Track expansion disease identification method based on subway track

Abstract

The invention provides a rail expansion disease identification method based on a subway rail, relates to the technical field of rail monitoring, and can continuously monitor the strain and temperature change of the rail by arranging fiber grating sensors in a sensitive area of the rail. Therefore, the track system can obtain temperature and strain data in real time, so that the track can be dynamically monitored. Compared with traditional periodic inspection, dynamic changes of the track in extreme environments such as high temperature and low temperature can be effectively captured through real-time monitoring, and the possible track expansion problem can be found in time. Risk backtracking judgment is carried out by analyzing the relation between strain sudden increase and temperature change of an orbit and introducing a nighttime'cold contraction-daily expansion 'fatigue cycle effect. The rail fatigue factor is constructed by backtracking the rail temperature and the strain fluctuation curve of nearly three days, and different monitoring results are corrected, so that misjudgment caused by monitoring time loss or data deviation is reduced, and the monitoring accuracy is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of track monitoring, and in particular to a method for identifying expansion defects of subway tracks. Background Art

[0002] As a vital component of urban transportation, the operational safety of subway track systems directly impacts both operational efficiency and passenger safety. Traditional subway track defect monitoring technologies primarily focus on detecting track cracks, turnout failures, and track defects. However, over time, the types of track defects have become increasingly diverse, with track expansion, in particular, becoming a significant issue impacting the safe operation of subway track systems.

[0003] Rail expansion defects are usually caused by the expansion or contraction of the track due to temperature changes, resulting in deformation, bending, and twisting of the track geometry. This defect usually occurs in hot summer weather. The rails expand under high temperature conditions. If not effectively managed, the rails may become misaligned, bent, or even damaged, seriously affecting the safe operation of subway trains. Rail expansion is prone to occur in areas where the pressure or temperature of the rails is relatively high, such as the junction of the expansion zone and the fixed zone, the simple fixed zone, in front of ballastless bridges and level crossings, which are locations prone to pressure peaks, as well as vertical curves, braking sections at the beginning and end of curves, and the bottom of slopes. In spring and summer, when the temperature gradually rises, when the daytime rail temperature is close to the locked rail temperature, the large temperature difference between day and night will also increase the temperature pressure inside the rails, affecting the stability of the line.

[0004] Traditional monitoring methods rely primarily on periodic manual inspections and regular track inspection vehicles, primarily focusing on static defects such as cracks and fissures. However, this traditional approach suffers from low detection efficiency, high labor costs, and difficulty monitoring the impact of temperature changes on the track in real time, making it difficult to promptly detect dynamic track defects such as track expansion. Traditional track defect monitoring systems often overlook the relationship between temperature and track expansion and fail to effectively predict the accumulation of track fatigue under temperature fluctuations. Therefore, traditional monitoring methods cannot meet the needs of real-time, accurate, and comprehensive monitoring of track defects. Especially in extreme temperature environments, the problem of track expansion and deformation becomes more prominent, resulting in the inability of traditional monitoring technologies to promptly identify track expansion defects and make appropriate repairs and adjustments.

[0005] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not form the prior art that is already known to a person of ordinary skill in the art. Summary of the Invention

[0006] The purpose of the present invention is to provide a method for identifying expansion defects of subway tracks to solve the problems raised in the above background technology.

[0007] To achieve the above object, the present invention provides the following technical solutions: The specific steps of the subway track expansion defect identification method include: Step 1: Divide the subway track into several sensitive areas and deploy fiber Bragg grating sensors in the sensitive areas to continuously monitor the strain and temperature of the i-th sensitive area, analyze the correlation between the sudden increase in strain and the temperature change, and form the first evaluation value of the corresponding point of the i-th sensitive area. ; Step 2: Integrate the nighttime "cold shrinkage-sun expansion" fatigue cycle effect to conduct risk retrospective judgment, trace the track temperature-strain fluctuation curve in the past three days, and construct the rail fatigue factor of the i-th sensitive area , and the first evaluation value of the corresponding point in the i-th sensitive area Correction is performed to correct the deviation between the daytime detection results and the missing data at night, and the first evaluation value of the corrected i-th sensitive area is obtained. After obtaining the first classification results, a key monitoring section group is established; Step 3: For the key monitoring section group, periodically collect the track 3D point cloud data and detect the track geometric distortion to form the track state index of the i-th sensitive area. and graded to obtain a second grade result; Step 4: A drone equipped with a high-resolution camera cruises and photographs the track, collects track surface images and performs intelligent recognition to construct the track damage risk coefficient of the i-th sensitive area at time t. and graded to obtain a third grade result; Step 5: Generate corresponding maintenance strategies for the first classification result, the second classification result, and the third classification result.

[0008] Furthermore, the step 1 includes: Step 101: According to the track design drawings, identify the joint sections, welding sections, curved sections and bridge intersection sections in the track, set them as target areas prone to track expansion, and divide them into several sensitive areas at equal intervals; each sensitive area is marked as the i-th monitoring section, and the length of the track section corresponding to the node is calculated as , the distance between sensor nodes is marked as , and record the track structure attribute labels at the node position, including rail type, fastener type and rail grade; Step 102: At the fishplate connection in each i-th sensitive area, a plurality of fiber Bragg grating sensor nodes are evenly arranged along the track direction; Multiple Fiber Bragg Grating sensor nodes include: The temperature sensor is used to collect the temperature data of the rail web and obtain the rail temperature value of the i-th sensitive area at time t. ; The strain sensor is used to collect the longitudinal strain changes of the rail and obtain the longitudinal strain value of the rail in the i-th sensitive area at time t. ; Vibration sensor, used to identify disturbances caused by trains, determine track structure looseness or vibration change trends, and obtain the rail vibration value of the i-th sensitive area at time t ; Extract the temperature curve and strain curve in the sensitive area or the second area, and construct the first evaluation value of the i-th sensitive area , the specific construction steps are: S111. Extract the longitudinal strain value of the rail in the i-th sensitive area at time t , the strain increase rate of the i-th sensitive area is calculated by the following formula : Where, represents the maximum rate of change of strain over time in the i-th region during the monitoring period, is the strain change rate, calculated by the inverse method, Represents the inverse, takes the derivative of t, represents the rate of change over time; and sets the deformation threshold. When the deformation threshold is exceeded, the time when the maximum value occurs is recorded ; S112, according to the time when the maximum value in S111 occurs , it is necessary to investigate the cause and extract the time when the maximum value occurs Calculate the critical fluctuation value of rail temperature in the i-th sensitive area according to the rail temperature at that time , the expression is as follows: Where, express When the deformation threshold is exceeded, the time when the maximum value occurs is recorded Rail temperature at ， Indicates the temperature value of the rail in the initial stable state of operation; S113. Calculate the hysteresis correlation factor of the i-th sensitive area : Where, Indicates the delay time of the effect of temperature change on strain. represents the Pearson correlation coefficient, represents the strain value at the delay time; S114: Combine the strain increase rate of the i-th sensitive area obtained in S111-S113 , critical fluctuation value of rail temperature and the hysteresis correlation factor of the i-th sensitive area After dimensionless processing, the first evaluation value of the i-th sensitive area is calculated by the following formula: : Where, 、 and are constants, and , Indicates that 、 and Weight, calculate the first evaluation value of the i-th sensitive area .

