A method and system for intelligent monitoring and early warning of urban rail vehicle doors based on PHM technology

Abstract

The application discloses a kind of city rail car door intelligent monitoring early warning method and system based on PHM technology, belong to intelligent monitoring early warning technical field;Multi-modal sensor is deployed at the key mechanical node of city rail car door, the operation data of city rail car door is collected, and time stamp is synchronized, and multi-source data interaction graph is constructed;Based on the multi-source data interaction graph, the health degree of single city rail car door at a single time stamp is calculated;Time stamp-health degree curve of single city rail car door is constructed;Based on time stamp-health degree curve, the health abnormality degree of single city rail car door is calculated;Similarity between different city rail car doors is calculated;Threshold is preset, and early warning is carried out.The application builds multi-source data interaction graph, combines the dynamic evaluation of health degree and health abnormality degree, supplemented by similarity analysis and early warning threshold determination between doors, effectively improves the accuracy of door system health state identification and the timeliness of abnormal detection.

Description

Technical Field

[0001] This invention relates to the field of intelligent monitoring and early warning technology, specifically to an intelligent monitoring and early warning method and system for urban rail vehicle doors based on PHM technology. Background Technology

[0002] With the rapid development of urban rail transit systems, the operational safety and reliability of urban rail trains, as an important carrier of public transportation, are receiving increasing attention. Among these, the door system, as a critical interface for passenger boarding and alighting, directly impacts train operation efficiency and passenger safety. Therefore, health monitoring and fault early warning for door systems have become important research directions for the intelligent operation and maintenance of rail transit equipment. In recent years, with the continuous development of information technology, sensing technology, and artificial intelligence, methods for monitoring the condition of door systems have gradually shifted from offline diagnosis based on manual inspections to online intelligent diagnosis based on data-driven approaches. Especially driven by the concept of PHM (Prognostics and Health Management), more and more research is attempting to apply multimodal sensor data acquisition, health status modeling, and intelligent prediction mechanisms to key subsystems of rail transit to achieve predictable management of equipment lifespan. However, PHM research for door systems is still in its early stages, lacking a systematic deployment plan and reliable anomaly early warning mechanisms, which restricts its widespread application in actual urban rail trains.

[0003] While some existing technologies have incorporated sensors and monitoring algorithms to assess door status, most methods suffer from the following limitations: First, the data dimension is limited, often relying on a single physical quantity (such as current or displacement) for analysis, making it difficult to comprehensively reflect the multidimensional changes in door operation and easily leading to misjudgments or omissions in fault identification. Second, the lack of a unified data synchronization mechanism prevents precise temporal alignment between data from different sensors, reducing the effectiveness of multi-source information fusion. Third, most existing models employ static analysis methods, lacking dynamic trajectory modeling and calculation of health evolution trends, thus failing to provide early warnings based on health change trends. Furthermore, traditional algorithms do not adequately consider the correlation between different doors, failing to effectively uncover the health evolution patterns between similar doors, resulting in weak systematic predictive capabilities. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for intelligent monitoring and early warning of urban rail vehicle doors based on PHM technology, so as to solve the problems mentioned in the background art.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0006] A method for intelligent monitoring and early warning of urban rail vehicle doors based on PHM technology is disclosed. The method includes the following steps: Step S1: Deploy multimodal sensors at key mechanical nodes of the urban rail vehicle door to collect its operational data and synchronize the timestamps; Step S2: Embed a communication interface in the PHM system of the urban rail vehicle door controller and transmit the received operational data to the ground PHM system; Step S3: Construct a multi-source data interaction graph based on the operational data at a single timestamp; calculate the door health of a single urban rail vehicle door at a single timestamp based on the multi-source data interaction graph; Step S4: Obtain the door health of all timestamps and construct a timestamp-door health curve for a single urban rail vehicle door; calculate the door health anomaly degree of a single urban rail vehicle door based on the timestamp-door health curve; calculate the similarity between different urban rail vehicle doors; preset a threshold and issue an early warning.

[0007] As a preferred embodiment of the intelligent monitoring and early warning method for urban rail vehicle doors based on PHM technology described in this invention, multimodal sensors are deployed at key mechanical nodes of the urban rail vehicle doors. The multimodal sensors include position sensors, current sensors, speed sensors, and distance sensors. The key mechanical nodes include the connection node between the slide rail and the door leaf, the door motor, the door leaf, and the slide rail.

