Contact rail expansion joint state information online detection method and system

Abstract

The invention provides an online detection method and system for state information of a contact rail expansion joint, and relates to the technical field of track detection, and the method comprises the steps: obtaining multi-dimensional state data, including temperature state data, gap distance data and sliding motion data, of the contact rail expansion joint; preprocessing the multi-dimensional state data to generate standard state data; according to different operation scene characteristics of the rail transit, a differential data transmission strategy is adopted to upload standard state data; constructing an edge-cloud collaborative analysis model, extracting multi-dimensional state features of the standard state data, performing state analysis, anomaly recognition and risk level evaluation on the multi-dimensional state features, and generating a state report; and based on the state report, triggering a corresponding multi-level early warning mechanism and pushing early warning information and maintenance suggestions to the management end, thereby realizing comprehensive and real-time online monitoring of the state information of the expansion joint of the contact rail.

Description

Technical Field

[0001] This invention relates to the field of track inspection technology, and in particular to an online detection method and system for the status information of contact rail expansion joints. Background Technology

[0002] With the rapid development of urban rail transit and the continuous increase in operating mileage, the contact rail power supply system, as a core component of rail transit, directly affects the safe and efficient operation of rail transit. The contact rail expansion joint, as a key component of the contact rail power supply system, plays a crucial role in adapting to the thermal expansion and contraction of the track and ensuring stable current transmission; therefore, the stability of its operating state is of paramount importance.

[0003] Currently, traditional contact rail expansion joint condition monitoring mainly relies on manual periodic inspections. This method not only struggles to achieve real-time monitoring of the expansion joint's operational status, easily overlooking potential safety hazards caused by abnormal clearance, slippage, or excessive temperature, but also suffers from inaccuracies and inefficiencies due to factors such as the inspector's skill level and working conditions, failing to meet the demands of efficient modern rail transit operations. Furthermore, existing research largely focuses on structural optimization and basic performance improvement of contact rail expansion joints, with research on multi-parameter online monitoring technology still in its early stages. Some studies only achieve single-parameter monitoring, exhibiting significant shortcomings in the real-time performance and stability of data transmission and processing, and a complete, large-scale applicable online monitoring system has yet to be developed.

[0004] Therefore, it is necessary to provide an online detection method and system for the status information of contact rail expansion joints to solve the above-mentioned technical problems. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention provides an online detection method and system for the status information of contact rail expansion joints. This system solves the problems of existing contact rail expansion joint detection methods, such as difficulty in achieving real-time monitoring, limited detection parameters, insufficient stability in data transmission and processing, and inability to meet the requirements of efficient and safe operation of rail transit for accurate monitoring of expansion joint status.

[0006] This invention provides an online detection method for the status information of a contact rail expansion joint, the method comprising: The multi-dimensional expansion joint status data of the contact rail expansion joint is acquired according to a preset acquisition frequency. The multi-dimensional expansion joint status data includes temperature status data, gap distance data, and sliding action data. The multi-dimensional expansion joint status data is preprocessed to generate standard expansion joint status data; Based on the characteristics of different operating scenarios of rail transit, a differentiated data transmission strategy is adopted to upload the status data of the standard expansion joint; An edge-cloud collaborative analysis model is constructed to extract multi-dimensional expansion joint status features from the standard expansion joint status data. The multi-dimensional expansion joint status features are then analyzed to identify expansion joint anomalies and assess expansion joint risk levels, generating an expansion joint status report. Based on the expansion joint status report, the corresponding multi-level early warning mechanism for the expansion joint is triggered, and early warning information and maintenance suggestions for the expansion joint are pushed to the management terminal.

[0007] Preferably, the temperature status data includes contact rail temperature data and ambient temperature data; the gap distance data is the real-time gap data at the mating point of the contact rail expansion joint; and the sliding action data is the smoothness correlation data of the relative sliding of the contact rail expansion joint. Specifically, when collecting the temperature status data, a contact-type platinum resistance sensor or a non-contact infrared temperature sensor is used; when collecting the gap distance data, a laser rangefinder is used.

[0008] Preferably, the step of preprocessing the multi-dimensional expansion joint status data to generate standard expansion joint status data specifically includes: Outlier identification and removal are performed on the multi-dimensional expansion joint status data based on the Grubbs criterion. Among them, the sample mean of the gap distance data and the sample standard deviation of the gap distance data Calculate the i-th data point in the gap distance data. Grubbs statistic The Grubbs statistic Compared with the preset significance level The corresponding Grubbs critical value Comparison, , n represents the amount of data in the gap distance data; if Then determine the i-th data point Outliers were identified and removed. Missing value detection is performed on the gap distance data after outlier removal, and linear interpolation is used to calculate the gap distance. Data completion for missing positions at specific times The corresponding calculation formula is as follows: In the formula, , , This indicates the time nodes for collecting three consecutive data points in the gap distance data. ; Indicating gap distance data in Valid data at any given time; Indicating gap distance data in Valid data at any given time; The calculation methods for the temperature status data and the sliding motion data are the same as above; The multi-dimensional expansion joint status data is normalized using the Z-score standardization method and then integrated according to the collection time nodes to generate the standard expansion joint status data.

[0009] Preferably, the step of uploading the standard expansion joint status data using a differentiated data transmission strategy based on the characteristics of different rail transit operating scenarios specifically includes: The typical operating scenarios of rail transit are divided according to the characteristics of different operating scenarios of rail transit. The typical operating scenarios of rail transit include normal operating peak hours, normal operating off-peak hours, nighttime shutdown and maintenance periods, and extreme weather operation periods. Construct a scenario-transmission parameter mapping relationship and dynamically adjust the transmission frequency under different typical rail transit operation scenarios. The corresponding calculation formula is as follows: In the formula, Indicates the preset reference transmission frequency; This represents the real-time demand coefficient for different typical rail transit operating scenarios, where r is the time period during extreme weather conditions. ;r represents peak operating hours. ;r represents the off-peak period during normal operation. When r represents the nighttime shutdown for maintenance, ; The urgency coefficient represents the condition data of the standard expansion joint; Indicates the weight of the real-time requirement coefficient for the scenario; This indicates the weight of the data urgency coefficient. .

