Distributed fiber based method and system for monitoring bridge breakage

Abstract

The application provides a kind of bridge fracture monitoring method and system based on distributed optical fiber, it is related to bridge monitoring technical field, first, the stress response correlation mapping table of distributed optical fiber and bridge structure is constructed, records the information such as laying position, sensing range parameter and response sensitivity curve, then based on stress response correlation mapping table dynamic adjustment collection frequency, generate the strain and environmental correlation data set with collection frequency identification, environmental interference stripping processing is carried out to strain and environmental correlation data set, form environmental correction after space strain deviation feature set, again the space strain deviation feature set is associated with time dimension strain data change trend modeling, combine response sensitivity curve and generate the space-time strain evolution characteristics with sensitivity weight, finally, through the abnormal evolution section marked by multi-fiber cooperative verification mechanism, generate the bridge fracture monitoring result containing verification result, so as to accurately, comprehensively monitor bridge fracture condition.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of bridge monitoring, in particular to a bridge fracture monitoring method and system based on distributed optical fibers. BACKGROUND

[0002] With the increase of bridge service life and the increasing traffic flow, the load borne by the bridge structure is constantly changing, and potential problems such as fatigue damage and crack propagation are prone to occur in the key stress areas of the bridge structure. If these problems are not discovered and addressed in a timely manner, they may cause serious safety accidents such as bridge fracture, which poses a great threat to people's life and property safety.

[0003] At present, the traditional bridge monitoring method mainly relies on periodic manual inspection and some conventional sensor monitoring. Manual inspection is not only inefficient, but also difficult to comprehensively and accurately detect the internal structural damage of the bridge, especially for some hidden damage. Although conventional sensor monitoring can obtain certain data, it often has problems such as limited monitoring range, low data accuracy, and inability to reflect the overall stress state of the bridge in real time. In addition, the existing monitoring methods do not fully consider the influence of environmental factors on the monitoring data, resulting in large errors in the monitoring results and making it difficult to accurately determine the true condition of the bridge. SUMMARY

[0004] In view of the above-mentioned problems, in combination with the first aspect of the present application, the present application embodiment provides a bridge fracture monitoring method based on distributed optical fibers, which comprises:

[0005] A stress response correlation mapping table of distributed optical fibers and bridge structure is constructed, which records the correspondence between the layout positions of the distributed optical fibers and the key stress areas of the bridge structure, the sensing range parameters of the distributed optical fibers at each layout position, and the response sensitivity curves of the distributed optical fibers under different stress actions;

[0006] Based on the stress response correlation mapping table of distributed optical fibers and bridge structure, the collection frequency of the distributed optical fibers is dynamically adjusted for different use states of the bridge, the strain data of the distributed optical fibers and the synchronous environmental influence factor data under the corresponding state are collected, and a strain and environment correlation data set with collection frequency identification is generated;

[0007] The strain and environment correlation data set with collection frequency identification is subjected to environmental interference stripping processing, the strain data of different distributed optical fibers in the same key stress area of the bridge is corrected according to the environmental influence factor data, the difference of the corrected strain data is extracted, and a spatial strain deviation feature set after environmental correction is formed;

[0008] Correlate the environment-corrected spatial strain deviation feature set with the strain data change trend in the time dimension, combine the response sensitivity curve in the stress response correlation mapping table of the distributed optical fiber and the bridge structure, generate a spatial and temporal strain evolution feature with sensitivity weight, the spatial and temporal strain evolution feature with sensitivity weight presents the change law of strain deviation of different bridge key stress areas with time and the response weight of the corresponding optical fiber;

[0009] Based on the spatial and temporal strain evolution feature with sensitivity weight, an abnormal evolution section conforming to the bridge pre-fracture feature is marked through a multi-fiber cooperative verification mechanism, a bridge fracture monitoring result containing verification results is generated, the bridge fracture monitoring result containing verification results contains the bridge key stress area corresponding to the abnormal evolution section, the strain deviation change details of the bridge key stress area and the verification consistency information of the multi-fiber data.

[0010] In still another aspect, the embodiment of the present application also provides a bridge fracture monitoring system based on a distributed optical fiber, characterized in that it comprises:

[0011] A processor; a machine readable storage medium for storing machine executable instructions of the processor; wherein the processor is configured to execute the above-mentioned bridge fracture monitoring method based on a distributed optical fiber by executing the machine executable instructions.