[0008] Based on multimodal sensors, the operation data of the urban rail vehicle door is collected. The operation data includes door mechanical position data, door motor current data, door movement data, door V-shape dimension data, and door swing dimension data.

[0009] Using the Bluetooth Mesh protocol, a communication connection is established between the position sensor, current sensor, speed sensor and distance sensor, and the mechanical position data of the door, the current data of the door motor, the movement data of the door, the V-shaped dimension data of the door and the outward swing dimension data are synchronized and aligned on the timestamp.

[0010] As a preferred embodiment of the intelligent monitoring and early warning method for urban rail vehicle doors based on PHM technology described in this invention, a communication interface is embedded in the PHM system of the urban rail vehicle door controller. The communication interface is used to receive door mechanical position data, door motor current data, door movement data, door V-shaped dimension data, and outward swing dimension data after synchronization and alignment based on timestamps; and to transmit the received door mechanical position data, door motor current data, door movement data, door V-shaped dimension data, and outward swing dimension data to the ground PHM system.

[0011] Record the t-th timestamp as TY. t The timestamp TY in the ground PHM system tThe mechanical position data, motor current data, movement data, V-shaped dimension data, and outward swing dimension data of the i-th urban rail vehicle door after synchronization are denoted as DMP. i (TY t DMC i (TY t ), DM i (TY t DVS i (TY t ) and ES i (TY t ).

[0012] As a preferred embodiment of the intelligent monitoring and early warning method for urban rail vehicle doors based on PHM technology described in this invention, based on timestamp TY... t DMP of the mechanical position data of the i-th urban rail vehicle door after synchronization and alignment i (TY t Door motor current data DMC i (TY t ), door movement data DM i (TY t Door V-shape dimension data DVS i (TY t ) and outer swing dimension data ES i (TY t Construct a multi-source data interaction graph, as follows:

[0013] DMP of mechanical position data of car door i (TY t Let V1(TY) be the mechanical position node of the multi-source data interaction graph. t ), DMC of the door motor current i (TY t ) is denoted as the current node V2(TY) of the multi-source data interaction graph. t ), DM the door movement data i (TY t Let V3(TY) be the moving trajectory node of the multi-source data interaction graph. t ), DVS the V-shaped dimensions of the car door i (TY t ) is denoted as V4(TY) of the V-shaped size node in the multi-source data interaction graph. t ), and transfer the outer swing dimension data ES i (TY t Let V5(TY) be the outer swing dimension node of the multi-source data interaction graph. t ).

[0014] The asymmetric coupling rate of change between any two nodes is taken as an edge of the multi-source data interaction graph. The formula for calculating the asymmetric coupling rate of change is as follows:

[0015]

[0016] ΔV j (TY t ) = V j (TY t )-V j (TY t-1 )

[0017] in, Indicates the timestamp TY t The rate of change of the asymmetric coupling between the j-th node and the k-th node, ΔV j (TY t ) indicates that the j-th node is from the timestamp TY t-1 To the timestamp TY t The node changes, ΔV k (TY t ) represents the k-th node starting from timestamp TY t-1 To the timestamp TY t The node changes, α represents the preset adjustment parameter, V j (TY t ) represents the timestamp Ty t The j-th node at time V j (TY t-1 ) represents the timestamp TY t-1 The j-th node at time.

[0018] Based on the multi-source data interaction graph, the time stamp TY of the i-th urban rail vehicle door is calculated. t The health status of the car door at that time is calculated using the following formula:

[0019]

[0020] Among them, FHSI i (TY t ) indicates that the i-th urban rail vehicle door is at timestamp TY t The health of the car door at that time, λ jk This represents the preset edge weight between the j-th node and the k-th node. Indicates the timestamp TY t The rate of change of the asymmetric coupling between the j-th node and the k-th node, V j (TY t ) represents the timestamp TY t The j-th node at time, μ j Let σ represent the mean of the j-th node.j Let represent the standard deviation of the j-th node, and ∈ represent the preset error term.