[0009] Furthermore, the step 2 includes: S201. Collect real-time monitoring data on track temperature and strain, covering at least the past three days. Collect daytime temperature rise data and nighttime temperature data. Use curve fitting methods to analyze the changes in the "cold shrinkage-sun expansion" stress cycle and construct the periodic change intensity of the track in the i-th sensitive area at time t. : Where, is the base stress, is the stress cycle amplitude, which indicates the intensity of the “cold contraction-sun expansion” effect; is the angular frequency at time t, representing the daily cycle change. When the temperature change is a 24-hour cycle, the angular frequency 2 ; Indicates phase shift; S202: The periodic variation intensity of the orbit of the i-th sensitive area at time t calculated according to S201 , combined with the maximum stress amplitude that the rail material in the i-th sensitive area continuously withstands , according to the material characteristics of the rail and experimental data, the rail fatigue factor of the i-th sensitive area is calculated by the following formula : Among them, the rail fatigue factor of i sensitive area is The larger the value, the greater the stress cycle amplitude caused by the "cold shrinkage-sun expansion" effect, and the higher the accumulated fatigue damage. S203, based on the rail fatigue factor of the i-th sensitive area , the first evaluation value of the i-th sensitive area After correction, the first evaluation value of the i-th sensitive area after correction is obtained by the following formula : Where, It represents the correction coefficient, which indicates the influence of fatigue factor on the evaluation value, and is determined based on actual monitoring data and experimental results.

[0010] Furthermore, the step 2 further includes: S204: Calculate the first evaluation value of the corrected i-th sensitive area Grading is performed to obtain the first assessment level, including: When 0.00< When ≤0.60, it means there is no risk of track deformation caused by thermal expansion and continuous monitoring is required; When 0.61< When ≤0.80, it indicates that there is a risk of track deformation caused by thermal expansion, generating Level I and triggering the first abnormality warning; 0.81< When ≤1.00, it indicates that there is a risk of track deformation caused by thermal expansion, which is higher than Level I, generating Level II and triggering the second abnormality warning; S205. The sensitive areas that trigger the first abnormal warning and the second abnormal warning are filed as a key monitoring section group.

[0011] Furthermore, the step 3 includes: S301. Use mobile laser scanning equipment, including vehicle-mounted LiDAR, to conduct track inspections within key monitoring sections at predetermined intervals, obtaining reference point cloud data of the track in its initial state and real-time point cloud data during the current inspection. S302: Segment the reference point cloud data into several sensitive areas according to step 101, and perform coordinate registration between the reference point cloud data and the real-time point cloud data during the current detection; S303: Extract two track top edge lines in each sensitive area, fit the center line, gauge line and rail top elevation curve for each track; output the set of center line points of the sensitive area, and output the lateral offset value of the i-th sensitive area. , longitudinal offset value , cumulative settlement value , Gauge change value , distortion rate change rate , geometric profile deviation and roughness , and the method of obtaining it is as follows: Compare the geometric centerline point set of the sensitive area with the reference point cloud data to obtain: lateral offset value , longitudinal offset value and cumulative settlement value ; Extract the left and right rail head edge points at the positioning point of the i-th sensitive area, calculate the horizontal distance between the two points, obtain the track gauge, and compare the track gauge with the design track gauge in the benchmark point cloud data to obtain the track gauge change value. ; Extract the left and right rail top elevations at the positioning point in the i-th sensitive area, calculate the left and right height difference, and perform differential slope fitting to obtain the distortion rate change rate. : At the mileage point of the i-th sensitive area, perpendicular to the track direction, a cross section is cut and the contour lines are fitted, including the rail head, rail waist and rail bottom, and the geometric contour deviation is obtained by point-by-point deviation calculation with the reference point cloud data. ; At the mileage point in the i-th sensitive area, the rail top elevation Z value is extracted along the track centerline from the real-time point cloud data during the current detection to form a continuous elevation sequence. The continuous elevation sequence is Fourier transformed to analyze the frequency range corresponding to the main peak in the frequency and output the roughness. .

[0012] Furthermore, the step 3 further includes: S304: Extract the lateral offset value corresponding to the i-th sensitive area , longitudinal offset value , cumulative settlement value , Gauge change value , distortion rate change rate , geometric profile deviation and roughness , after dimensionless processing, the orbital state index of the i-th sensitive area is calculated : Where, 、 、 、 、 、 and All are weights, and the sum of the weights is 1; S305: Track status index for the i-th sensitive area Grading to obtain a second assessment level includes: When 0.00< When ≤0.40, it means that the track deviation is within the preset expected range and is continuously monitored; When 0.41< When the value is ≤0.70, it indicates that there is a moderate risk of track deviation due to wear or aging, generating Level III and triggering the third abnormality warning; 0.71< When ≤1.00, it indicates that there is a risk of height deviation caused by wear or settlement of the track, and the risk is higher than Level III, generating Level IV and triggering the fourth abnormal warning.

[0013] Furthermore, the step 4 includes: S401. Use a drone with a high-resolution camera to regularly cruise and photograph the track, collect all-round image data of the track surface, and perform preliminary preprocessing on the collected image data, including noise removal, distortion correction, and color correction. S402: Using image recognition technology, extract the rail gap position of each sensitive area in the image data and mark it, and extract the actual width of the rail gap at the jth position in the i-th sensitive area. , when the actual width of the j-th rail gap in the i-th sensitive area ≤ the closing judgment threshold, the rail gap is judged to be in a closed state; When the actual width of the j-th rail gap in the i-th sensitive area > closing judgment threshold, the rail gap is judged to be in an unclosed state; S403: Count the number of closed rail joints in the i-th sensitive area and mark them accordingly. Calculate the total number of closed rail joints in the i-th sensitive area and then calculate the ratio of the total number of rail joints in the i-th sensitive area to obtain the rail joint closure rate of the i-th sensitive area. ; The closing determination threshold is set to 2mm.

[0014] Furthermore, the step 4 further includes: S404: Extract the rail vibration value of the i-th sensitive area at time t in step 102 Before the train arrives, a camera or laser scanner is used to capture images of each sensitive area, and an image recognition algorithm is used to locate the rail gap position of each sensitive area, and the initial width of the j-th rail gap in the i-th sensitive area is determined. , the vibration sensor monitors the rail vibration value of the i-th sensitive area at time t in real time , preset vibration threshold, when the rail vibration value of the i-th sensitive area at time t is When the vibration threshold is reached, it is determined that a train is approaching and the camera starts recording. When the train wheel passes through the rail gap, the disturbance width of the jth rail gap in the i-th sensitive area after the train passes at time t is extracted. Calculate the track gap shrinkage rate of the jth track gap in the i-th sensitive area at time t : Where, Indicates the time interval when the train wheels pass through the rail gap; S405: Based on the track gap shrinkage rate of the jth track gap in the i-th sensitive area at time t , calculate the average track gap shrinkage rate of the i-th sensitive area at time t : S406: Combine the track gap closure rate of the i-th sensitive area , the average track gap shrinkage rate of the i-th sensitive area at time t and the rail vibration value of the i-th sensitive area at time t , after dimensionless processing, the track damage risk coefficient of the i-th sensitive area at time t is obtained : Where, 、 and are constants, and , Indicates that 、 and Weight, calculate the track damage risk coefficient of the i-th sensitive area at time t .

[0015] Furthermore, the step 4 further includes: S407. Track damage risk coefficient for the i-th sensitive area at time t Grading to obtain a third assessment level includes: When 0.00< When ≤0.60, it means that the track gap change in the sensitive area is within the expected normal range, the track is in good condition, and continuous monitoring is required; when When it is greater than 0.61, it indicates that there is an overload risk in the sensitive area, and there is a risk of fatigue of the rails and track structure due to long-term rail gap closure, generating Level V and triggering the fifth abnormal warning.