[0012] In still another aspect, the embodiment of the present application also provides a computer program product, the computer program product comprises machine executable instructions, the machine executable instructions are stored in a computer readable storage medium, a processor of a computer device reads the machine executable instructions from the computer readable storage medium, the processor executes the machine executable instructions, so that the computer device executes the above-mentioned bridge fracture monitoring method based on a distributed optical fiber.

[0013] Based on the above aspects, by constructing a stress response correlation mapping table of the distributed optical fiber and the bridge structure, the correspondence relationship between the distributed optical fiber layout position and the bridge key stress area, the sensing range parameter and the response sensitivity curve are recorded. Based on the stress response correlation mapping table, the collection frequency is dynamically adjusted and the strain and environment correlation data set with collection frequency identification is generated. The data can be flexibly obtained according to different use states of the bridge, which not only ensures the timeliness of the data, but also avoids unnecessary data redundancy, and improves the efficiency and pertinence of data collection. The environmental interference stripping processing is performed on the data set, which effectively eliminates the interference of environmental factors on the monitoring data, and the extracted environmental corrected spatial strain deviation feature set is more real and reliable, which can accurately reflect the strain condition of the bridge key stress area. The spatial strain deviation feature set and the strain data change trend in time dimension are correlated and modeled, and the space-time strain evolution feature with sensitivity weight is generated combined with the response sensitivity curve, which can comprehensively present the strain deviation change law of different bridge key stress areas and the response weight of the corresponding optical fiber from the space-time two dimensions. Finally, the abnormal evolution section conforming to the bridge early fracture characteristics is marked through the multi-optical fiber cooperative verification mechanism, and the bridge fracture monitoring result containing the verification result is generated, which can timely find the potential fracture risk of the bridge, and provide the bridge key stress area corresponding to the abnormal evolution section, the strain deviation change details and the verification consistency information of the multi-optical fiber data in detail, greatly improve the accuracy and reliability of the bridge fracture monitoring, and effectively guarantee the safe operation of the bridge. BRIEF DESCRIPTION OF DRAWINGS

[0014] Figure 1 is the execution flow diagram of the bridge fracture monitoring method based on the distributed optical fiber provided by the embodiment of the application.

[0015] Figure 2 is the schematic diagram of exemplary hardware and software components of the bridge fracture monitoring system based on the distributed optical fiber provided by the embodiment of the application. DETAILED DESCRIPTION

[0016] The application will be described in detail below with reference to the accompanying drawings of the specification, Figure 1 is the flow diagram of the bridge fracture monitoring method based on the distributed optical fiber provided by an embodiment of the application, and the bridge fracture monitoring method based on the distributed optical fiber will be described in detail below.

[0017] Step S110: constructing a stress response correlation mapping table of the distributed optical fiber and the bridge structure, the stress response correlation mapping table of the distributed optical fiber and the bridge structure recording the correspondence relationship between the distributed optical fiber layout position and the bridge key stress area, the sensing range parameter of the distributed optical fiber at each layout position, and the response sensitivity curve of the distributed optical fiber under different stress actions.

[0018] Step S111: Obtain the structural design file of the bridge and the material mechanics parameters, extract the key stress region information of the bridge based on the structural design file, the material mechanics parameters include the elastic modulus, Poisson's ratio and tensile strength parameters of the bridge main structure material, and the key stress region information of the bridge includes the main beam support point region, the mid-span stress region, the pier connection region and the support stress region.

[0019] In this embodiment, the digital delivery system of the bridge design unit is used to obtain the structural design file including two-dimensional drawings, three-dimensional models and material lists, wherein the material list is stored in a structured data format, and the concrete strength grade, steel type and other parameters can be directly parsed. For the elastic modulus, Poisson's ratio and tensile strength parameters, the standard test data in the material mechanics performance test report is used to ensure that the parameter accuracy meets the requirements of finite element analysis. When extracting the key stress region information, the load combination description in the structural design file is combined, the specific position and influence range of each support point in the main beam support point region need to be clearly marked, the different sub-regions of the upper and lower edges, webs and bottom plates in the mid-span stress region need to be divided, the pier top anchoring segment, pier body transition segment and bottom pile cap connection segment in the pier connection region need to be included, and the steel pad, rubber pad and surrounding concrete region of each support need to be marked respectively in the support stress region.