[0021] As a preferred embodiment of the intelligent monitoring and early warning method for urban rail vehicle doors based on PHM technology described in this invention, based on the timestamp TY of the i-th urban rail vehicle door... t Door health index (FHSI) i (TY t To obtain the health status of the i-th urban rail vehicle door across all timestamps, construct the timestamp-door health status curve for the i-th urban rail vehicle door, as follows:

[0022] Construct a timestamp-door health coordinate system. Use all T sequentially arranged timestamps as the x-coordinate of the system, and record the door health corresponding to each of the T timestamps as the y-coordinate. Connect all coordinate points in the timestamp-door health coordinate system sequentially to construct a timestamp-door health curve, denoted as TDH[FHSI]. i (TY t )).

[0023] Based on timestamp-door health curve TDH[FHSI] i (TY t ]] Calculate the door health anomaly degree of the i-th urban rail vehicle door using the following formula:

[0024]

[0025]

[0026] Among them, S i (TY t ) indicates that the i-th urban rail vehicle door is at timestamp TY t-1 To the timestamp TY t The rate of descent of the curve at time FHSI i (TY t-1 ) indicates that the i-th urban rail vehicle door is at timestamp TY t-1 The health of the car door at that time, C i (TY t ) indicates that the i-th urban rail vehicle door is at timestamp TY t Local curvature factor at time, FHSI i (TY t+1 ) indicates that the i-th urban rail vehicle door is at timestamp TY t+1 The health of the car door at that time, A i (TY t ) represents the door health abnormality degree of the i-th urban rail vehicle door. ω and ω represent the weighting coefficients of the preset curve descent rate and local curvature factor, respectively.

[0027] The similarity between different urban rail vehicle doors is calculated using the formula: ΔA i (TY t )=|A i (TY t )-A i+1 (TY t )|, where ΔA i (TY t A represents the similarity between the i-th urban rail vehicle door and the (i+1)-th urban rail vehicle door. i+1 (TY t ) represents the door health abnormality of the (i+1)th urban rail vehicle door.

[0028] Preset thresholds for door health anomaly and similarity; if the door health anomaly A of the i-th urban rail vehicle door... i (TY t If the number of times the door health anomaly threshold is greater than or equal to the preset threshold, and the similarity between the door and the (i+1)th urban rail door is less than or equal to the similarity threshold, then the i-th urban rail door is determined to have a health anomaly at the next time stamp and is similar to the (i+1)th urban rail door. If the i-th urban rail door has a health anomaly at the next time stamp, then all doors similar to the i-th urban rail door are obtained, and a unified early warning is issued to remind staff to carry out maintenance in advance.

[0029] A smart monitoring and early warning system for urban rail vehicle doors based on PHM technology. The system includes: a data acquisition and synchronization module, a data receiving and transmission module, an interactive graph and health calculation module, and a curve construction and anomaly analysis and early warning module.

[0030] The data acquisition and synchronization module deploys multimodal sensors at key mechanical nodes of the urban rail vehicle door to collect the operating data of the urban rail vehicle door and synchronize it with timestamps.

[0031] The data receiving and transmission module: embeds a communication interface in the PHM system of the urban rail vehicle door controller, and transmits the received operational data to the ground PHM system;

[0032] The interaction graph and health calculation module: constructs a multi-source data interaction graph based on the operating data at a single timestamp; and calculates the health of a single urban rail vehicle door at a single timestamp based on the multi-source data interaction graph.

[0033] The curve construction and anomaly analysis and early warning module: obtains the door health status under all timestamps, constructs a timestamp-door health status curve for a single urban rail vehicle door; calculates the door health anomaly status for a single urban rail vehicle door based on the timestamp-door health status curve; calculates the similarity between different urban rail vehicle doors; presets a threshold and provides early warning.

[0034] Compared with existing technologies, the beneficial effects achieved by this invention are as follows: The intelligent monitoring and early warning method and system for urban rail vehicle doors based on PHM technology provided by this invention deploys multimodal sensors, including position, current, speed, and distance sensors, at key mechanical nodes to collect multi-dimensional dynamic data during door operation in real time. It also utilizes the Bluetooth Mesh protocol to achieve precise time synchronization between data, thereby ensuring information integrity and timing accuracy, laying a data foundation for subsequent processing. Furthermore, a communication interface is embedded in the door controller to transmit the synchronized data to the ground PHM system in real time, realizing data linkage and remote access capabilities between the vehicle and ground ends, enhancing the system's scalability and real-time monitoring capabilities. Based on this, the system constructs a system containing multiple feature nodes and asymmetric... The system employs a multi-source data interaction graph with coupled relationships, and further introduces a rate-of-change calculation method. By integrating parameters such as node mean and standard deviation, it quantifies the health status of the doors at a single point in time, achieving accurate modeling of single-point health. Subsequently, by constructing a complete timestamp-door health curve, it not only depicts the operational stability of the doors throughout their entire lifecycle but also calculates the door health anomaly degree using indicators such as curve descent rate and local curvature factor, thereby characterizing its operational trend changes. Finally, based on a dual judgment mechanism of anomaly degree and similarity to adjacent doors, the system identifies potentially abnormal doors and, in conjunction with similar doors, issues unified early warnings in advance. This effectively improves the accuracy of fault prediction and the timeliness of response, providing strong technical support for the intelligent maintenance and operational safety of urban rail transit doors. Attached Figure Description