[0016] Furthermore, the step 5 includes: Generate corresponding strategies for Levels I to V, including: Level I generates the first maintenance strategy, including: in this sensitive area, when the temperature exceeds 35°C in summer, the track in this area will be sprayed with water every 1.5 to 2 hours to cool it down; when the temperature drops below -30°C in winter, the track heating system will be used to maintain the track surface temperature between -10°C and +5°C; Level II generates the second maintenance strategy, including: in this sensitive area, when the temperature exceeds 35°C in summer, the track in this area will be sprayed with water every 30 minutes to 1 hour to cool it down; when the temperature drops below -30°C in winter, the track heating system will be used to maintain the track surface temperature between -5°C and +10°C; Level III generates the third maintenance strategy, which includes: local repair of the sensitive area, including replacing severely worn rails or reinforcing track joints, increasing the grouting coverage by 30% to reinforce the sensitive area, and conducting inspections every two months; Level IV generates the fourth maintenance strategy, which includes: taking foundation reinforcement measures for the soft soil layer or waterlogged areas in the sensitive area to prevent further settlement, using track correction equipment to adjust the offset parts to restore the normal condition of the track, and conducting monthly inspections; Level V generates the fifth maintenance strategy, which includes: using a rail gap opener or rail gap expander in the sensitive area to expand 80% of the closed rail gaps in the sensitive area, gradually increasing the gap between the rail gaps to restore the track structure. When severely closed rail gaps cannot be restored to normal through mechanical adjustment, the rail joints and rail gap plates need to be replaced or reinforced; and overloaded vehicles are monitored, and measures such as speed limit or restriction of the passage of overloaded vehicles are taken.

[0017] Compared with the prior art, the present invention has the following beneficial effects: This method dynamically captures track expansion or contraction by monitoring track temperature changes in real time. By calculating the hysteresis correlation factor for the i-th sensitive region, it accurately analyzes the delay in the impact of temperature changes on rail strain, thereby determining the temporal relationship between temperature and rail deformation. This analysis helps identify track defects caused by thermal expansion or other environmental factors, effectively distinguishing deformation caused by heat sources from structural instability caused by non-heat sources, and accurately identifying and assessing the source and type of defects.

[0018] The present invention also introduces the nighttime "cold shrinkage-daily expansion" fatigue cycle effect to conduct risk retrospective judgment. This method can accurately assess the fatigue accumulation of the track under temperature changes and predict the possible expansion defects of the track. It can not only monitor temperature changes, but also timely obtain changes in the geometric shape of the track, such as dynamic defects such as bending and twisting, avoiding the low efficiency of traditional monitoring methods that rely on manual inspections and periodic testing. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 Schematic diagram of the overall method steps of the present invention. DETAILED DESCRIPTION

[0020] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to specific embodiments.

[0021] It should be noted that, unless otherwise defined, the technical or scientific terms used in the present invention should have the usual meanings understood by people with ordinary skills in the field to which the present invention belongs. The "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative position relationships. When the absolute position of the object being described changes, the relative position relationship may also change accordingly.

[0022] Example 1: See also Figure 1 The present invention provides a technical solution: a method for identifying expansion defects of subway tracks, the specific steps of which include: Step 1: Divide the subway track into several sensitive areas and deploy fiber Bragg grating sensors in the sensitive areas to continuously monitor the strain and temperature of the i-th sensitive area, analyze the correlation between the sudden increase in strain and the temperature change, and form the first evaluation value of the corresponding point of the i-th sensitive area. ; Step 2: Integrate the nighttime "cold shrinkage-sun expansion" fatigue cycle effect to conduct risk retrospective judgment, trace the track temperature-strain fluctuation curve in the past three days, and construct the rail fatigue factor of the i-th sensitive area , and the first evaluation value of the corresponding point in the i-th sensitive area Correction is performed to correct the deviation between the daytime detection results and the missing data at night, and the first evaluation value of the corrected i-th sensitive area is obtained. After obtaining the first classification results, a key monitoring section group is established; Step 3: For the key monitoring section group, periodically collect the track 3D point cloud data and detect the track geometric distortion to form the track state index of the i-th sensitive area. and graded to obtain a second grade result; Step 4: A drone equipped with a high-resolution camera cruises and photographs the track, collects track surface images and performs intelligent recognition to construct the track damage risk coefficient of the i-th sensitive area at time t. and graded to obtain a third grade result; Step 5: Generate corresponding maintenance strategies for the first classification result, the second classification result, and the third classification result.

[0023] In this embodiment, this method continuously monitors track strain and temperature changes by deploying fiber Bragg grating (FBG) sensors in sensitive areas of the track. This enables the track system to obtain real-time temperature and strain data, enabling dynamic monitoring of the track. Compared to traditional periodic inspections, real-time monitoring can effectively capture dynamic changes in the track under extreme conditions such as high and low temperatures, enabling timely identification of potential track expansion issues. By analyzing the relationship between sudden increases in track strain and temperature changes, and incorporating the nighttime "cold contraction-daily expansion" fatigue cycle effect for risk retrospective assessment, this method can accurately assess track fatigue accumulation under temperature fluctuations. This process helps improve the accuracy of track condition assessment, reduce misjudgments caused by missing monitoring time or data bias, and enhance monitoring precision. By retrospectively tracing the track temperature and strain fluctuation curves over the past three days to construct a rail fatigue factor, and then correcting and integrating different monitoring results, this method can comprehensively consider multiple data sources to form a comprehensive and accurate track condition assessment. This data fusion not only reduces the risks associated with missing or biased single data sources but also provides a solid basis for subsequent decision-making, helping relevant managers formulate effective track maintenance and repair strategies. This method periodically collects 3D track point cloud data to detect track geometric distortion. It effectively identifies changes in track surface curvature, settlement, and other factors, and can proactively detect geometric distortion caused by track expansion. This detection process helps prevent unstable train travel or track damage caused by track geometric distortion, further ensuring train safety.

[0024] Using drones equipped with high-resolution cameras to cruise and photograph the track surface and perform intelligent image recognition, they can identify damage, cracks, and other track defects in real time and assess potential safety hazards by calculating the risk of track damage. This intelligent technology not only improves the accuracy of track inspections but also reduces the workload of manual inspections, saving significant human resources.

[0025] By grading the monitoring data for each sensitive area, the system can promptly assess the track's risk level and trigger corresponding warnings based on the different grading results. This grading and warning mechanism effectively distinguishes different levels of risk, providing managers with accurate decision-making support, helping them take appropriate maintenance measures based on the actual track conditions and reduce the risk of accidents. Based on real-time data and intelligent monitoring, managers can take timely maintenance measures when potential track problems arise, avoiding the delayed response and untimely repairs that occur with traditional methods. By promptly detecting and addressing track expansion problems, the incidence of accidents can be greatly reduced, thereby improving maintenance efficiency and reducing repair costs. More importantly, timely maintenance and repairs can extend the service life of the track and reduce long-term maintenance costs.