[0020] Step S112: Perform three-dimensional stress distribution simulation analysis on each key stress region of the bridge, construct a stress field model of the key stress region by using the finite element analysis method, and determine the stress concentration sub-region in each key stress region of the bridge by using the stress field model of the key stress region, wherein the stress concentration sub-region in each key stress region of the bridge is a local continuous region with higher stress value distribution than the surrounding region in the key stress region of the bridge.

[0021] In this embodiment, for the main beam support point region, a local fine model including the main beam, support and pier top is established by using the finite element analysis software, the model size is scaled according to the actual bridge structure proportion, the hexahedral element is used for mesh division, and the mesh size is set to match the characteristic length of the material mechanics parameters. The simulation working conditions include automobile load, temperature load, support settlement and earthquake action, and the load parameters of each working condition are determined according to the bridge design specification. By analyzing the numerical distribution of the stress nephogram, the local continuous region with more than 15% stress value than the surrounding region is extracted as the stress concentration sub-region, the boundary of the stress concentration sub-region is determined by mathematical fitting through the mutation point of the stress contour line, and the spatial range of the sub-region is consistent with the geometric characteristics of the weak stress part of the actual bridge structure.

[0022] Step S113: For each stress concentration sub-region, determine the layout path direction of the distributed optical fiber in the stress concentration sub-region according to the stress distribution gradient output by the stress field model of the key stress region. The layout path direction of the distributed optical fiber in the stress concentration sub-region needs to be consistent with the maximum change direction of the stress distribution gradient, and the layout path needs to cover the entire spatial range of the stress concentration sub-region.

[0023] In this embodiment, for the stress concentration sub-region of the main beam midspan stress region, the maximum change direction of the stress distribution gradient is determined by calculating the eigenvalue of the stress tensor, which is consistent with the principal stress direction of the main beam bending stress. The layout path direction of the distributed optical fiber is set to extend along the principal stress direction from the midspan to both ends, covering the upper and lower edges and web regions of the entire stress concentration sub-region. When determining the layout path, a three-dimensional path planning algorithm is used in combination with the surface curvature of the bridge structure, the installation space and the fiber bending radius limit to generate a continuous spatial path, ensuring that the fiber can closely adhere to the surface of the bridge structure while avoiding interference with steel bars, embedded parts and other components.

[0024] Step S114: Select multiple test points on the layout path, and perform stress response testing on the distributed optical fiber at each test point. By applying different gradients of standard stress, record the strain output value of the distributed optical fiber under each standard stress, and based on the correspondence between the strain output value and the standard stress, generate a response sensitivity curve for the distributed optical fiber at each test point.

[0025] In this embodiment, multiple test points are selected on the selected layout path in an equal or variable interval manner. The number of test points is determined according to the spatial size and stress variation complexity of the stress concentration sub-region. A standard stress loading device is used to load each test point in one direction or multiple directions. The loading gradient is determined according to the design stress range of the bridge structure, and the loading time of each gradient is kept consistent to ensure that the fiber strain reaches a stable state. The distributed optical fiber sensing system is used to record the strain output value of each test point under different standard stresses. The strain output value is stored in time series form, including the start time, end time and strain data at each time. Based on the correspondence between the strain output value and the standard stress, a response sensitivity curve for the distributed optical fiber at each test point is generated using the least squares method. The fitting accuracy of the curve is verified by residual analysis to ensure that the curve can accurately reflect the stress-strain characteristics of the fiber.

[0026] Step S115: Calculate the average curve of the response sensitivity curves of all test points on each layout path, and take the average curve as the distributed optical fiber response sensitivity curve corresponding to the layout path. Meanwhile, record the distributed optical fiber layout position, layout path direction, sensing range parameters and response sensitivity curve corresponding to each bridge key stress area, form a distributed optical fiber and bridge structure stress response correlation mapping table, and establish a corresponding relationship between each bridge key stress area and the layout position of at least three distributed optical fibers in the distributed optical fiber and bridge structure stress response correlation mapping table. The layout positions of the three distributed optical fibers are located in different stress gradient areas of the stress concentration sub-area, so as to realize multi-gradient strain data acquisition in the same bridge key stress area.