[0035] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof.

[0036] Figure 1 This is a schematic diagram illustrating the steps of an intelligent monitoring and early warning method for urban rail vehicle doors based on PHM technology according to the present invention.

[0037] Figure 2 This is a schematic diagram of the structure of an intelligent monitoring and early warning system for urban rail vehicle doors based on PHM technology according to the present invention. Detailed Implementation

[0038] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0039] Please see Figure 1 In this first embodiment: a method for intelligent monitoring and early warning of urban rail vehicle doors based on PHM technology is provided, which includes the following steps:

[0040] Step S1: Deploy multimodal sensors at key mechanical nodes of the urban rail vehicle door to collect the operating data of the urban rail vehicle door and synchronize the timestamps.

[0041] Specifically, multimodal sensors are deployed at key mechanical nodes of the urban rail vehicle doors. These multimodal sensors include position sensors, current sensors, speed sensors, and distance sensors. The key mechanical nodes include the connection node between the slide rail and the door panel, the door motor, the door panel, and the slide rail.

[0042] Based on multimodal sensors, the operation data of the urban rail vehicle door is collected. The operation data includes door mechanical position data, door motor current data, door movement data, door V-shape dimension data, and door swing dimension data.

[0043] Furthermore, using the Bluetooth Mesh protocol, a communication connection is established between the position sensor, current sensor, speed sensor, and distance sensor, and the door mechanical position data, door motor current data, door movement data, door V-shaped dimension data, and outward swing dimension data are synchronized and aligned on the timestamp.

[0044] It should be noted that by deploying multimodal sensors for position, current, speed, and distance at key mechanical nodes such as the connection points between the rail track and the door panel, the door motor, and the door panel and rail track, comprehensive perception of the mechanical, electrical, and motion characteristics of the door during operation is achieved. Furthermore, a communication link is established between the multi-source sensors via the Bluetooth Mesh protocol, enabling data timestamp synchronization. This not only improves the spatial resolution and temporal consistency of the door's operational status data but also constructs a complete and dynamically consistent perception foundation. This step supports the accuracy and reliability of data in subsequent data modeling and health analysis processes, avoiding data distortion or synchronization issues. It provides a high-quality, fusionable, and time-consistent data foundation for door status monitoring and intelligent early warning, improving the accuracy and stability of the monitoring model.

[0045] Step S2: Embed a communication interface in the PHM system of the urban rail vehicle door controller and transmit the received operation data to the ground PHM system.

[0046] Specifically, a communication interface is embedded in the PHM system of the urban rail vehicle door controller. The communication interface is used to receive door mechanical position data, door motor current data, door movement data, door V-shaped dimension data and outward swing dimension data after synchronization and alignment based on timestamps; and transmits the received door mechanical position data, door motor current data, door movement data, door V-shaped dimension data and outward swing dimension data to the ground PHM system.

[0047] Furthermore, the t-th timestamp is denoted as TY. t The timestamp TY in the ground PHM system t The mechanical position data, motor current data, movement data, V-shaped dimension data, and outward swing dimension data of the i-th urban rail vehicle door after synchronization are denoted as DMP. i (TY t DMC i (TY t ), DM i (TY t DVS i (TY t ) and ES i (TY t ).