[0026] Example 2 This embodiment is explained in Example 1. Specifically, the step 1 includes: Step 101: According to the track design drawings, identify the joint sections, welding sections, curved sections and bridge intersection sections in the track, set them as target areas prone to track expansion, and divide them into several sensitive areas at equal intervals; each sensitive area is marked as the i-th monitoring section, and the length of the track section corresponding to the node is calculated as , the distance between sensor nodes is marked as , and record the track structure attribute labels at the node position, including rail type, fastener type and rail grade; Step 102: At the fishplate connection in each i-th sensitive area, a plurality of fiber Bragg grating sensor nodes are evenly arranged along the track direction; Multiple Fiber Bragg Grating sensor nodes include: The temperature sensor is used to collect the temperature data of the rail web and obtain the rail temperature value of the i-th sensitive area at time t. ; The strain sensor is used to collect the longitudinal strain changes of the rail and obtain the longitudinal strain value of the rail in the i-th sensitive area at time t. ; Vibration sensor, used to identify disturbances caused by trains, determine track structure looseness or vibration change trends, and obtain the rail vibration value of the i-th sensitive area at time t ; Extract the temperature curve and strain curve in the sensitive area or the second area, and construct the first evaluation value of the i-th sensitive area , the specific construction steps are: S111. Extract the longitudinal strain value of the rail in the i-th sensitive area at time t , the strain increase rate of the i-th sensitive area is calculated by the following formula : Where, represents the maximum rate of change of strain over time in the i-th region during the monitoring period, is the strain change rate, calculated by the inverse method, Represents the inverse, takes the derivative of t, represents the rate of change over time; and sets the deformation threshold. When the deformation threshold is exceeded, the time when the maximum value occurs is recorded ; S112, according to the time when the maximum value in S111 occurs , it is necessary to investigate the cause and extract the time when the maximum value occurs Calculate the critical fluctuation value of rail temperature in the i-th sensitive area according to the rail temperature at that time , the expression is as follows: Where, express When the deformation threshold is exceeded, the time when the maximum value occurs is recorded Rail temperature at ， Indicates the temperature value of the rail in the initial stable state of operation; The difference between the two is a critical temperature fluctuation. If the difference is very large, it means that the mutation is likely to be a structural effect caused by thermal expansion or environmental temperature rise. If the temperature difference is not significant, there may be other problems, such as track derailment, loose foundation, or concentrated force.

[0027] This method first calculates the strain increase rate , detect whether there is an abnormal sudden change response of the rail in unit time, so as to identify the potential structural deformation risk. Subsequently, the temperature critical fluctuation index is further introduced , used to determine whether the mutation is driven by temperature increase, and then assist in distinguishing deformation caused by thermal expansion from non-thermal structural instability, to achieve accurate diagnosis and attribution analysis of abnormal structural response; S113. Calculate the hysteresis correlation factor of the i-th sensitive area : Where, Indicates the delay time of the effect of temperature change on strain. represents the Pearson correlation coefficient, represents the strain value at the delay time; The calculation formula of Pearson correlation coefficient is: Where n represents the total number of sampling points, represents the average rail temperature, It represents the average value of the hysteresis strain, the numerator is the covariance, and the denominator is the product of the standard deviations; hysteresis: emphasizes that the relationship between "temperature change" and "strain response" is not immediate, but there is a delayed effect; high correlation, such as >0.8 indicates that the temperature change is likely to occur After a period of time, it leads to a strain response - this is called "thermal stagflation"; ≈0, which indicates low correlation, indicating that the structural strain is not dominated by temperature changes, and there may be mechanical damage or foundation movement factors.

[0028] S114: Combine the strain increase rate of the i-th sensitive area obtained in S111-S113 , critical fluctuation value of rail temperature and the hysteresis correlation factor of the i-th sensitive area After dimensionless processing, the first evaluation value of the i-th sensitive area is calculated by the following formula: : Where, 、 and are constants, and , Indicates that 、 and Weight, calculate the first evaluation value of the i-th sensitive area .

[0029] The raw data sampling information of the sensor in the third sensitive area is shown in Table 1: Table 1 Step S111: The deformation threshold is set to 1.0με / s. When the deformation threshold is exceeded, the analysis is triggered and the maximum value is recorded at t=1030 seconds. The corresponding rail temperature is 27.1°C. is 24.3°C, then the critical fluctuation value of rail temperature is °C; Assuming the lag time is 10s, use the sliding window Pearson correlation calculation, and take the corresponding values before and after as shown in Table 2: Table 2 , ; Calculate the hysteresis correlation factor of the i-th sensitive area : First calculate the mean: Dimensionless processing, the maximum normalized value is set to: The maximum value of is 2.0; The maximum value is 5.0°C; The maximum value is 1.0; Setting weights , ; ; First evaluation value In this embodiment, fiber Bragg grating sensors are deployed in key sensitive areas of the track to accurately monitor strain, temperature, and vibration data in real time. This solution not only captures the expansion or contraction of the rails due to temperature changes, but also effectively identifies vibration or structural loosening caused by trains. By calculating the strain spike rate and extracting the correlation between temperature changes and strain responses, it is possible to quickly identify abnormal changes in rails during operation and accurately diagnose potential structural deformation problems. Using the critical temperature fluctuation index, it is possible to further clarify whether the source of abnormal mutations is caused by thermal expansion or other factors such as track debonding or loose foundations. This helps to reduce misdiagnosis and ensure the accuracy and operability of the identification results. By calculating the hysteresis correlation factor of the i-th sensitive area, the delay time of the impact of temperature changes on rail strain can be accurately analyzed, thereby determining the temporal relationship between temperature and rail deformation. This analysis helps to identify track defects caused by thermal expansion or other environmental factors, effectively distinguishing deformation caused by heat sources from structural instability caused by non-heat sources, and accurately distinguishing and evaluating the source and type of the defect.

[0030] By comprehensively considering the strain surge rate, critical temperature fluctuation, and the hysteresis correlation factor of the i-th sensitive area, and utilizing dimensionless normalization, a first assessment value can be calculated for each sensitive area. This assessment value provides a quantitative indicator for monitoring track defects, enabling a scientific assessment of track health. This assessment value allows for the classification of track expansion risks in different areas, providing data support for subsequent decision-making and enabling timely implementation of appropriate measures to prevent accidents.

[0031] Example 3 This embodiment is explained in Example 1. Specifically, step 2 includes: S201. Collect real-time monitoring data on track temperature and strain, covering at least the past three days. Collect daytime temperature rise data and nighttime temperature data. Use curve fitting methods to analyze the changes in the "cold shrinkage-sun expansion" stress cycle and construct the periodic change intensity of the track in the i-th sensitive area at time t. : Where, is the base stress, is the stress cycle amplitude, which indicates the intensity of the “cold contraction-sun expansion” effect; is the angular frequency at time t, representing the daily cycle change. When the temperature change is a 24-hour cycle, the angular frequency 2 ; Indicates phase shift; Thermal expansion (daytime): When the temperature rises, the rail expands, generating positive stress (dilation stress); Thermal contraction (nighttime): When the temperature drops, the rail contracts, generating reverse stress (compression stress); Simple harmonic oscillation: The above thermal expansion and contraction effects are periodic in time, that is, the changes in temperature and stress are repeatable. Therefore, a sine wave (i.e., simple harmonic oscillation) is used to approximate the model and obtain the periodic stress change value: S202: The periodic variation intensity of the orbit of the i-th sensitive area at time t calculated according to S201 , combined with the maximum stress amplitude that the rail material in the i-th sensitive area continuously withstands , according to the material characteristics of the rail and experimental data, the rail fatigue factor of the i-th sensitive area is calculated by the following formula : Among them, the rail fatigue factor of i sensitive area is The larger the value, the greater the stress cycle amplitude caused by the "cold shrinkage-sun expansion" effect, and the higher the accumulated fatigue damage. S203, based on the rail fatigue factor of the i-th sensitive area , the first evaluation value of the i-th sensitive area After correction, the first evaluation value of the i-th sensitive area after correction is obtained by the following formula : Where, It represents the correction coefficient, which indicates the influence of fatigue factor on the evaluation value, and is determined based on actual monitoring data and experimental results.

[0032] S204: Calculate the first evaluation value of the corrected i-th sensitive area Grading is performed to obtain the first assessment level, including: When 0.00< When ≤0.60, it means there is no risk of track deformation caused by thermal expansion and continuous monitoring is required; When 0.61< When ≤0.80, it indicates that there is a risk of track deformation caused by thermal expansion, generating Level I and triggering the first abnormality warning; 0.81< When ≤1.00, it indicates that there is a risk of track deformation caused by thermal expansion, which is higher than Level I, generating Level II and triggering the second abnormality warning; S205. The sensitive areas that trigger the first abnormal warning and the second abnormal warning are filed as a key monitoring section group.