[0027] In this embodiment, for each layout path, the response sensitivity curves of all test points are converted into the same stress interval, the average value of the strain output values of all curves in each stress interval is calculated, and the average response sensitivity curve corresponding to the layout path is generated. The distributed optical fiber layout position, layout path direction, sensing range parameters and response sensitivity curve corresponding to each bridge key stress area are stored in a structured database, wherein the sensing range parameters include the spatial coverage of the optical fiber, the strain measurement range and the measurement accuracy. A corresponding relationship is established between each bridge key stress area and the layout position of at least three distributed optical fibers, and the three layout positions are located in the high stress gradient area, the medium stress gradient area and the low stress gradient area of the stress concentration sub-area, respectively, to ensure that strain data under different stress levels can be collected.

[0028] Step S120: Based on the distributed optical fiber and bridge structure stress response correlation mapping table, dynamically adjust the collection frequency of the distributed optical fiber for different bridge use states, collect the distributed optical fiber strain data and the synchronous environmental influence factor data under the corresponding state, and generate strain and environmental correlation data set with collection frequency identification.

[0029] Step S121: Extract the layout position, corresponding bridge key stress area and response sensitivity curve of each distributed optical fiber from the distributed optical fiber and bridge structure stress response correlation mapping table, analyze the stress change characteristics of the key stress area under different bridge use states, and the different bridge use states include the bridge empty load state, the bridge full load state, the bridge temporary load state, the bridge wind load state in the environmental load and the bridge temperature load state in the environmental load.

[0030] In this embodiment, by querying the stress response correlation mapping table of the distributed optical fiber and the bridge structure, the unique identifier, layout position coordinates, corresponding bridge key stress area name and response sensitivity curve data of each distributed optical fiber are extracted. For the bridge empty state, the stress change characteristics of the key stress area are analyzed. Under the bridge empty state, the bridge only bears self-weight load, the stress change amplitude is small and stable. For the bridge full load state, the stress change characteristics of the key stress area under the action of vehicle load are analyzed. Under the bridge full load state, the stress change amplitude is large and has dynamic characteristics. For the bridge under temporary load state, the stress change characteristics of the key stress area under the action of construction load or sudden event load are analyzed. Under the temporary load state, the stress change has suddenness. For the bridge under wind load state in environmental load, the stress change characteristics of the key stress area under the action of wind load are analyzed. Under the wind load state, the stress change has periodic fluctuation characteristics. For the bridge under temperature load state in environmental load, the stress change characteristics of the key stress area under the action of temperature change are analyzed. Under the temperature load state, the stress change has correlation with temperature change.

[0031] Step S122: For each bridge use state, the collection frequency reference value of strain data is determined according to the stress change characteristics of the corresponding bridge key stress area. Under the bridge empty state, the stress change is gentle, and the first type of collection frequency reference value is set. Under the bridge full load state and the bridge under temporary load state, the stress change is severe, and the second type of collection frequency reference value is set. The time interval of the first type of collection frequency reference value is longer than the time interval of the second type of collection frequency reference value.

[0032] In this embodiment, for the bridge empty state, since the stress change is gentle and the change rate of strain data is slow, the first type of collection frequency reference value is set, and the time interval of the first type of collection frequency reference value is set to a long time, which ensures that the change of long-term stress state of the bridge can be covered. For the bridge full load state and the bridge under temporary load state, since the stress change is severe and the change rate of strain data is fast, the second type of collection frequency reference value is set, and the time interval of the second type of collection frequency reference value is set to a short time, which ensures that the fast change of strain data can be captured. When determining the collection frequency reference value, the sampling rate limit of the distributed optical fiber sensing system needs to be considered to ensure that the collection frequency does not exceed the maximum sampling rate of the system.

[0033] Step S123: Establish the adjustment rule of environmental factors and collection frequency. When the environmental temperature change rate in the environmental factor data exceeds the preset change rate, or the wind speed in the wind force data exceeds the preset wind speed, the collection frequency of the corresponding distributed optical fiber is increased based on the adjustment rule of environmental factors and collection frequency. The increase amplitude is determined according to the difference value of the environmental temperature change rate exceeding the preset change rate and the difference value of the wind speed exceeding the preset wind speed. The greater the difference value, the greater the increase amplitude of the collection frequency.