[0048] It should be noted that by embedding a PHM communication interface in the door controller, the aforementioned synchronized multimodal operational data can be uploaded to the ground PHM system in real time and efficiently. This step achieves effective collaboration and closed-loop communication between the vehicle-mounted and ground-based sides, thereby opening up the data path between the perception and analysis layers. This ensures the ground system's remote analysis capability of the door health status, improves the centralization and integrity of data processing, and gives the entire system the advantages of a distributed architecture of "edge acquisition + cloud analysis," improving processing efficiency and reducing the load on the front-end controller. At the same time, it provides a basis for centralized comparison of data from multiple doors.

[0049] Step S3: Construct a multi-source data interaction graph based on the operational data at a single timestamp; calculate the door health of a single urban rail vehicle door at a single timestamp based on the multi-source data interaction graph.

[0050] Specifically, based on timestamp TY t DMP of the mechanical position data of the i-th urban rail vehicle door after synchronization and alignment i (TY t Door motor current data DMC i (TYt ), door movement data DM i (TY t Door V-shape dimension data DVS i (TY t ) and outer swing dimension data ES i (TY t Construct a multi-source data interaction graph, as follows:

[0051] DMP of mechanical position data of car door i (Ty t Let V1(Ty) be the mechanical position node of the multi-source data interaction graph. t ), DMC of the door motor current i (Ty t ) is denoted as the current node V2(TY) of the multi-source data interaction graph. t ), DM the door movement data i (TY t Let V3(Ty) be the moving trajectory node of the multi-source data interaction graph. t ), DVS the V-shaped dimensions of the car door i (TY t ) is denoted as V4(TY) of the V-shaped size node in the multi-source data interaction graph. t ), and transfer the outer swing dimension data ES i (TY t Let V5(TY) be the outer swing dimension node of the multi-source data interaction graph. t ).

[0052] Furthermore, the asymmetric coupling rate of change between any two nodes is taken as an edge of the multi-source data interaction graph, and the formula for calculating the asymmetric coupling rate of change is as follows:

[0053]

[0054] ΔV j (TY t ) = V j (TY t )-V j (TY t-1 )

[0055] in, Indicates the timestamp TY t The rate of change of the asymmetric coupling between the j-th node and the k-th node, ΔV j (TY t ) indicates that the j-th node is from the timestamp TY t-1 To the timestamp TY t The node changes, ΔV k (TYt ) represents the k-th node starting from timestamp TY t-1 To the timestamp TY t The node changes, α represents the preset adjustment parameter, V j (TY t ) represents the timestamp TY t The j-th node at time V j (TY t-1 ) represents the timestamp TY t-1 The j-th node at time.

[0056] Furthermore, based on the aforementioned multi-source data interaction graph, the timestamp TY of the i-th urban rail vehicle door is calculated. t The health status of the car door at that time is calculated using the following formula:

[0057]

[0058] Among them, FHSI i (TY t ) indicates that the i-th urban rail vehicle door is at timestamp TY t The health of the car door at that time, λ jk This represents the preset edge weight between the j-th node and the k-th node. Indicates the timestamp TY t The rate of change of the asymmetric coupling between the j-th node and the k-th node, V j (TY t ) represents the timestamp TY t The j-th node at time, μ j Let σ represent the mean of the j-th node. j Let represent the standard deviation of the j-th node, and ∈ represent the preset error term.

[0059] It should be noted that by constructing a multi-source data interaction graph based on multimodal data at the same timestamp, different types of sensor data are abstracted into different nodes in the graph, and the asymmetric coupling rate of change between nodes is defined as an attribute of the edges, reflecting the dynamic correlation between nodes. This interaction graph effectively extracts the potential coupling and interdependence relationships between various sensor data under complex operating conditions, realizing the transformation from "independent data quantities" to "associated structures".

[0060] Furthermore, by performing weighted analysis on the edges in the interaction graph, a comprehensive judgment of the door's operating status was achieved, namely, calculating the door's health status at the current time point. It can also explore the time-varying coupling mechanism between multimodal data, improve the multidimensional perception capability and dynamic sensitivity of the door status assessment, and form a health assessment mechanism with higher robustness and accuracy. This effectively reduces the probability of misjudgment caused by abnormal single indicators and provides an accurate basis for subsequent trend prediction.

[0061] Step S4: Obtain the door health status under all timestamps, and construct the timestamp-door health status curve for a single urban rail vehicle door; calculate the door health anomaly status for a single urban rail vehicle door based on the timestamp-door health status curve; calculate the similarity between different urban rail vehicle doors; preset thresholds and issue early warnings.