[0033] The effectiveness of this embodiment lies in its ability to more accurately assess track health through corrections to cyclic stress and fatigue factors. Combined with analysis of the "cold shrinkage-daily expansion" effect, it can identify stress changes caused by temperature fluctuations, thereby predicting and providing early warning of potential fatigue damage. This method provides dynamic monitoring of rails, effectively preventing track damage caused by excessive fatigue and ensuring safe track operation, with significant preventative and diagnostic benefits. These areas, due to their higher risk of deformation, warrant priority monitoring. Sensitive areas that trigger the first and second abnormality warnings will be filed as key monitoring sections and subject to more rigorous tracking and management. This means that these filed sections will be included in a long-term monitoring plan and may require regular inspection and maintenance to ensure track safety. By grading the corrected assessment values, track areas with different risk levels can be more accurately identified. This provides clear guidance for managers, allowing resources and monitoring to be focused on higher-risk areas. By establishing an abnormality warning mechanism, potential track deformation issues can be identified and addressed in advance. Level I and Level II warnings not only enable timely responses to risks caused by thermal expansion, but also ensure that emergency measures are quickly initiated when problems arise, thereby reducing the impact on railway safety.

[0034] Example 4 This embodiment is explained in Example 1. Specifically, step 3 includes: S301. Use mobile laser scanning equipment, including vehicle-mounted LiDAR, to conduct track inspections within key monitoring sections at predetermined intervals, obtaining reference point cloud data of the track in its initial state and real-time point cloud data during the current inspection. S302: Segment the reference point cloud data into several sensitive areas according to step 101, and perform coordinate registration between the reference point cloud data and the real-time point cloud data during the current detection; S303: Extract two track top edge lines in each sensitive area, fit the center line, gauge line and rail top elevation curve for each track; output the set of center line points of the sensitive area, and output the lateral offset value of the i-th sensitive area. , longitudinal offset value , cumulative settlement value , Gauge change value , distortion rate change rate , geometric profile deviation and roughness , and the method of obtaining it is as follows: Compare the geometric centerline point set of the sensitive area with the reference point cloud data to obtain: lateral offset value , longitudinal offset value and cumulative settlement value ; Extract the left and right rail head edge points at the positioning point of the i-th sensitive area, calculate the horizontal distance between the two points, obtain the track gauge, and compare the track gauge with the design track gauge in the benchmark point cloud data to obtain the track gauge change value. ; Extract the left and right rail top elevations at the positioning point in the i-th sensitive area, calculate the left and right height difference, and perform differential slope fitting to obtain the distortion rate change rate. : At the mileage point of the i-th sensitive area, perpendicular to the track direction, a cross section is cut and the contour lines are fitted, including the rail head, rail waist and rail bottom, and the geometric contour deviation is obtained by point-by-point deviation calculation with the reference point cloud data. ; At the mileage point in the i-th sensitive area, the rail top elevation Z value is extracted along the track centerline from the real-time point cloud data during the current detection to form a continuous elevation sequence. The continuous elevation sequence is Fourier transformed to analyze the frequency range corresponding to the main peak in the frequency and output the roughness. .

[0035] Output the set of centerline points of the sensitive area, as shown in Table 3: Table 3 At each mileage point in the sensitive area, the rail top elevation Z value is extracted along the track centerline from the real-time point cloud data during the current detection to form a continuous elevation sequence. The continuous elevation sequence is Fourier transformed, and the frequency range corresponding to the main peak in the frequency is analyzed to output the roughness Freq: Specifically, the rail top elevation sequence on the track centerline is extracted. The rail top elevation Z value is extracted along the track centerline at 0.1m intervals from the real-time point cloud data at the time of current detection to form: Where n represents the total number of sampling points, is the Z coordinate value of the rail top extracted from the point cloud at the i-th sampling point; Indicates the sampling interval, which is set to 0.1m interval; represents the mileage coordinates of the i-th sampling point along the track centerline; Use polynomial fitting, sliding average or high-pass filtering to remove macro slope changes (low-frequency trends) and only retain micro fluctuations: ; Represents the overall change in low frequency. If Trend(x) is not removed, the Fourier transform will be "contaminated" by the low-frequency long wave; Residual after removal To reflect the true unevenness characteristics; The elevation signal is transformed using the Fast Fourier Transform (FFT) Convert to frequency domain signal F and find the frequency corresponding to the main peak amplitude Identify the frequency corresponding to the main peak amplitude The frequency band in which the roughness Freq is obtained, for example, 0.2–0.4Hz, 0.3–0.6Hz; The frequency sampling information of the smooth state is shown in Table 4: Table 4 The following is the sample data collected in step 3, as shown in Table 5: Table 5 In this embodiment, the on-board LiDAR device can accurately obtain three-dimensional point cloud data of the track, with the characteristics of high precision and high efficiency. Through regular inspections, dynamic updates of the track status can be provided to ensure the continuity of long-term monitoring. By obtaining the baseline point cloud data of the track in its initial state and the real-time point cloud data during the current inspection and comparing them, problems such as track deformation, settlement, and misalignment can be clearly identified. This data comparison provides an important basis for subsequent risk assessment. By extracting the top edge line of the track and fitting the geometric curve of the track, the geometric shape of the track can be accurately described, including the track centerline, gauge line, and rail top elevation curve. Through the analysis of these geometric features, problems such as track deformation, misalignment, and unevenness can be identified. After extracting the rail top elevation data, methods such as polynomial fitting, sliding average, or high-pass filtering are used to remove macro-slope changes (low-frequency trends) to ensure that only the micro-undulations of the track are retained, so that the subtle deformation and unevenness characteristics of the track can be clearly analyzed. Using a Fast Fourier Transform (FFT), the track elevation signal is converted into a frequency domain signal, identifying the frequency band corresponding to the dominant frequency and accurately identifying track irregularities. Frequency analysis helps understand track vibration characteristics and potential wear issues, and is particularly important for predicting high-frequency fluctuations (such as track irregularities).

[0036] Example 5 This embodiment is explained in Example 4. Specifically, step 3 further includes: S304: Extract the lateral offset value corresponding to the i-th sensitive area , longitudinal offset value , cumulative settlement value , Gauge change value , distortion rate change rate , geometric profile deviation and roughness , after dimensionless processing, the orbital state index of the i-th sensitive area is calculated : Where, 、 、 、 、 、 and All are weights, and the sum of the weights is 1; S305: Track status index for the i-th sensitive area Grading to obtain a second assessment level includes: When 0.00< When ≤0.40, it means that the track deviation is within the preset expected range and is continuously monitored; When 0.41< When the value is ≤0.70, it indicates that there is a moderate risk of track deviation due to wear or aging, generating Level III and triggering the third abnormality warning; 0.71< When ≤1.00, it indicates that there is a risk of height deviation caused by wear or settlement of the track, and the risk is higher than Level III, generating Level IV and triggering the fourth abnormal warning.

[0037] In this embodiment, the lateral offset value, longitudinal offset value, cumulative settlement value, gauge change value, twist rate change rate, geometric profile deviation and roughness are dimensionlessly processed to ensure that various parameters can be uniformly calculated and compared under different unit systems. This processing method eliminates the impact of different measurement units on data integration and ensures the comparability and uniformity of the data. By calculating the track condition index and dividing it into different levels, managers can understand the health status of the track at the first time. This makes monitoring more timely and efficient, and can deal with potential problems before they develop into serious failures. The graded assessment of the track condition index provides managers with clear risk boundaries, helping to assess different levels of risk and take targeted measures. For example, in the case of moderate offset risk, short-term repairs and long-term monitoring can be planned, while for serious risks, emergency intervention should be carried out immediately.