[0034] In this embodiment, the adjustment rule of environmental factors and collection frequency is established. First, the preset change rate of the environmental temperature change rate and the preset wind speed of the wind speed are determined. The preset value is determined according to the climate condition of the region where the bridge is located and the design specification. When the environmental temperature change rate in the environmental factor data exceeds the preset change rate, the difference value of the environmental temperature change rate exceeding the preset change rate is calculated, and the increase amplitude of the collection frequency is determined according to the difference value. The greater the difference value, the greater the increase amplitude. When the wind speed in the wind force data exceeds the preset wind speed, the difference value of the wind speed exceeding the preset wind speed is calculated, and the increase amplitude of the collection frequency is determined according to the difference value. The greater the difference value, the greater the increase amplitude. The increase amplitude of the collection frequency is realized by adjusting the collection period, and the collection period is inversely proportional to the collection frequency.

[0035] Step S124: When the bridge is in any one of the use states and meets the collection trigger condition, the distributed optical fiber at the corresponding layout position is triggered to collect strain data at the adjusted collection frequency. During the collection process, the collection time, collection frequency and layout position of the corresponding distributed optical fiber of each strain data are recorded, and a distributed optical fiber strain sequence with collection frequency identification under the use state is formed.

[0036] In this embodiment, the bridge health monitoring system is used to monitor the use state of the bridge in real time. When the bridge is in any one of the use states and meets the collection trigger condition, the distributed optical fiber at the corresponding layout position is triggered to collect strain data. The collection trigger condition includes the change of the bridge use state, the abnormal change of the environmental factor and the manual trigger instruction. During the collection process, the distributed optical fiber sensing system records the collection time, collection frequency and layout position of the corresponding distributed optical fiber of each strain data. The strain data is stored in the form of time sequence, and each data point contains strain value, collection time, collection frequency and layout position information.

[0037] Step S125: While collecting the distributed optical fiber strain sequence with the acquisition frequency identifier, the environmental influence factor data at the corresponding time is collected by the environmental monitoring device deployed around the bridge, which includes environmental temperature data, environmental humidity data, wind data and precipitation data, so that the collection time of the environmental influence factor data is synchronized with the collection time of the strain data, and the time reference of the environmental monitoring device and the distributed optical fiber collection system is associated to achieve synchronization.

[0038] In this embodiment, while collecting the distributed optical fiber strain sequence with the acquisition frequency identifier, the environmental influence factor data at the corresponding time is collected by the environmental monitoring device deployed around the bridge, which includes temperature sensors, humidity sensors, anemometers and rain gauges. The collection time of the environmental influence factor data is synchronized with the collection time of the strain data through the global positioning system time reference, ensuring the time consistency of the data. The environmental influence factor data is stored in the form of time series, and each data point includes environmental temperature value, environmental humidity value, wind speed value, wind direction value, precipitation value and collection time information.

[0039] Step S126: The distributed optical fiber strain sequence with the acquisition frequency identifier under the same use state is associated and bound with the environmental influence factor data at the corresponding time to generate the strain and environmental association sub-data set with the acquisition frequency identifier under the use state.

[0040] In this embodiment, the distributed optical fiber strain sequence with the acquisition frequency identifier under the same use state and the environmental influence factor data at the corresponding time are associated and bound by using the timestamp matching algorithm, and the strain data and the environmental influence factor data at the same collection time are combined into one data entry. The data entry includes the distributed optical fiber layout position, the strain value, the collection time, the acquisition frequency, the environmental temperature value, the environmental humidity value, the wind speed value, the wind direction value, the precipitation value and the bridge use state identifier.

[0041] Step S127: The strain and environmental association sub-data set with the acquisition frequency identifier under all use states of the bridge is summarized to form a complete strain and environmental association data set with the acquisition frequency identifier, and each data entry in the strain and environmental association data set with the acquisition frequency identifier includes the distributed optical fiber layout position, the strain data, the collection time, the acquisition frequency, the environmental influence factor data and the corresponding bridge use state identifier.

[0042] In this embodiment, the strain and environment correlation data set with collection frequency identification in all use states of the bridge is summarized, and a data merging algorithm is used to merge all the sub-data sets into a complete strain and environment correlation data set with collection frequency identification. Each data entry in the data set contains the distributed fiber layout position, strain data, collection time, collection frequency, environmental impact factor data, and corresponding bridge use state identification. The data entries are organized in the distributed fiber layout position dimension, which facilitates subsequent data processing and analysis.

[0043] Step S130: Perform environmental interference stripping processing on the strain and environment correlation data set with collection frequency identification, correct the strain data of different distributed optical fibers in the same bridge key stress area according to the environmental impact factor data, extract the difference of the corrected strain data, and form a spatial strain deviation feature set after environmental correction.