[0010] Preferably, the construction of the edge-cloud collaborative analysis model to extract multi-dimensional expansion joint status features from the standard expansion joint status data specifically includes: The edge-cloud collaborative analysis model includes an edge processing end and a cloud analysis end, and adopts a layered collaborative architecture of real-time edge preprocessing and global cloud fusion. The edge processing terminal receives the standard expansion joint status data, divides the data into segments according to a preset time window, extracts the spatial coupling features of the standard expansion joint status data through an improved CNN model, and extracts the temporal evolution features of the standard expansion joint status data through a Transformer model. A two-dimensional fusion strategy based on dynamic weight adaptation and feature attention optimization is used to fuse the spatial coupling features and the temporal evolution features to generate a local feature vector of the expansion joint. The corresponding calculation formula is as follows: In the formula, This represents the spatial coupling features extracted by the improved CNN model; This represents the temporal evolution features extracted by the Transformer model; Dynamic weights representing spatial coupling characteristics; Dynamic weights representing temporal evolution characteristics ; Feature attention weights representing spatial coupling characteristics; Feature attention weights represent temporal evolution characteristics.

[0011] Preferably, the cloud-based analysis terminal analyzes the local feature vector of the expansion joint. Global feature fusion is performed to generate the multi-dimensional expansion joint state features. Specifically, it includes: The cloud-based analysis terminal receives the local feature vector of the expansion joint uploaded by the edge processing terminal. An expansion joint feature vector index library was established based on the contact rail track partitioning and expansion joint numbering. Based on the preset contact rail area coupling coefficient and historical fault propagation weight parameters, an improved graph attention network is used to analyze the local feature vectors of expansion joints in the expansion joint feature vector index library. Spatial correlation mining was performed to extract global spatial features of the expansion joint. ; The spatial global features of the expansion joint at different time points are analyzed using a time-series sliding window mechanism. Perform temporal correlation fusion to generate the multi-dimensional expansion joint state features. .

[0012] Preferably, the step of performing expansion joint status analysis, expansion joint anomaly identification, and expansion joint risk level assessment on the multi-dimensional expansion joint status characteristics to generate an expansion joint status report specifically includes: Based on the preset expansion joint feature-state mapping rule library, the multi-dimensional expansion joint state features are analyzed in layers to generate an expansion joint feature-state relationship matrix, and the expansion joint state analysis results are output based on the expansion joint feature-state relationship matrix. The isolated forest algorithm is used to perform abnormal pattern matching and identification on the multi-dimensional expansion joint state features by combining the feature-state relationship matrix of the expansion joint and the preset threshold range of the normal state features of the expansion joint, marking the abnormal feature type and expansion joint number, and outputting the abnormal identification result of the expansion joint. A multi-factor risk assessment model is constructed. The abnormal identification results of the expansion joint are input, and the severity of the abnormal characteristics of the expansion joint, the service life of the expansion joint, the operating load of the line where the expansion joint is located, and the historical fault propagation coefficient are integrated. The weight of the assessment indicators is determined by the analytic hierarchy process, the comprehensive risk assessment value of the expansion joint is calculated, and the risk level of the expansion joint is determined according to the preset risk level classification standard.

[0013] Preferably, the step of performing hierarchical analysis on the multi-dimensional expansion joint state features based on a preset expansion joint feature-state mapping rule base to generate an expansion joint feature-state relationship matrix specifically includes: Among them, the relationship value between the v-th multi-dimensional expansion joint state feature and the j-th expansion joint operating state in the expansion joint feature-state relationship matrix M. The calculation formula is as follows: In the formula, The hierarchical weight coefficient represents the state feature of the v-th multi-dimensional expansion joint. Represents the dynamic correction coefficient for the v-th multidimensional expansion joint state characteristic; The sensitivity adjustment coefficient representing the feature-state mapping; This represents the basic correlation between the v-th multidimensional expansion joint state characteristics and the j-th expansion joint operating state.

[0014] An online detection system for the status information of contact rail expansion joints, the system comprising: The data acquisition module is used to acquire multi-dimensional expansion joint status data of the contact rail expansion joint according to a preset acquisition frequency. The multi-dimensional expansion joint status data includes temperature status data, gap distance data, and sliding action data. The data processing module is used to preprocess the multi-dimensional expansion joint status data to generate standard expansion joint status data. The data upload module is used to upload the status data of the standard expansion joint according to the characteristics of different operating scenarios of rail transit, using a differentiated data transmission strategy. The status detection module is used to construct an edge-cloud collaborative analysis model, extract multi-dimensional expansion joint status features from the standard expansion joint status data, and perform expansion joint status analysis, expansion joint anomaly identification, and expansion joint risk level assessment on the multi-dimensional expansion joint status features to generate an expansion joint status report. The early warning and maintenance module is used to trigger the corresponding multi-level early warning mechanism for the expansion joint based on the expansion joint status report and push the expansion joint early warning information and expansion joint maintenance suggestions to the management terminal.

[0015] Compared with related technologies, the online detection method and system for the status information of contact rail expansion joints provided by the present invention has the following beneficial effects: This invention acquires multi-dimensional expansion joint status data of the contact rail expansion joint according to a preset acquisition frequency. This multi-dimensional expansion joint status data includes temperature status data, gap distance data, and sliding action data. The multi-dimensional expansion joint status data is preprocessed to generate standard expansion joint status data. Based on the characteristics of different rail transit operating scenarios, a differentiated data transmission strategy is adopted to upload the standard expansion joint status data. An edge-cloud collaborative analysis model is constructed to extract the multi-dimensional expansion joint status features from the standard expansion joint status data. This model is then used to perform expansion joint status analysis, expansion joint anomaly identification, and expansion joint risk level assessment, generating an expansion joint status report. Based on the expansion joint status report, a corresponding multi-level early warning mechanism is triggered, and early warning information and maintenance suggestions are pushed to the management end. This achieves comprehensive and real-time online monitoring of the contact rail expansion joint status information, effectively overcoming the limitations of traditional manual inspection and existing single-parameter detection technologies.