[0038] Example 6 The closure of rail gaps can cause localized bending or twisting of the track. Long-term closure of rail gaps can lead to track fatigue, especially on sections with high-frequency train traffic. When the gaps close, the contact surface between the wheel and rail changes, increasing friction or slip between the wheel and rail, leading to increased wear.

[0039] This embodiment is explained in Example 1. Specifically, step 4 includes: S401. Use a drone with a high-resolution camera to regularly cruise and photograph the track, collect all-round image data of the track surface, and perform preliminary preprocessing on the collected image data, including noise removal, distortion correction, and color correction. S402: Using image recognition technology, extract the rail gap position of each sensitive area in the image data and mark it, and extract the actual width of the rail gap at the jth position in the i-th sensitive area. , when the actual width of the j-th rail gap in the i-th sensitive area ≤ the closing judgment threshold, the rail gap is judged to be in a closed state; When the actual width of the j-th rail gap in the i-th sensitive area > closing judgment threshold, the rail gap is judged to be in an unclosed state; S403: Count the number of closed rail joints in the i-th sensitive area and mark them accordingly. Calculate the total number of closed rail joints in the i-th sensitive area and then calculate the ratio of the total number of rail joints in the i-th sensitive area to obtain the rail joint closure rate of the i-th sensitive area. The closure determination threshold is set at 2mm. Image recognition technology automatically detects rail gaps on the track surface, avoiding subjective errors associated with manual labeling and improving detection accuracy and efficiency. By setting a specific closure determination threshold, the determination of rail gap status is more scientific and standardized, avoiding misjudgments due to the track environment or human factors.

[0040] S404: Extract the rail vibration value of the i-th sensitive area at time t in step 102 Before the train arrives, a camera or laser scanner is used to capture images of each sensitive area, and an image recognition algorithm is used to locate the rail gap position of each sensitive area, and the initial width of the j-th rail gap in the i-th sensitive area is determined. , the vibration sensor monitors the rail vibration value of the i-th sensitive area at time t in real time , preset vibration threshold, when the rail vibration value of the i-th sensitive area at time t is When the vibration threshold is reached, it is determined that a train is approaching and the camera starts recording. When the train wheel passes through the rail gap, the disturbance width of the jth rail gap in the i-th sensitive area after the train passes at time t is extracted. Calculate the track gap shrinkage rate of the jth track gap in the i-th sensitive area at time t : Where, Indicates the time interval when the train wheels pass through the rail gap; S405: Based on the track gap shrinkage rate of the jth track gap in the i-th sensitive area at time t , calculate the average track gap shrinkage rate of the i-th sensitive area at time t : S406: Combine the track gap closure rate of the i-th sensitive area , the average track gap shrinkage rate of the i-th sensitive area at time t and the rail vibration value of the i-th sensitive area at time t , after dimensionless processing, the track damage risk coefficient of the i-th sensitive area at time t is obtained : Where, 、 and are constants, and , Indicates that 、 and Weight, calculate the track damage risk coefficient of the i-th sensitive area at time t .

[0041] S407. Track damage risk coefficient for the i-th sensitive area at time t Grading to obtain a third assessment level includes: When 0.00< When ≤0.60, it means that the track gap change in the sensitive area is within the expected normal range, the track is in good condition, and continuous monitoring is required; when When it is greater than 0.61, it indicates that there is an overload risk in the sensitive area, and there is a risk of fatigue of the rails and track structure due to long-term rail gap closure, generating Level V and triggering the fifth abnormal warning.

[0042] In this embodiment, by calculating the rail gap closure rate, the track status of each sensitive area can be clearly understood, and it can be determined whether the track has serious problems such as deviation, settlement or fatigue damage. Before the train approaches, the vibration sensor is used to monitor the vibration value of the rail in real time, and a vibration threshold is preset. When the vibration threshold is reached, the system automatically records the disturbance width when the train passes through the rail gap and calculates the rail gap reduction rate. The vibration sensor, combined with the disturbance width when the train passes, can evaluate in detail the actual impact of the train on the track and provide strong evidence for the dynamic health status of the track. The average reduction rate of the rail gap in each sensitive area after the train passes is calculated, and this is used as a basis for evaluating the degree of damage to the track. By comprehensively considering data from multiple dimensions such as the rail gap closure rate, reduction rate and vibration, the risk of track damage can be assessed more comprehensively and accurately.

[0043] Example 7 This embodiment is explained in Example 6. Specifically, step 5 includes: Generate corresponding strategies for Levels I to V, including: Level I generates the first maintenance strategy, including: In sensitive areas, when summer temperatures exceed 35°C, water cooling systems are used to spray the track every 1.5 to 2 hours. When winter temperatures fall below -30°C, track heating systems are used to maintain track surface temperatures between -10°C and +5°C. By using the water cooling system for cooling in the summer and the heating system for insulation in the winter, the track is effectively protected from thermal expansion and contraction caused by large temperature fluctuations, thereby preventing deformation or damage. This temperature management reduces the impact of external temperature fluctuations on the track and extends its service life.

[0044] Level II generates a second maintenance strategy, including: In sensitive areas, when summer temperatures exceed 35°C, a water cooling system is used to spray water on the track every 30 minutes to one hour; when winter temperatures fall below -30°C, a track heating system is used to maintain the track surface temperature between -5°C and +10°C; and enhanced summer cooling measures and winter heating measures are implemented to prevent excessive physical damage to the track in high and low temperature environments, reducing thermal cracks or shrinkage caused by temperature fluctuations. Compared to Level I, frequent cooling and heating allow for more precise regulation of track temperature, thereby reducing the impact of temperature extremes and ensuring track stability under long-term, high-load operation.

[0045] Level III generates the third maintenance strategy, which includes: localized repairs to sensitive areas, including replacing severely worn rails or reinforcing track joints, increasing grouting coverage by 30% to reinforce sensitive areas, and conducting bimonthly inspections. Replacing severely worn rails and reinforcing track joints effectively restores the track structure to normal operation, prevents further localized damage, and extends the overall track life. Localized reinforcement and repairs avoid large-scale track replacement, saving costs and quickly resolving the problem.

[0046] Level IV generates the fourth maintenance strategy, which includes: implementing foundation reinforcement measures for sensitive areas with soft soil or waterlogging to prevent further settlement, using track correction equipment to adjust deviated sections and restore the track to normal condition, and conducting monthly inspections. Foundation reinforcement in soft soil or waterlogged areas effectively prevents track unevenness or deviation caused by foundation settlement, ensuring long-term track stability. Using track correction equipment to adjust deviated sections restores the track to normal condition, eliminating safety hazards posed by deviated sections to train operations.

[0047] Level V generates the fifth maintenance strategy, which includes: using a rail gap opener or expander to open 80% of the closed rail gaps in sensitive areas, gradually increasing the gap clearance and restoring the track structure. When severely closed rail gaps cannot be restored mechanically, track joints and gap plates must be replaced or reinforced. Overloaded vehicles must be monitored and speed limits or restrictions on the passage of overloaded vehicles must be implemented. Using a rail gap opener or expander to restore closed rail gaps to normal prevents train failures or derailments caused by closed rail gaps. Restricting the passage of overloaded vehicles or implementing speed limits effectively prevents damage to the track structure caused by overloading, ensuring that trains operate within the track's carrying capacity. When rail gaps are severely closed or cannot be restored mechanically, track joints and gap plates must be replaced or reinforced to ensure the structural integrity and safety of the track system. Monitoring and managing overloads not only protects the safety of the track but also safeguards the lives and property of train passengers, reducing major accidents caused by track anomalies.