[0044] Step S131: From the strain and environment correlation data set with collection frequency identification, all distributed optical fiber strain data corresponding to the same bridge key stress area and associated environmental impact factor data are screened out. The screening basis is the corresponding relationship between the bridge key stress area and the distributed optical fiber layout position in the stress response correlation mapping table of the distributed optical fiber and the bridge structure.

[0045] In this embodiment, by querying the stress response correlation mapping table of the distributed optical fiber and the bridge structure, a list of distributed optical fiber layout positions corresponding to each bridge key stress area is obtained. According to the list, all distributed optical fiber strain data corresponding to the same bridge key stress area and associated environmental impact factor data are screened out from the strain and environment correlation data set with collection frequency identification. The screening process uses a distributed data query algorithm to ensure the efficiency and accuracy of data screening.

[0046] Step S132: For each group of strain data and environmental impact factor data in the same bridge key stress area, an interference correlation model of environmental impact factor and strain data is established. By analyzing the change correlation between historical environmental impact factor data and strain data, the interference coefficients of environmental temperature, environmental humidity, and wind force on strain data are determined.

[0047] In this embodiment, for each group of strain data and environmental influence factor data in the same key stress area of the bridge screened out, a multivariate linear regression algorithm is used to establish an interference correlation model of the environmental influence factor and the strain data. First, the historical environmental influence factor data and the strain data are extracted, and the data is preprocessed, including data cleaning, outlier processing and normalization processing. Then, the least square method is used to estimate the interference coefficients of environmental temperature, environmental humidity and wind force on the strain data, which reflect the influence of environmental influence factors on strain data. The goodness of fit of the model is verified by the determination coefficient to ensure that the model can accurately reflect the correlation between environmental influence factors and strain data.

[0048] Step S133: Based on the interference correlation model of the environmental influence factor and the strain data and the interference coefficients, the environmental interference of each group of strain data is stripped, the strain interference value caused by the environmental influence factor is calculated, and the original strain data is subtracted from the strain interference value to obtain the strain data after environmental correction, which only reflects the strain caused by the stress of the bridge structure.

[0049] In this embodiment, based on the interference correlation model of the environmental influence factor and the strain data and the interference coefficients, the environmental interference of each group of strain data is stripped. First, the strain interference value caused by the environmental influence factor is calculated according to the environmental influence factor data and the interference coefficients, and the model prediction algorithm is used to calculate the strain interference value. Then, the original strain data is subtracted from the strain interference value to obtain the strain data after environmental correction, which only reflects the strain caused by the stress of the bridge structure, eliminating the interference of the environmental influence factor.

[0050] Step S134: Group all distributed optical fiber strain data after environmental correction in the same key stress area of the bridge according to the same collection time to obtain multiple time groupings, each time grouping containing the strain data after environmental correction of all distributed optical fibers in the key stress area of the bridge at that time.

[0051] In this embodiment, all distributed optical fiber strain data after environmental correction in the same key stress area of the bridge are grouped according to the collection time, and the timestamp matching algorithm is used in the grouping process to group the strain data of the same collection time into one time grouping. Each time grouping contains the strain data after environmental correction of all distributed optical fibers in the key stress area of the bridge at that time, and the strain data is organized in the dimension of the distributed optical fiber layout position.

[0052] Step S135: Calculate the numerical difference between the strain data after environmental correction of any two distributed optical fibers in each time grouping, and record all numerical differences and the layout positions and collection frequencies of the corresponding two distributed optical fibers.

[0053] In this embodiment, for each time grouping, the numerical difference between the strain data of any two distributed optical fibers after environmental correction is calculated, and the numerical difference is calculated using an absolute difference algorithm. Record all numerical differences and corresponding distributed optical fiber layout positions, collection frequencies, and numerical differences in matrix form. The rows and columns of the matrix correspond to the layout positions of the distributed optical fibers.

[0054] Step S136: Statistics of the distribution characteristics of the numerical difference in each time grouping, determine the maximum, minimum, average and variance of the numerical difference in the time grouping, and associate the statistical results of the maximum, minimum, average and variance of the numerical difference with the environmental influence factor data and the collection frequency corresponding to the time grouping.

[0055] In this embodiment, for the numerical difference in each time grouping, the maximum, minimum, average and variance of the numerical difference are calculated using a statistical analysis algorithm. The statistical results of the maximum, minimum, average and variance of the numerical difference are associated with the environmental influence factor data and the collection frequency corresponding to the time grouping, and the association process uses a data merging algorithm to ensure the consistency and integrity of the data.