[0016] This invention overcomes the limitations of traditional single-parameter detection by collecting multi-dimensional expansion joint status data, including temperature, gap distance, and sliding motion data. It comprehensively covers key operational indicators of expansion joints, enabling timely detection of potential safety hazards such as temperature anomalies, gap deviations, and sliding jamming, avoiding the drawbacks of manual periodic inspections which struggle to monitor in real-time and are prone to overlooking problems. This invention ensures data accuracy and consistency through outlier removal, missing value completion, and standardized preprocessing of the multi-dimensional expansion joint status data. Based on different scenarios such as peak and off-peak operation, nighttime maintenance shutdowns, and extreme weather, this invention employs differentiated data transmission strategies to dynamically adapt to the real-time requirements of each scenario, balancing data transmission efficiency and stability, and avoiding data delays or redundant transmission. Finally, this invention extracts multi-dimensional expansion joint status features through an edge-cloud collaborative analysis model, accurately performing expansion joint status analysis, anomaly identification, and risk level assessment, replacing subjective manual judgment and improving the accuracy of detection and analysis. This invention triggers a multi-level early warning mechanism for expansion joints based on expansion joint status reports and pushes early warning information and maintenance suggestions for expansion joints. This helps maintenance personnel to handle faults in advance, ensures the safe and stable operation of the contact rail power supply system, reduces unnecessary maintenance work and costs, and meets the needs of efficient and safe operation of modern rail transit. Attached Figure Description

[0017] Figure 1 A flowchart of an online detection method for the status information of a contact rail expansion joint provided in an embodiment of the present invention; Figure 2 A system block diagram of an online detection system for the status information of a contact rail expansion joint provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of the hardware structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, 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.

[0019] like Figure 1 The diagram shown is a flowchart of an online detection method for the status information of a contact rail expansion joint provided by an embodiment of the present invention. Figure 1 The execution entity of the method shown can be a software and / or hardware device. The execution entity of this application can include, but is not limited to, at least one of the following: user equipment, network equipment, etc. User equipment can include, but is not limited to, computers, smartphones, personal digital assistants (PDAs), and the aforementioned electronic devices. Network equipment can include, but is not limited to, a single network server, a server group consisting of multiple network servers, or a cloud based on cloud computing consisting of a large number of computers or network servers. Cloud computing is a type of distributed computing, consisting of a super virtual computer composed of a group of loosely coupled computers. This embodiment does not limit this. Steps S1 to S5 are detailed as follows: S1. Acquire multi-dimensional expansion joint status data of the contact rail expansion joint according to a preset acquisition frequency. The multi-dimensional expansion joint status data includes temperature status data, gap distance data, and sliding action data. The temperature status data includes contact rail temperature data and ambient temperature data; the gap distance data is the real-time spacing data at the mating point of the contact rail expansion joint; the sliding motion data is the smoothness correlation data of the relative sliding of the contact rail expansion joint. Specifically, when collecting the temperature status data, a contact-type platinum resistance sensor or a non-contact infrared temperature sensor is used; when collecting the gap distance data, a laser rangefinder is used.

[0020] When collecting temperature status data, a contact-type platinum resistance sensor is preferentially used. This sensor directly adheres to the contact rail surface, meeting the direct acquisition requirements of contact rail temperature data and ensuring stable contact with the contact rail to obtain accurate temperature information. When high voltage interference or space limitations in local areas of the contact rail make direct contact inconvenient, a non-contact infrared temperature sensor is used. This non-contact detection method simultaneously acquires ambient temperature data and indirect temperature information from the contact rail, with the two types of sensors working together to achieve comprehensive coverage of temperature status data. When collecting gap distance data, a laser rangefinder sensor is used. This sensor is precisely aligned with the contact rail expansion joint mating point, adapting to the dynamic acquisition scenario of real-time gap data and avoiding the impact of mechanical contact on the normal sliding of the expansion joint. The acquisition of sliding motion data is achieved through a detection component linked to the sliding parts of the expansion joint, capturing in real time the resistance, jamming, and other smoothness-related information during the relative sliding process, ensuring the complete acquisition of multi-dimensional expansion joint status data.

[0021] S2, perform data preprocessing on the multi-dimensional expansion joint status data to generate standard expansion joint status data; The step of preprocessing the multi-dimensional expansion joint status data to generate standard expansion joint status data specifically includes: Outlier identification and removal are performed on the multi-dimensional expansion joint status data based on the Grubbs criterion. Among them, the sample mean of the gap distance data and the sample standard deviation of the gap distance data Calculate the i-th data point in the gap distance data. Grubbs statistic The Grubbs statistic Compared with the preset significance level The corresponding Grubbs critical value Comparison, , n represents the amount of data in the gap distance data; if Then determine the i-th data point Outliers were identified and removed. Missing value detection is performed on the gap distance data after outlier removal, and linear interpolation is used to calculate the gap distance. Data completion for missing positions at specific times The corresponding calculation formula is as follows: In the formula, , , This indicates the time nodes for collecting three consecutive data points in the gap distance data. ; Indicating gap distance data in Valid data at any given time; Indicating gap distance data in Valid data at any given time; The calculation methods for the temperature status data and the sliding motion data are the same as above; The multi-dimensional expansion joint status data is normalized using the Z-score standardization method and then integrated according to the collection time nodes to generate the standard expansion joint status data.

[0022] Among these methods, the Grubbs standard identifies outliers based on the statistical characteristics of multi-dimensional expansion joint status data, effectively eliminating invalid data caused by temporary sensor malfunctions, instantaneous external environmental interference, and other factors, thus ensuring data reliability from the source. Linear interpolation addresses the potential data gaps that may arise after outlier removal by using adjacent valid data before and after the missing time point, combined with time series correlations, to fill in the missing data, maintaining the temporal continuity of the data and avoiding the impact of data gaps on the analysis results. The Z-score standardization method eliminates analytical biases caused by unit differences in different types of data by unifying the dimensions of multi-dimensional data. Finally, the standardized data is correlated and integrated according to the acquisition time nodes, ensuring the consistency and comparability of the generated standard expansion joint status data.

[0023] In practical applications, taking the detection of contact rail expansion joints in urban subway tunnels as an example, the high-voltage environment inside the tunnel can easily cause abnormal values ​​in the gap distance data collected by the laser ranging sensor. In this case, the Grubbs criterion is used to accurately identify and remove such abnormal data. If the acquisition unit experiences a short-term power outage due to temporary maintenance inside the tunnel, resulting in missing temperature status data for a certain period, linear interpolation is used to fill in the missing parts based on the effective temperature status data before and after the power outage. After Z-score standardization to unify the data dimensions and integration according to the acquisition time, the generated standard expansion joint status data can accurately support the cloud-based analysis of the operating status of the expansion joints in this tunnel section, ensuring that the detection results meet the actual working conditions.