[0048] In this embodiment, maintenance strategies 1 through 5 provide progressively stronger response measures for varying levels of track damage, ensuring comprehensive protection from routine management to major repairs. Whether in routine maintenance or extreme repairs, regular monitoring, localized repairs, reinforcement measures, and meticulous management ensure the safety, stability, and long-term effectiveness of the track. This not only improves the operational efficiency of the railway system but also helps reduce the risk of accidents caused by track damage.

[0049] It should be noted that all calculation formulas in this application document utilize, including but not limited to, regression analysis within machine learning algorithms to deeply analyze the collected parameters and identify their natural trends and interrelationships. Professional software, such as Python's Scikit-learn library or the R language, is used to automatically generate mathematical models that match the data. Model performance is then objectively evaluated through methods such as cross-validation, combined with continuous feedback and optimization to ensure that the created formulas truly reflect the inherent laws of the data, thereby guaranteeing their validity and accuracy, and ensuring that the calculation process complies with the constraints of natural laws rather than being based on artificially set rules.

[0050] The technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory (FLASH), hard disk or optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute the methods of various embodiments of the present invention.

[0051] The logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing the logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device). For purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0052] It should be noted that the above embodiments are only used to illustrate the technical solutions 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 preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

[0053] It should be noted that the above embodiments are only used to illustrate the technical solutions 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 preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for identifying expansion defects of subway tracks, characterized in that: The specific steps include: Step 1: Divide the subway track into several sensitive areas and deploy fiber Bragg grating sensors in the sensitive areas to continuously monitor the strain and temperature of the i-th sensitive area, analyze the correlation between the sudden increase in strain and the temperature change, and form the first evaluation value of the corresponding point of the i-th sensitive area. ; Step 2: Integrate the nighttime "cold shrinkage-sun expansion" fatigue cycle effect to conduct risk retrospective judgment, trace the track temperature-strain fluctuation curve in the past three days, and construct the rail fatigue factor of the i-th sensitive area. , and the first evaluation value of the corresponding point in the i-th sensitive area Correction is performed to correct the deviation between the daytime detection results and the missing data at night, and the first evaluation value of the corrected i-th sensitive area is obtained. After obtaining the first classification results, a key monitoring section group is established; Step 3: For the key monitoring section group, periodically collect the track 3D point cloud data and detect the track geometric distortion to form the track state index of the i-th sensitive area. and graded to obtain a second grade result; Step 4: A drone equipped with a high-resolution camera cruises and photographs the track, collects track surface images and performs intelligent recognition to construct the track damage risk coefficient of the i-th sensitive area at time t. and graded to obtain a third grade result; Step 5: Generate corresponding maintenance strategies for the first classification result, the second classification result, and the third classification result.

2. The method for identifying expansion defects on subway tracks according to claim 1, characterized in that: The step 1 comprises: Step 101: According to the track design drawings, identify the joint sections, welding sections, curved sections and bridge intersection sections in the track, set them as target areas prone to track expansion, and divide them into several sensitive areas at equal intervals; each sensitive area is marked as the i-th monitoring section, and the length of the track section corresponding to the node is calculated as , the distance between sensor nodes is marked as , and record the track structure attribute labels at the node position, including rail type, fastener type and rail grade; Step 102: At the fishplate connection in each i-th sensitive area, a plurality of fiber Bragg grating sensor nodes are evenly arranged along the track direction; Multiple Fiber Bragg Grating sensor nodes include: The temperature sensor is used to collect the temperature data of the rail web and obtain the rail temperature value of the i-th sensitive area at time t. ; The strain sensor is used to collect the longitudinal strain changes of the rail and obtain the longitudinal strain value of the rail in the i-th sensitive area at time t. ; Vibration sensor, used to identify disturbances caused by trains, determine track structure looseness or vibration change trends, and obtain the rail vibration value of the i-th sensitive area at time t ; Extract the temperature curve and strain curve in the sensitive area or the second area, and construct the first evaluation value of the i-th sensitive area , the specific construction steps are: S111. Extract the longitudinal strain value of the rail in the i-th sensitive area at time t , the strain increase rate of the i-th sensitive area is calculated by the following formula : Where, represents the maximum rate of change of strain over time in the i-th region during the monitoring period, is the strain change rate, calculated by the inverse method, Represents the inverse, takes the derivative of t, represents the rate of change over time; and sets the deformation threshold. When the deformation threshold is exceeded, the time when the maximum value occurs is recorded ; S112, according to the time when the maximum value in S111 occurs , it is necessary to investigate the cause and extract the time when the maximum value occurs Calculate the critical fluctuation value of rail temperature in the i-th sensitive area according to the rail temperature at that time , the expression is as follows: Where, express When the deformation threshold is exceeded, the time when the maximum value occurs is recorded Rail temperature at ， Indicates the temperature value of the rail in the initial stable state of operation; S113. Calculate the hysteresis correlation factor of the i-th sensitive area : Where, Indicates the delay time of the effect of temperature change on strain. represents the Pearson correlation coefficient, represents the strain value at the delay time; S114: Combine the strain increase rate of the i-th sensitive area obtained in S111-S113 , critical fluctuation value of rail temperature and the hysteresis correlation factor of the i-th sensitive area After dimensionless processing, the first evaluation value of the i-th sensitive area is calculated by the following formula: : Where, 、 and are constants, and , Indicates that 、 and Weight, calculate the first evaluation value of the i-th sensitive area .

3. The method for identifying expansion defects on subway tracks according to claim 2, characterized in that: The step 2 includes: S201. Collect real-time monitoring data on track temperature and strain, covering at least the past three days. Collect daytime temperature rise data and nighttime temperature data. Use curve fitting methods to analyze the changes in the "cold contraction-sunday expansion" stress cycle and construct the periodic change intensity of the track in the i-th sensitive area at time t. : Where, is the base stress, is the stress cycle amplitude, which indicates the intensity of the "cold shrinkage-sun expansion" effect; is the angular frequency at time t, representing the daily cycle change. When the temperature change is a 24-hour cycle, the angular frequency ; Indicates phase shift; S202: The periodic variation intensity of the orbit of the i-th sensitive area at time t calculated according to S201 , combined with the maximum stress amplitude that the rail material in the i-th sensitive area continuously withstands , according to the material characteristics of the rail and experimental data, the rail fatigue factor of the i-th sensitive area is calculated by the following formula : Among them, the rail fatigue factor of i sensitive area is The larger the value, the greater the stress cycle amplitude caused by the "cold shrinkage-sun expansion" effect, and the higher the accumulated fatigue damage. S203, based on the rail fatigue factor of the i-th sensitive area , the first evaluation value of the i-th sensitive area After correction, the first evaluation value of the i-th sensitive area after correction is obtained by the following formula : Where, It represents the correction coefficient, which indicates the influence of fatigue factor on the evaluation value, and is determined based on actual monitoring data and experimental results.

4. The method for identifying expansion defects on subway tracks according to claim 3, characterized in that: The step 2 further comprises: S204: Calculate the first evaluation value of the corrected i-th sensitive area Grading is performed to obtain the first assessment level, including: When 0.00< When ≤0.60, it means there is no risk of track deformation caused by thermal expansion and continuous monitoring is required; When 0.61< When ≤0.80, it indicates that there is a risk of track deformation caused by thermal expansion, generating Level I and triggering the first abnormality warning; 0.81< When ≤1.00, it indicates that there is a risk of track deformation caused by thermal expansion, which is higher than Level I, generating Level II and triggering the second abnormality warning; S205. The sensitive areas that trigger the first abnormal warning and the second abnormal warning are filed as a key monitoring section group.

