one of claims 1 to 7.

10. A maglev train characterized by It includes an environmental sensing device, a processor, a monitoring probe, and a storage device, for implementing the method described in any one of claims 1 to 7.

Images