[0062] Specifically, based on the i-th urban rail vehicle door at timestamp TY t Door health index (FHSI) i (TY t To obtain the health status of the i-th urban rail vehicle door across all timestamps, construct the timestamp-door health status curve for the i-th urban rail vehicle door, as follows:

[0063] Construct a timestamp-door health coordinate system. Use all T sequentially arranged timestamps as the x-coordinate of the system, and record the door health corresponding to each of the T timestamps as the y-coordinate. Connect all coordinate points in the timestamp-door health coordinate system sequentially to construct a timestamp-door health curve, denoted as TDH[FHSI]. i (TY t )).

[0064] Furthermore, based on the timestamp-door health curve TDH[FHSI] i (TY t ]] Calculate the door health anomaly degree of the i-th urban rail vehicle door using the following formula:

[0065]

[0066]

[0067] Among them, S i (TY t ) indicates that the i-th urban rail vehicle door is at timestamp TY t-1 To the timestamp TY t The rate of descent of the curve at time FHSI i (TY t-1 ) indicates that the i-th urban rail vehicle door is at timestamp TY t-1 The health of the car door at that time, C i (TY t ) indicates that the i-th urban rail vehicle door is at timestamp TY t Local curvature factor at time, FHSI i (TY t+1 ) indicates that the i-th urban rail vehicle door is at timestamp TY t+1 The health of the car door at that time, A i (TY t) represents the door health abnormality degree of the i-th urban rail vehicle door. ω and ω represent the weighting coefficients of the preset curve descent rate and local curvature factor, respectively.

[0068] The similarity between different urban rail vehicle doors is calculated using the formula: ΔA i (TY t )=|A i (TY t )-A i+1 (TY t )|, where ΔA i (TY t A represents the similarity between the i-th urban rail vehicle door and the (i+1)-th urban rail vehicle door. i+1 (TY t ) represents the door health abnormality of the (i+1)th urban rail vehicle door.

[0069] Furthermore, preset thresholds for door health anomaly and similarity are defined. If the door health anomaly A of the i-th urban rail vehicle door is... i (TY t If the number of times the door health anomaly threshold is greater than or equal to the preset threshold, and the similarity between the door and the (i+1)th urban rail door is less than or equal to the similarity threshold, then the i-th urban rail door is determined to have a health anomaly at the next time stamp and is similar to the (i+1)th urban rail door. If the i-th urban rail door has a health anomaly at the next time stamp, then all doors similar to the i-th urban rail door are obtained, and a unified early warning is issued to remind staff to carry out maintenance in advance.

[0070] It should be noted that by aggregating the health information of individual doors across all timestamps, a timestamp-health curve was constructed. This curve reflects the complete trajectory of the door's health status evolution over time. Based on this, a local curvature factor and a rate of decline factor were further introduced to comprehensively evaluate the volatility and downward trend of the health curve, calculating the door's health anomaly degree. This step uses curve trends and change characteristics for dynamic discrimination and identifies potential group failure risks through similarity analysis between doors, constructing a more predictive early warning mechanism and implementing a risk propagation model between similar doors, thus enhancing the system's ability to identify sudden failures and systemic degradation.

[0071] This step not only enables precise early warning at the single-door level, but also allows for early identification of door failure risks with similar structures or operating conditions, supporting maintenance personnel in making batch inspection and planned maintenance decisions, thereby reducing the overall system's operational risks.

[0072] Please see Figure 2In this second embodiment, a smart monitoring and early warning system for urban rail vehicle doors based on PHM technology is provided. The system includes: a data acquisition and synchronization module, a data receiving and transmission module, an interactive graph and health calculation module, and a curve construction and anomaly analysis and early warning module.

[0073] Furthermore, the data acquisition and synchronization module deploys multimodal sensors at key mechanical nodes of the urban rail vehicle door to collect the operating data of the urban rail vehicle door and synchronize the timestamps.

[0074] Furthermore, the data receiving and transmission module embeds a communication interface in the PHM system of the urban rail vehicle door controller and transmits the received operational data to the ground PHM system.

[0075] Furthermore, the interaction graph and health calculation module: constructs a multi-source data interaction graph based on the operating data at a single timestamp; and calculates the health of a single urban rail vehicle door at a single timestamp based on the multi-source data interaction graph.