5. The method for identifying expansion defects on subway tracks according to claim 1, characterized in that: The step 3 comprises: S301. Use mobile laser scanning equipment, including vehicle-mounted LiDAR, to conduct track inspections within key monitoring sections at predetermined intervals, obtaining reference point cloud data of the track in its initial state and real-time point cloud data during the current inspection. S302: Segment the reference point cloud data into several sensitive areas according to step 101, and perform coordinate registration between the reference point cloud data and the real-time point cloud data during the current detection; S303: Extract two track top edge lines in each sensitive area, fit the center line, gauge line and rail top elevation curve for each track; output the set of center line points of the sensitive area, and output the lateral offset value of the i-th sensitive area. , longitudinal offset value , cumulative settlement value , Gauge change value , distortion rate change rate , geometric profile deviation and roughness , and the method of obtaining it is as follows: Compare the geometric centerline point set of the sensitive area with the reference point cloud data to obtain: lateral offset value , longitudinal offset value and cumulative settlement value ; Extract the left and right rail head edge points at the positioning point of the i-th sensitive area, calculate the horizontal distance between the two points, obtain the track gauge, and compare the track gauge with the design track gauge in the benchmark point cloud data to obtain the track gauge change value. ; Extract the left and right rail top elevations at the positioning point in the i-th sensitive area, calculate the left and right height difference, and perform differential slope fitting to obtain the distortion rate change rate. : At the mileage point of the i-th sensitive area, perpendicular to the track direction, a cross section is cut and the contour lines are fitted, including the rail head, rail waist and rail bottom, and the geometric contour deviation is obtained by point-by-point deviation calculation with the reference point cloud data. ; At the mileage point in the i-th sensitive area, the rail top elevation Z value is extracted along the track centerline from the real-time point cloud data during the current detection to form a continuous elevation sequence. The continuous elevation sequence is Fourier transformed to analyze the frequency range corresponding to the main peak in the frequency and output the roughness. .

6. The method for identifying expansion defects on subway tracks according to claim 5, characterized in that: The step 3 further comprises: S304: Extract the lateral offset value corresponding to the i-th sensitive area , longitudinal offset value , cumulative settlement value , Gauge change value , distortion rate change rate , geometric profile deviation and roughness , after dimensionless processing, the orbital state index of the i-th sensitive area is calculated : Where, 、 、 、 、 、 and All are weights, and the sum of the weights is 1; S305: Track status index for the i-th sensitive area Grading to obtain a second assessment level includes: When 0.00< When ≤0.40, it means that the track deviation is within the preset expected range and is continuously monitored; When 0.41< When the value is ≤0.70, it indicates that there is a moderate risk of track deviation due to wear or aging, generating Level III and triggering the third abnormality warning; 0.71< When ≤1.00, it indicates that there is a risk of height deviation caused by wear or settlement of the track, and the risk is higher than Level III, generating Level IV and triggering the fourth abnormal warning.

7. The method for identifying expansion defects on subway tracks according to claim 5, characterized in that: The step 4 comprises: S401. Use a drone with a high-resolution camera to regularly cruise and photograph the track, collect all-round image data of the track surface, and perform preliminary preprocessing on the collected image data, including noise removal, distortion correction, and color correction. S402: Using image recognition technology, extract the rail gap position of each sensitive area in the image data and mark it, and extract the actual width of the rail gap at the jth position in the i-th sensitive area. , when the actual width of the jth rail gap in the i-th sensitive area ≤ the closing judgment threshold, the rail gap is judged to be in a closed state; When the actual width of the j-th rail gap in the i-th sensitive area > closing judgment threshold, the rail gap is judged to be in an unclosed state; S403: Count the number of closed rail joints in the i-th sensitive area and mark them accordingly. Calculate the total number of closed rail joints in the i-th sensitive area and then calculate the ratio of the total number of rail joints in the i-th sensitive area to obtain the rail joint closure rate of the i-th sensitive area. ; The closing determination threshold is set to 2mm.

8. The method for identifying expansion defects on subway tracks according to claim 7, characterized in that: The step 4 further comprises: S404: Extract the rail vibration value of the i-th sensitive area at time t in step 102 Before the train arrives, a camera or laser scanner is used to capture images of each sensitive area, and an image recognition algorithm is used to locate the rail gap position of each sensitive area, and the initial width of the j-th rail gap in the i-th sensitive area is determined. , the vibration sensor monitors the rail vibration value of the i-th sensitive area at time t in real time , preset vibration threshold, when the rail vibration value of the i-th sensitive area at time t is When the vibration threshold is reached, it is determined that a train is approaching and the camera starts recording. When the train wheel passes through the rail gap, the disturbance width of the jth rail gap in the i-th sensitive area after the train passes at time t is extracted. Calculate the track gap shrinkage rate of the jth track gap in the i-th sensitive area at time t : Where, Indicates the time interval when the train wheels pass through the rail gap; S405: Based on the track gap shrinkage rate of the jth track gap in the i-th sensitive area at time t , calculate the average track gap shrinkage rate of the i-th sensitive area at time t : S406: Combine the track gap closure rate of the i-th sensitive area , the average track gap shrinkage rate of the i-th sensitive area at time t and the rail vibration value of the i-th sensitive area at time t , after dimensionless processing, the track damage risk coefficient of the i-th sensitive area at time t is obtained : Where, 、 and are constants, and , Indicates that 、 and Weight, calculate the track damage risk coefficient of the i-th sensitive area at time t .

9. The method for identifying expansion defects on subway tracks according to claim 8, characterized in that: The step 4 further comprises: S407. Track damage risk coefficient for the i-th sensitive area at time t Grading to obtain a third assessment level includes: When 0.00< When ≤0.60, it means that the track gap change in the sensitive area is within the expected normal range, the track is in good condition, and continuous monitoring is required; when When it is greater than 0.61, it indicates that there is an overload risk in the sensitive area, and there is a risk of fatigue of the rails and track structure due to long-term rail gap closure, generating Level V and triggering the fifth abnormal warning.

10. The method for identifying expansion defects of subway tracks according to claim 9, characterized in that: The step 5 comprises: Generate corresponding strategies for Levels I to V, including: Level I generates the first maintenance strategy, including: in this sensitive area, when the temperature exceeds 35°C in summer, the track in this area will be sprayed with water every 1.5 to 2 hours to cool it down; when the temperature drops below -30°C in winter, the track heating system will be used to maintain the track surface temperature between -10°C and +5°C; Level II generates the second maintenance strategy, including: in this sensitive area, when the temperature exceeds 35°C in summer, the track in this area will be sprayed with water every 30 minutes to 1 hour to cool it down; when the temperature drops below -30°C in winter, the track heating system will be used to maintain the track surface temperature between -5°C and +10°C; Level III generates the third maintenance strategy, which includes: local repair of the sensitive area, including replacing severely worn rails or reinforcing track joints, increasing the grouting coverage by 30% to reinforce the sensitive area, and conducting inspections every two months; Level IV generates the fourth maintenance strategy, which includes: taking foundation reinforcement measures for the soft soil layer or waterlogged areas in the sensitive area to prevent further settlement, using track correction equipment to adjust the offset parts to restore the normal condition of the track, and conducting monthly inspections; Level V generates the fifth maintenance strategy, which includes: using a rail gap opener or rail gap expander in the sensitive area to expand 80% of the closed rail gaps in the sensitive area, gradually increasing the gap between the rail gaps to restore the track structure. When severely closed rail gaps cannot be restored to normal through mechanical adjustment, the rail joints and rail gap plates need to be replaced or reinforced; and overloaded vehicles are monitored, and measures such as speed limit or restriction of the passage of overloaded vehicles are taken.

Images