[0024] S3. Based on the characteristics of different operating scenarios of rail transit, the standard expansion joint status data is uploaded using a differentiated data transmission strategy. The method of uploading the standard expansion joint status data using a differentiated data transmission strategy based on the characteristics of different rail transit operating scenarios specifically includes: The typical operating scenarios of rail transit are divided according to the characteristics of different operating scenarios of rail transit. The typical operating scenarios of rail transit include normal operating peak hours, normal operating off-peak hours, nighttime shutdown and maintenance periods, and extreme weather operation periods. Construct a scenario-transmission parameter mapping relationship and dynamically adjust the transmission frequency under different typical rail transit operation scenarios. The corresponding calculation formula is as follows: In the formula, Indicates the preset reference transmission frequency; This represents the real-time demand coefficient for different typical rail transit operating scenarios, where r is the time period during extreme weather conditions. ;r represents peak operating hours. ;r represents the off-peak period during normal operation. When r represents the nighttime shutdown for maintenance, ; The urgency coefficient represents the condition data of the standard expansion joint; Indicates the weight of the real-time requirement coefficient for the scenario; This indicates the weight of the data urgency coefficient. .

[0025] The classification of typical rail transit operation scenarios is based on the differences in operating load, safety requirements, and external environmental conditions of the contact rail power supply system at different times, ensuring that the scenario classification closely matches actual operating conditions. The scenario-transmission parameter mapping relationship is the core link for implementing the strategy. By establishing a correspondence between scenarios and transmission frequencies, transmission parameter adjustments can be dynamically adapted to the scenario, avoiding resource waste or data delays caused by traditional transmission modes.

[0026] The scenario-based real-time requirement coefficient is set according to the scenario's requirements for data timeliness. For example, expansion joints are more prone to failure during extreme weather conditions, while during peak operating hours, high-frequency data transmission is necessary to monitor the status of expansion joints in real time, as the safety of a large number of passengers and operational efficiency are crucial. Therefore, the coefficient is set higher. The data urgency coefficient is used to supplement scenario requirements. If there are signs of anomalies in the standard expansion joint status data, even in scenarios with lower real-time requirements, the transmission priority can still be increased by raising this coefficient.

[0027] Understandably, the setting of the weights for the scenario real-time demand coefficient and the data urgency coefficient is to balance the inherent needs of the scenario with sudden data events, ensuring that the transmission strategy conforms to regular operating rules while also being able to cope with emergency situations at the data level, and ultimately achieving a balance between the real-time performance, stability, and resource rationality of standard expansion joint status data uploads in different scenarios.

[0028] By employing the above methods, the strategy for uploading standard expansion joint status data can be precisely matched with the characteristics of typical rail transit operating scenarios. This not only meets the differentiated requirements for real-time data in different scenarios but also allows for flexible adjustment of priorities based on data urgency, avoiding delays in critical data or unnecessary redundant data transmission.

[0029] S4. Construct an edge-cloud collaborative analysis model, extract multi-dimensional expansion joint status features from the standard expansion joint status data, and perform expansion joint status analysis, expansion joint anomaly identification, and expansion joint risk level assessment on the multi-dimensional expansion joint status features to generate an expansion joint status report. The construction of the edge-cloud collaborative analysis model extracts multi-dimensional expansion joint status features from the standard expansion joint status data, specifically including: The edge-cloud collaborative analysis model includes an edge processing end and a cloud analysis end, and adopts a layered collaborative architecture of real-time edge preprocessing and global cloud fusion. The edge processing terminal receives the standard expansion joint status data, divides the data into segments according to a preset time window, extracts the spatial coupling features of the standard expansion joint status data through an improved CNN model, and extracts the temporal evolution features of the standard expansion joint status data through a Transformer model. A two-dimensional fusion strategy based on dynamic weight adaptation and feature attention optimization is used to fuse the spatial coupling features and the temporal evolution features to generate a local feature vector of the expansion joint. The corresponding calculation formula is as follows: In the formula, This represents the spatial coupling features extracted by the improved CNN model; This represents the temporal evolution features extracted by the Transformer model; Dynamic weights representing spatial coupling characteristics; Dynamic weights representing temporal evolution characteristics ; Feature attention weights representing spatial coupling characteristics; Feature attention weights represent temporal evolution characteristics.

[0030] The edge-cloud collaborative analysis model employs a layered collaborative architecture that combines real-time edge preprocessing with global fusion in the cloud, clearly defining the functional division between the edge processing and cloud analysis ends. Leveraging its proximity deployment, the edge processing end rapidly receives standard expansion joint status data, avoiding the latency issues associated with long-distance data transmission and enabling real-time preliminary data processing. Meanwhile, the cloud analysis end, with its powerful computing and storage capabilities, performs large-scale, multi-node global feature fusion, compensating for the limited computing power and analytical perspective of the edge processing end. This collaborative approach ensures both the timeliness of data processing and the comprehensiveness of feature analysis.

[0031] By dividing the data into segments using a preset time window and considering the temporal continuity of the contact rail expansion joint status data, the continuously acquired data stream is cut into regular processing units. This avoids processing delays caused by excessively large data volumes in a single segment, and also prevents fragmentation of feature information caused by excessively short segments.

[0032] An improved CNN model is used for extracting spatial coupling features. Targeting the spatial correlation between temperature state data, gap distance data, and sliding action data, it strengthens the capture of the spatial coupling relationship of standard expansion joint state data by optimizing the network convolution kernel and feature mapping mechanism. Compared to traditional CNNs, it is more suitable for extracting state features of multi-parameter synergy in expansion joints. The Transformer model is used for extracting temporal evolution features. Leveraging its ability to capture long-term dependencies in temporal data, it uncovers the changing patterns of standard expansion joint state data over time, such as the evolution trend of sliding action smoothness at different time periods, fully presenting the temporal dynamic characteristics of the expansion joint state.

[0033] A dual-dimensional fusion strategy combining dynamic weight adaptation and feature attention optimization is key to achieving efficient fusion of spatial and temporal features. Dynamic weight adaptation adjusts the weights of the two types of features based on the actual proportion of features in the data. For example, when temporal changes are more significant in a certain period, the dynamic weight of the temporal evolution feature is increased, avoiding feature bias caused by fixed weights. Feature attention optimization automatically focuses on feature components that significantly affect the state of the expansion joint. For instance, when the gap distance is abnormal, the role of the gap-related component in the spatial coupling feature is strengthened. The resulting local feature vector of the expansion joint accurately reflects the core state information of a single-node expansion joint.

[0034] By adopting the above approach and relying on the layered collaborative architecture of edge-cloud collaboration, the edge processing end can quickly receive standard expansion joint status data. After dividing the data into regular segments with the help of a preset time window, the spatial coupling features and temporal evolution features of the data are accurately captured by the improved CNN model and the Transformer model respectively. Then, the weights of the two types of features are dynamically balanced by a two-dimensional fusion strategy and key components are focused to generate a local feature vector that can accurately reflect the core status of a single-node expansion joint.