[0076] Furthermore, the curve construction and anomaly analysis and early warning module: obtains the door health status under all timestamps, constructs a timestamp-door health status curve for a single urban rail vehicle door; calculates the door health anomaly status for a single urban rail vehicle door based on the timestamp-door health status curve; calculates the similarity between different urban rail vehicle doors; presets a threshold, and provides early warning.

[0077] It should 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 process, method, article, or apparatus.

[0078] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A PHM technology-based intelligent monitoring and early warning method for urban rail vehicle doors, characterized in that, The method comprises the following steps: Step S1: deploying multi-modal sensors at key mechanical nodes of the urban rail vehicle door, collecting operation data of the urban rail vehicle door, and performing timestamp synchronization; Step S2: embedding a communication interface in a PHM system in the urban rail vehicle door controller, and transmitting the received operation data to a ground PHM system; Step S3: based on the operation data at a single timestamp, constructing a multi-source data interaction graph; based on the multi-source data interaction graph, calculating the door health degree of a single urban rail vehicle door at a single timestamp; Step S4: obtaining the door health degree at all timestamps, constructing a timestamp-door health degree curve of a single urban rail vehicle door; based on the timestamp-door health degree curve, calculating the door health abnormality degree of the single urban rail vehicle door; calculating the similarity between different urban rail vehicle doors; presetting a threshold, and performing early warning; The specific implementation process of step S3 comprises: Based on the tth time stamp Door mechanical position data of the ith urban rail vehicle door after the down-sync alignment Door motor current data Door movement data V-shaped size data of the door and the outer swing size data , construct a multi-source data interaction graph, as follows: translating the vehicle door mechanical position data a mechanical position node of the multi-source data interaction graph translating the vehicle door motor current data a current node of the multi-source data interaction graph translating the vehicle door movement data a movement trajectory node of the multi-source data interaction graph translating the vehicle door V-dimension data a V-dimension node of the multi-source data interaction graph translating the vehicle door outswing dimension data an outswing dimension node of the multi-source data interaction graph ; The asymmetric coupling change rate between any two nodes is taken as an edge of the multi-source data interaction graph, and the calculation formula of the asymmetric coupling change rate is as follows: wherein, denotes the rate of change of the asymmetric coupling between the jth node and the kth node at the time stamp , denotes the node change of the jth node from the time stamp to the time stamp , denotes the node change of the kth node from the time stamp to the time stamp , denotes a preset adjustment parameter, denotes the jth node at the time stamp , denotes the jth node at the time stamp , denotes the t-1th time stamp; The specific implementation process of step S3 further comprises: Based on the multi-source data interaction graph, the timestamp of the i-th urban rail vehicle door is calculated. The health status of the car door at that time is calculated using the following formula: wherein, denotes the door health of the ith urban rail door at the timestamp denotes the door health of the ith urban rail door at the timestamp denotes the preset edge weight between the jth node and the kth node, denotes the asymmetric coupling change rate between the jth node and the kth node at the timestamp denotes the asymmetric coupling change rate between the jth node and the kth node at the timestamp denotes the jth node at the timestamp denotes the jth node at the timestamp denotes the mean of the jth node, denotes the standard deviation of the jth node, denotes the preset error term; The specific implementation process of step S4 further comprises: Based on the timestamp-door health degree curve, the door health abnormality degree of the i-th urban rail vehicle door is calculated, and the calculation formula is as follows: wherein, represents a curve drop rate of the ith urban rail vehicle door at a time stamp to a time stamp , represents a door health of the ith urban rail vehicle door at a time stamp , represents a local curvature factor of the ith urban rail vehicle door at a time stamp , represents a door health of the ith urban rail vehicle door at a time stamp , represents a door health abnormality of the ith urban rail vehicle door, and respectively represent a preset weight coefficient of the curve drop rate and the local curvature factor. The similarity between different urban rail car doors is calculated, and the calculation formula is: wherein, represents the similarity between the ith urban rail car door and the i+1th urban rail car door, represents the door health abnormality degree of the i+1th urban rail car door; The preset door health abnormality threshold and the similarity threshold, if the door health abnormality of the ith urban rail door The number of times greater than or equal to the door health abnormality threshold is greater than or equal to the preset number threshold, and the similarity between the ith urban rail door and the i+1th urban rail door is less than or equal to the similarity threshold, it is determined that the ith urban rail door has a health abnormality at the next timestamp, and is similar to the i+1th urban rail door; if the ith urban rail door has a health abnormality at the next timestamp, all doors similar to the ith urban rail door are obtained, and unified early warning is performed in advance to remind the staff to perform maintenance in advance. 2.The intelligent monitoring and early warning method for urban rail vehicle door based on PHM technology according to claim 1, characterized in that, The specific implementation process of step S1 comprises: Deploying multi-modal sensors at key mechanical nodes of the urban rail vehicle door, wherein the multi-modal sensors comprise position sensors, current sensors, speed sensors, and distance sensors, and the key mechanical nodes comprise a sliding rail and door leaf connection node, a door motor, a door leaf, and a sliding rail; Based on the multi-modal sensors, operation data of the urban rail vehicle door is collected, wherein the operation data comprises door mechanical position data, door motor current data, door movement data, door V-shaped size data, and outer swing size data; Using a Bluetooth Mesh protocol, communication connections are established between the position sensors, current sensors, speed sensors, and distance sensors, and the door mechanical position data, door motor current data, door movement data, door V-shaped size data, and outer swing size data are synchronized and aligned in time stamps. 3.The intelligent monitoring and early warning method for urban rail vehicle door based on PHM technology according to claim 2, characterized in that, The specific implementation process of step S2 comprises: Embedding a communication interface in a PHM system in the urban rail vehicle door controller, wherein the communication interface is used to receive the door mechanical position data, door motor current data, door movement data, door V-shaped size data, and outer swing size data that are synchronized and aligned in time stamps; and transmitting the received door mechanical position data, door motor current data, door movement data, door V-shaped size data, and outer swing size data to a ground PHM system; Let the t-th time stamp be denoted as Let the time stamp in the ground PHM system be denoted as Let the door mechanical position data, the door motor current data, the door movement data, the door V-shaped size data and the door outside swing size data of the i-th urban rail door after the synchronization alignment be denoted as , , , and .