[0035] The cloud-based analysis terminal analyzes the local feature vector of the expansion joint. Global feature fusion is performed to generate the multi-dimensional expansion joint state features. Specifically, it includes: The cloud-based analysis terminal receives the local feature vector of the expansion joint uploaded by the edge processing terminal. An expansion joint feature vector index library was established based on the contact rail track partitioning and expansion joint numbering. Based on the preset contact rail area coupling coefficient and historical fault propagation weight parameters, an improved graph attention network is used to analyze the local feature vectors of expansion joints in the expansion joint feature vector index library. Spatial correlation mining was performed to extract global spatial features of the expansion joint. ; The spatial global features of the expansion joint at different time points are analyzed using a time-series sliding window mechanism. Perform temporal correlation fusion to generate the multi-dimensional expansion joint state features. .

[0036] Among them, the contact rail track zoning and expansion joint numbering are used to assign clear spatial identifiers to the local feature vectors of expansion joints, ensuring that each vector accurately corresponds to its physical location on the actual track. The contact rail track zoning is divided according to the operating section of the track, and the expansion joint numbering assigns a unique identifier to each expansion joint. The combination of these two allows the local feature vectors of expansion joints to be traced back to the specific inspection object. Based on this, the expansion joint feature vector index library can systematically store the local feature vectors of expansion joints at each node, avoiding data confusion among multiple nodes.

[0037] The contact rail region coupling coefficient is used to quantify the degree of state correlation of expansion joints within the same or adjacent track sections, such as the state impact of adjacent joints due to track structure linkage. The historical fault propagation weight parameter is set based on the propagation path and impact range of past faults among different expansion joints, highlighting the priority of expansion joint associations along easily propagating fault paths. The improved graph attention network relies on these two parameters to specifically mine the spatial correlation of local feature vectors of expansion joints in the expansion joint feature vector index library. It automatically focuses on the correlations that play a key role in the overall state, rather than indiscriminately calculating all correlations, thereby accurately extracting the global spatial features of expansion joints.

[0038] The time-series sliding window mechanism extracts the spatial global features of different nodes along the time dimension and integrates their evolution over time. By sliding the window to cover continuous time segments, it captures the temporal change trend of spatial global features, so that the final multi-dimensional expansion joint state features include not only the spatial correlation information between joints, but also the dynamic changes in state over time, comprehensively reflecting the overall operating status of the contact rail expansion joint.

[0039] In practical applications, taking the contact rail inspection of Metro Line 2 in a certain city as an example, the line is divided into three contact rail sections: the eastern section, the central section, and the western section. Each section's expansion joints are assigned a unique number in the format "section number - joint sequence number." The cloud-based analysis terminal receives the local feature vectors of the expansion joints uploaded by each edge processing terminal and establishes an expansion joint feature vector index library. When a local feature vector of an expansion joint in the central section shows an abnormal temperature, the cloud-based analysis terminal, based on preset contact rail area coupling coefficients and historical fault propagation weight parameters (such as higher correlation between joints within the central section than across sections and past instances of temperature anomalies spreading to adjacent joints in the central section), quickly mines the spatial correlation between this expansion joint and five surrounding expansion joints using an improved graph attention network. Then, through a temporal sliding window mechanism, it integrates the global spatial features of expansion joints at various time points within the past hour, ultimately generating multi-dimensional expansion joint status features that clearly present the diffusion trend of temperature anomalies.

[0040] The process involves analyzing the expansion joint status characteristics, identifying expansion joint anomalies, and assessing the risk level of the expansion joints to generate an expansion joint status report. Specifically, this includes: Based on the preset expansion joint feature-state mapping rule library, the multi-dimensional expansion joint state features are analyzed in layers to generate an expansion joint feature-state relationship matrix, and the expansion joint state analysis results are output based on the expansion joint feature-state relationship matrix. The isolated forest algorithm is used to perform abnormal pattern matching and identification on the multi-dimensional expansion joint state features by combining the feature-state relationship matrix of the expansion joint and the preset threshold range of the normal state features of the expansion joint, marking the abnormal feature type and expansion joint number, and outputting the abnormal identification result of the expansion joint. A multi-factor risk assessment model is constructed. The abnormal identification results of the expansion joint are input, and the severity of the abnormal characteristics of the expansion joint, the service life of the expansion joint, the operating load of the line where the expansion joint is located, and the historical fault propagation coefficient are integrated. The weight of the assessment indicators is determined by the analytic hierarchy process, the comprehensive risk assessment value of the expansion joint is calculated, and the risk level of the expansion joint is determined according to the preset risk level classification standard.

[0041] The expansion joint feature-state mapping rule base is built based on industry safety standards for contact rail expansion joints, long-term historical operating data, and the experience of operation and maintenance experts. It predefines the correspondence between multi-dimensional expansion joint state features and various operating states. For example, it matches temperature-related features with "normal thermal expansion" and "excessive thermal expansion" states, and gap distance features with "compliant docking" and "gap exceeding standards" states. The expansion joint feature-state relationship matrix intuitively reflects the correlation strength between each feature and different operating states through matrix elements, avoiding subjective bias in feature analysis and making the expansion joint state analysis results more objective and traceable.

[0042] The Isolation Forest algorithm focuses on the efficiency and accuracy of anomaly identification. It does not rely on a large number of normal samples for training. By constructing an isolated tree, it quickly locates multi-dimensional features that deviate from the normal distribution. Combined with the feature-state relationship matrix of the expansion joint, it clarifies the state problem corresponding to the feature anomaly. Then, it compares with the preset threshold range of normal state features of the expansion joint. Based on the feature statistical range set under normal working conditions, it achieves accurate matching of anomaly patterns and marks the anomaly feature type and the corresponding expansion joint number.

[0043] The core of the multi-factor risk assessment model lies in comprehensively considering the key variables affecting the safety of expansion joints. The severity of abnormal characteristics of the expansion joint directly reflects the urgency of the current potential fault; the service life of the expansion joint is related to the basic risks brought about by equipment aging; the operating load of the line where the expansion joint is located determines the potential impact range of the fault; and the historical fault propagation coefficient reflects the potential probability of fault spread. The integration of these four factors comprehensively covers the dimensions of risk assessment. The analytic hierarchy process (AHP) scientifically determines the weight of different indicators by comparing the importance of each assessment indicator pairwise. For example, the weight of the severity of abnormal characteristics of the expansion joint is higher than the weight of the service life of the expansion joint, avoiding a single factor dominating the assessment results. This ensures that the calculated comprehensive risk assessment value of the expansion joint closely matches the actual risk level. Then, the level is determined according to the preset risk level classification standard, making the comprehensive risk assessment results of the expansion joint more instructive.

[0044] In practical applications, taking the detection of expansion joints in the contact rail of Metro Line 3 in a certain city as an example, the line has constructed a feature-state mapping rule library for expansion joints adapted to its operation based on historical operational data of expansion joints since its inception, industry safety standards, and maintenance experience. After receiving the multi-dimensional expansion joint state features uploaded by each edge terminal of the line, the cloud-based analysis terminal generates a feature-state relationship matrix through hierarchical parsing of the rule library, and outputs the state analysis result "the gap feature of a certain expansion joint in the western section of the city slightly exceeds the normal correlation range".

[0045] Subsequently, the isolated forest algorithm was used, combined with the aforementioned feature-state relationship matrix and the preset normal state feature threshold range, to further match and identify the "gap anomaly" of the expansion joint, and accurately mark the anomaly feature type and expansion joint number. Then, the anomaly identification result of the expansion joint was input into a multi-factor risk assessment model, which integrates the anomaly severity of "minor gap excess", the equipment age of "3 years of operation", the line operating load of "high load during peak hours", and the historical fault propagation coefficient of "gap anomaly propagation has occurred in this section". After determining the weight of each indicator through the analytic hierarchy process, the comprehensive risk assessment value of the expansion joint was calculated, and it was finally determined to be of medium risk. All relevant results were included in the expansion joint status report, providing a clear basis for maintenance personnel to arrange targeted nighttime maintenance.

[0046] The system, based on a preset expansion joint feature-state mapping rule base, performs hierarchical analysis on the multi-dimensional expansion joint state features to generate an expansion joint feature-state relationship matrix, specifically including: Among them, the relationship value between the v-th multi-dimensional expansion joint state feature and the j-th expansion joint operating state in the expansion joint feature-state relationship matrix M. The calculation formula is as follows: In the formula, The hierarchical weight coefficient represents the state feature of the v-th multi-dimensional expansion joint. Represents the dynamic correction coefficient for the v-th multidimensional expansion joint state characteristic; The sensitivity adjustment coefficient representing the feature-state mapping; This represents the basic correlation between the v-th multidimensional expansion joint state characteristics and the j-th expansion joint operating state.

[0047] The hierarchical weight coefficient of the multi-dimensional expansion joint status features is set according to the importance of the multi-dimensional expansion joint status features in the determination of the expansion joint's operating status. For example, the temperature status features have a more critical impact on the determination of abnormal thermal expansion status, so their hierarchical weight coefficient will be higher, thereby reflecting the difference in the primary and secondary importance of multi-dimensional features in status analysis.

[0048] The dynamic correction coefficient is used to adapt to fluctuations in actual operating conditions. When a multi-dimensional expansion joint's state characteristic is affected by temporary environmental disturbances, this coefficient fine-tunes its contribution to the relational value, avoiding analytical bias caused by fixed parameters. The sensitivity adjustment coefficient of the feature-state mapping is used to control the impact of changes in the basic correlation degree on the relational value, ensuring that the relational value can accurately respond to subtle changes in the relationship between features and states. The basic correlation degree is predetermined based on historical operating data of expansion joints and industry standards, reflecting the inherent correspondence between features and states. Finally, through the synergistic effect of various coefficients, a relational value that accurately quantifies the strength of the correlation between features and states is generated.

[0049] By employing the above methods and leveraging the synergistic effect of various coefficients, the correlation strength between the multi-dimensional expansion joint status characteristics and operational status can be accurately quantified. This not only reflects the primary and secondary differences and dynamic adaptability of features in hierarchical analysis, but also ensures that the relationship values ​​conform to the actual correlation patterns.

[0050] S5. Based on the expansion joint status report, trigger the corresponding multi-level early warning mechanism for the expansion joint and push the expansion joint early warning information and maintenance suggestions to the management terminal.

[0051] The multi-level early warning mechanism for expansion joints is based on the risk level classification of expansion joints in the status report, with corresponding early warning levels of different response intensities. Different levels are matched with differentiated triggering rules and handling priorities. For example, high risk corresponds to an emergency early warning, triggering an immediate response process; medium risk corresponds to a routine early warning, associated with planned handling; and low risk corresponds to a suggestive early warning, requiring only continuous monitoring. This achieves precise matching of risk and early warning measures, avoiding over-warning or under-warning. The early warning information for expansion joints includes the specific location of the expansion joint, the type of abnormal characteristics, and the risk level of the expansion joint, ensuring that management can quickly identify the problem object and its urgency. Maintenance recommendations for expansion joints are formulated based on the abnormal characteristics of the expansion joint, the risk level of the expansion joint, and the rail transit operation scenario, providing feasible handling solutions for different anomaly types and risk levels.

[0052] like Figure 2 The diagram shown is a system block diagram of an online detection system for the status information of a contact rail expansion joint provided in an embodiment of the present invention. The system includes: The data acquisition module is used to acquire multi-dimensional expansion joint status data of the contact rail expansion joint according to a preset acquisition frequency. The multi-dimensional expansion joint status data includes temperature status data, gap distance data, and sliding action data. The data processing module is used to preprocess the multi-dimensional expansion joint status data to generate standard expansion joint status data. The data upload module is used to upload the status data of the standard expansion joint according to the characteristics of different operating scenarios of rail transit, using a differentiated data transmission strategy. The status detection module is used to construct an edge-cloud collaborative analysis model, extract multi-dimensional expansion joint status features from the standard expansion joint status data, and perform expansion joint status analysis, expansion joint anomaly identification, and expansion joint risk level assessment on the multi-dimensional expansion joint status features to generate an expansion joint status report. The early warning and maintenance module is used to trigger the corresponding multi-level early warning mechanism for the expansion joint based on the expansion joint status report and push the expansion joint early warning information and expansion joint maintenance suggestions to the management terminal.

[0053] Figure 2 The apparatus of the illustrated embodiment can be used to perform corresponding actions. Figure 1 The steps in the method embodiments shown are implemented in a similar manner and have similar technical effects, and will not be repeated here.

[0054] An electronic device includes a memory and a processor, wherein the memory stores a computer program, and when the processor runs the computer program stored in the memory, the processor performs the steps of the online detection method for the status information of a contact rail expansion joint as described in any of the above claims.