4. The intelligent monitoring and early warning method for urban rail vehicle doors based on PHM technology according to claim 3, characterized in that, The specific implementation process of step S4 comprises: Based on the timestamp of the i-th urban rail vehicle door Car door health at time To obtain the health status of the i-th urban rail vehicle door across all timestamps, construct the timestamp-door health status curve for the i-th urban rail vehicle door, as follows: A time stamp-door health coordinate system is constructed, all T time stamps in sequence are taken as horizontal coordinates of the time stamp-door health coordinate system, and door health corresponding to all T time stamps in sequence is taken as vertical coordinates of the time stamp-door health coordinate system, all coordinate points in the time stamp-door health coordinate system are sequentially connected, a time stamp-door health curve is constructed, and is denoted as .

5. A PHM technology-based intelligent monitoring and early warning system for urban rail vehicle doors, which implements a PHM technology-based intelligent monitoring and early warning method for urban rail vehicle doors according to any one of claims 1-4, characterized in that, The system comprises a data acquisition and synchronization module, a data receiving and transmission module, an interaction graph and health degree calculation module, and a curve construction and abnormality analysis and early warning module; The data acquisition and synchronization module: deploying multi-modal sensors at key mechanical nodes of the urban rail vehicle door, collecting operation data of the urban rail vehicle door, and performing timestamp synchronization; The data receiving and transmitting module: embeds a communication interface in a PHM system in a city rail vehicle door controller, and transmits the received operation data to a ground PHM system; The interaction graph and health degree calculation module: constructs a multi-source data interaction graph based on operation data at a single timestamp; and calculates a door health degree of a single city rail vehicle door at the single timestamp based on the multi-source data interaction graph; The curve construction and abnormality analysis and early warning module: obtains door health degrees at all timestamps, constructs a timestamp-door health degree curve of a single city rail vehicle door, calculates a door health abnormality degree of the single city rail vehicle door based on the timestamp-door health degree curve, calculates similarities between different city rail vehicle doors, and pre-sets a threshold value and performs early warning. 6.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is characterized in that: The processor executes the computer program to implement the steps of the city rail vehicle door intelligent monitoring and early warning method based on the PHM technology in any one of claims 1-4.

7. A computer readable storage medium having stored thereon a computer program, characterized in that: The computer program is executed by the processor to implement the steps of the city rail vehicle door intelligent monitoring and early warning method based on the PHM technology in any one of claims 1-4.

Images