[0055] like Figure 3 The diagram shown is a hardware structure schematic of an electronic device according to an embodiment of the present invention. The electronic device 30 includes: a processor 31, a memory 32, and a computer program; wherein... The memory 32 is used to store the computer program, and the memory may also be flash memory. The computer program is, for example, an application program or functional module that implements the above method.

[0056] The processor 31 is configured to execute the computer program stored in the memory to implement the various steps performed by the device in the above method. For details, please refer to the relevant descriptions in the preceding method embodiments.

[0057] Alternatively, the memory 32 can be either standalone or integrated with the processor 31.

[0058] When the memory 32 is a device independent of the processor 31, the device may further include: Bus 33 is used to connect the memory 32 and the processor 31.

[0059] A readable storage medium storing a computer program, which, when executed by a processor, is used to implement the steps of an online detection method for the status information of a contact rail expansion joint as described in any of the above claims.

[0060] The readable storage medium can be a computer storage medium or a communication medium. A communication medium includes any medium that facilitates the transfer of computer programs from one location to another. A computer storage medium can be any available medium accessible to a general-purpose or special-purpose computer. For example, a readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application-Specific Integrated Circuit (ASIC). Alternatively, the ASIC can be located in a user equipment. Of course, the processor and the readable storage medium can also exist as discrete components in a communication device. The readable storage medium can be a read-only memory (ROM), random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.

[0061] The present invention also provides a program product including executable instructions stored in a readable storage medium. At least one processor of the device can read the executable instructions from the readable storage medium, and the at least one processor executes the executable instructions to cause the device to implement the methods provided in the various embodiments described above.

[0062] In the embodiments of the above-described device, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly manifested as execution by a hardware processor, or execution by a combination of hardware and software modules within the processor.

[0063] Through the above embodiments, this invention acquires multi-dimensional expansion joint status data of the contact rail expansion joint according to a preset acquisition frequency. This multi-dimensional expansion joint status data includes temperature status data, gap distance data, and sliding motion data. The multi-dimensional expansion joint status data is preprocessed to generate standard expansion joint status data. Based on the characteristics of different rail transit operating scenarios, a differentiated data transmission strategy is adopted to upload the standard expansion joint status data. An edge-cloud collaborative analysis model is constructed to extract multi-dimensional expansion joint status features from the standard expansion joint status data. This model is then used to perform expansion joint status analysis, expansion joint anomaly identification, and expansion joint risk level assessment, generating an expansion joint status report. Based on the expansion joint status report, a corresponding multi-level early warning mechanism for the expansion joint is triggered, and early warning information and maintenance suggestions are pushed to the management end. This achieves comprehensive and real-time online monitoring of the contact rail expansion joint status information, effectively solving the limitations of traditional manual inspection and existing single-parameter detection technologies.

[0064] This invention overcomes the limitations of traditional single-parameter detection by collecting multi-dimensional expansion joint status data, including temperature, gap distance, and sliding motion data. It comprehensively covers key operational indicators of expansion joints, enabling timely detection of potential safety hazards such as temperature anomalies, gap deviations, and sliding jamming, avoiding the drawbacks of manual periodic inspections which struggle to monitor in real-time and are prone to overlooking problems. This invention ensures data accuracy and consistency through outlier removal, missing value completion, and standardized preprocessing of the multi-dimensional expansion joint status data. Based on different scenarios such as peak and off-peak operation, nighttime maintenance shutdowns, and extreme weather, this invention employs differentiated data transmission strategies to dynamically adapt to the real-time requirements of each scenario, balancing data transmission efficiency and stability, and avoiding data delays or redundant transmission. Finally, this invention extracts multi-dimensional expansion joint status features through an edge-cloud collaborative analysis model, accurately performing expansion joint status analysis, anomaly identification, and risk level assessment, replacing subjective manual judgment and improving the accuracy of detection and analysis. This invention triggers a multi-level early warning mechanism for expansion joints based on expansion joint status reports and pushes early warning information and maintenance suggestions for expansion joints. This helps maintenance personnel to handle faults in advance, ensures the safe and stable operation of the contact rail power supply system, reduces unnecessary maintenance work and costs, and meets the needs of efficient and safe operation of modern rail transit.

[0065] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for online detection of the status information of a contact rail expansion joint, characterized in that, The method includes: The multi-dimensional expansion joint status data of the contact rail expansion joint is acquired according to a preset acquisition frequency. The multi-dimensional expansion joint status data includes temperature status data, gap distance data, and sliding action data. The multi-dimensional expansion joint status data is preprocessed to generate standard expansion joint status data; Based on the characteristics of different operating scenarios of rail transit, a differentiated data transmission strategy is adopted to upload the status data of the standard expansion joint; An edge-cloud collaborative analysis model is constructed to extract multi-dimensional expansion joint status features from the standard expansion joint status data. The multi-dimensional expansion joint status features are then analyzed to identify expansion joint anomalies and assess expansion joint risk levels, generating an expansion joint status report. Based on the expansion joint status report, the corresponding multi-level early warning mechanism for the expansion joint is triggered, and early warning information and maintenance suggestions for the expansion joint are pushed to the management terminal.

2. The online detection method for the status information of a contact rail expansion joint according to claim 1, characterized in that, The temperature status data includes contact rail temperature data and ambient temperature data; the gap distance data is the real-time spacing data at the mating point of the contact rail expansion joint; the sliding motion data is the smoothness correlation data of the relative sliding of the contact rail expansion joint. Specifically, when collecting the temperature status data, a contact-type platinum resistance sensor or a non-contact infrared temperature sensor is used; when collecting the gap distance data, a laser rangefinder is used.

3. The online detection method for the status information of a contact rail expansion joint according to claim 1, characterized in that, The step of preprocessing the multi-dimensional expansion joint status data to generate standard expansion joint status data specifically includes: Outlier identification and removal are performed on the multi-dimensional expansion joint status data based on the Grubbs criterion. Among them, the sample mean of the gap distance data and the sample standard deviation of the gap distance data Calculate the i-th data point in the gap distance data. Grubbs statistic The Grubbs statistic Compared with the preset significance level The corresponding Grubbs critical value Comparison, , n represents the amount of data in the gap distance data; if Then determine the i-th data point Outliers were identified and removed. Missing value detection is performed on the gap distance data after outlier removal, and linear interpolation is used to calculate the gap distance. Data completion for missing positions at any time The corresponding calculation formula is as follows: In the formula, , , This indicates the time nodes for collecting three consecutive data points in the gap distance data. ; Indicating gap distance data in Valid data at any given time; Indicating gap distance data in Valid data at any given time; The calculation methods for the temperature status data and the sliding motion data are the same as above; The multi-dimensional expansion joint status data is normalized using the Z-score standardization method and then integrated according to the collection time nodes to generate the standard expansion joint status data.

4. The online detection method for the status information of a contact rail expansion joint according to claim 1, characterized in that, The method of uploading the standard expansion joint status data using a differentiated data transmission strategy based on the characteristics of different rail transit operating scenarios specifically includes: The typical operating scenarios of rail transit are divided according to the characteristics of different operating scenarios of rail transit. The typical operating scenarios of rail transit include normal operating peak hours, normal operating off-peak hours, nighttime shutdown and maintenance periods, and extreme weather operation periods. Construct a scenario-transmission parameter mapping relationship and dynamically adjust the transmission frequency under different typical rail transit operation scenarios. The corresponding calculation formula is as follows: In the formula, Indicates the preset reference transmission frequency; This represents the real-time demand coefficient for different typical rail transit operating scenarios, where r is the time period during extreme weather conditions. ;r represents peak operating hours. ;r represents the off-peak period during normal operation. When r represents the nighttime shutdown for maintenance, ; The urgency coefficient represents the condition data of the standard expansion joint; Indicates the weight of the real-time requirement coefficient for the scenario; This indicates the weight of the data urgency coefficient. .

5. The online detection method for the status information of a contact rail expansion joint according to claim 1, characterized in that, The construction of the edge-cloud collaborative analysis model extracts multi-dimensional expansion joint status features from the standard expansion joint status data, specifically including: The edge-cloud collaborative analysis model includes an edge processing end and a cloud analysis end, and adopts a layered collaborative architecture of real-time edge preprocessing and global cloud fusion. The edge processing terminal receives the standard expansion joint status data, divides the data into segments according to a preset time window, extracts the spatial coupling features of the standard expansion joint status data through an improved CNN model, and extracts the temporal evolution features of the standard expansion joint status data through a Transformer model. A two-dimensional fusion strategy based on dynamic weight adaptation and feature attention optimization is used to fuse the spatial coupling features and the temporal evolution features to generate a local feature vector of the expansion joint. The corresponding calculation formula is as follows: In the formula, This represents the spatial coupling features extracted by the improved CNN model; This represents the temporal evolution features extracted by the Transformer model; Dynamic weights representing spatial coupling characteristics; Dynamic weights representing temporal evolution characteristics ; Feature attention weights representing spatial coupling characteristics; Feature attention weights represent temporal evolution characteristics.

6. The online detection method for the status information of a contact rail expansion joint according to claim 5, characterized in that, The cloud-based analysis terminal analyzes the local feature vector of the expansion joint. Global feature fusion is performed to generate the multi-dimensional expansion joint state features. Specifically, it includes: The cloud-based analysis terminal receives the local feature vector of the expansion joint uploaded by the edge processing terminal. An expansion joint feature vector index library was established based on the contact rail track partitioning and expansion joint numbering. Based on the preset contact rail area coupling coefficient and historical fault propagation weight parameters, an improved graph attention network is used to analyze the local feature vectors of expansion joints in the expansion joint feature vector index library. Spatial correlation mining is performed to extract global spatial features of the expansion joint. ; The spatial global features of the expansion joint at different time points are analyzed using a time-series sliding window mechanism. Perform temporal correlation fusion to generate the multi-dimensional expansion joint state features. .

7. The online detection method for the status information of a contact rail expansion joint according to claim 1, characterized in that, The process involves analyzing the expansion joint status characteristics, identifying expansion joint anomalies, and assessing the risk level of the expansion joints to generate an expansion joint status report. Specifically, this includes: Based on the preset expansion joint feature-state mapping rule library, the multi-dimensional expansion joint state features are analyzed in layers to generate an expansion joint feature-state relationship matrix, and the expansion joint state analysis results are output based on the expansion joint feature-state relationship matrix. The isolated forest algorithm is used to perform abnormal pattern matching and identification on the multi-dimensional expansion joint state features by combining the feature-state relationship matrix of the expansion joint and the preset threshold range of the normal state features of the expansion joint, marking the abnormal feature type and expansion joint number, and outputting the abnormal identification result of the expansion joint. A multi-factor risk assessment model is constructed. The abnormal identification results of the expansion joint are input, and the severity of the abnormal characteristics of the expansion joint, the service life of the expansion joint, the operating load of the line where the expansion joint is located, and the historical fault propagation coefficient are integrated. The weight of the assessment indicators is determined by the analytic hierarchy process, the comprehensive risk assessment value of the expansion joint is calculated, and the risk level of the expansion joint is determined according to the preset risk level classification standard.

8. The online detection method for the status information of a contact rail expansion joint according to claim 7, characterized in that, The system, based on a preset expansion joint feature-state mapping rule base, performs hierarchical analysis on the multi-dimensional expansion joint state features to generate an expansion joint feature-state relationship matrix. Specifically... include: Among them, the relationship value between the v-th multi-dimensional expansion joint state feature and the j-th expansion joint operating state in the expansion joint feature-state relationship matrix M. The calculation formula is as follows: In the formula, The hierarchical weight coefficient represents the state feature of the v-th multi-dimensional expansion joint. Represents the dynamic correction coefficient for the v-th multidimensional expansion joint state characteristic; The sensitivity adjustment coefficient representing the feature-state mapping; This represents the basic correlation between the v-th multidimensional expansion joint state characteristics and the j-th expansion joint operating state.

9. An online detection system for the status information of a contact rail expansion joint, applied to the online detection method for the status information of a contact rail expansion joint as described in any one of claims 1-8, characterized in that, The system includes: The data acquisition module is used to acquire multi-dimensional expansion joint status data of the contact rail expansion joint according to a preset acquisition frequency. The multi-dimensional expansion joint status data includes temperature status data, gap distance data, and sliding action data. The data processing module is used to preprocess the multi-dimensional expansion joint status data to generate standard expansion joint status data. The data upload module is used to upload the status data of the standard expansion joint according to the characteristics of different operating scenarios of rail transit, using a differentiated data transmission strategy. The status detection module is used to construct an edge-cloud collaborative analysis model, extract multi-dimensional expansion joint status features from the standard expansion joint status data, and perform expansion joint status analysis, expansion joint anomaly identification, and expansion joint risk level assessment on the multi-dimensional expansion joint status features to generate an expansion joint status report. The early warning and maintenance module is used to trigger the corresponding multi-level early warning mechanism for the expansion joint based on the expansion joint status report and push the expansion joint early warning information and expansion joint maintenance suggestions to the management terminal.

Images