[0056] Step S137: Summarize the numerical difference of all time groupings, the corresponding distributed optical fiber layout position, the collection frequency, the statistical results of the maximum, minimum, average and variance of the numerical difference, and the associated environmental influence factor data to form an environmental corrected spatial strain deviation feature set. Each entry in the environmental corrected spatial strain deviation feature set corresponds to spatial strain deviation related information of a time grouping, and each entry includes an identification of the environmental interference stripping process.

[0057] In this embodiment, the numerical difference of all time groupings, the corresponding distributed optical fiber layout position, the collection frequency, the statistical results of the maximum, minimum, average and variance of the numerical difference, and the associated environmental influence factor data are summarized, and all data are aggregated into an environmental corrected spatial strain deviation feature set using a data aggregation algorithm. Each entry in the feature set corresponds to spatial strain deviation related information of a time grouping, and each entry includes numerical difference, distributed optical fiber layout position, collection frequency, maximum, minimum, average, variance of numerical difference, environmental influence factor data and identification of environmental interference stripping process.

[0058] Step S140: Correlate the environmental corrected spatial strain deviation feature set with the strain data change trend in the time dimension, and combine the response sensitivity curve in the stress response correlation mapping table of the distributed optical fiber and the bridge structure to generate a time-space strain evolution feature with sensitivity weight. The time-space strain evolution feature with sensitivity weight presents the change law of strain deviation of different bridge key stress regions with time and the response weight of the corresponding optical fiber.

[0059] Step S141: Extract the time series of the environment-corrected strain data of each distributed optical fiber from the strain and environment association data set with band acquisition frequency identification, arrange them in chronological order of acquisition time, and form a distributed optical fiber corrected strain time series.

[0060] In this embodiment, the environment-corrected strain data of each distributed optical fiber is extracted from the strain and environment association data set with band acquisition frequency identification, arranged in chronological order of acquisition time, and a distributed optical fiber corrected strain time series is formed. The time series is organized in dimensions of the distributed optical fiber layout position, and each time series contains the acquisition time and the corresponding environment-corrected strain data.

[0061] Step S142: Trend extraction processing is performed on the distributed optical fiber corrected strain time series, the average change rate of strain data in each time window is calculated using the sliding window method, and the overall change trend of strain data is determined based on the average change rate. The overall change trend of strain data includes strain increasing trend, strain decreasing trend and strain stable trend.

[0062] Step S1421: According to the acquisition frequency and data length of the distributed optical fiber corrected strain time series, the time length of the sliding window is determined, which needs to cover at least three consecutive acquisition times.

[0063] In this embodiment, the acquisition frequency of the distributed optical fiber corrected strain time series is first obtained, which is expressed in the number of acquisitions per unit time, and the data length is expressed in the number of acquisition times. According to the acquisition frequency, the time interval of each acquisition time is calculated, and the time length of the sliding window is set to cover the time range of at least three consecutive acquisition times, to ensure that the change trend of strain data can be captured. For example, if the acquisition frequency is once per minute, the time length of three consecutive acquisition times is two minutes, and the time length of the sliding window is set to two minutes.

[0064] Step S1422: Starting from the starting acquisition time of the distributed optical fiber corrected strain time series, slide the window by a preset window moving step, which corresponds to one acquisition time, and each window contains all the distributed optical fiber corrected strain data in the corresponding time period.

[0065] In this embodiment, starting from the starting acquisition time of the distributed optical fiber corrected strain time series, slide the window by a preset window moving step, which is set to one acquisition time, to ensure that there is an overlapping part between the windows, and to improve the accuracy of trend extraction. Each window contains all the distributed optical fiber corrected strain data in the corresponding time period, and the data is organized in dimensions of acquisition time, which is convenient for subsequent calculation.

[0066] Step S1423: For the distributed fiber modified strain data in each sliding window, the strain data difference between the first acquisition time and the last acquisition time in the window is calculated, and the strain data difference is divided by the time interval in the window, which is the time difference between the last acquisition time and the first acquisition time in the window, to obtain the average change rate of the strain data in the sliding window.

[0067] In this embodiment, for the distributed fiber modified strain data in each sliding window, the strain data at the first acquisition time and the last acquisition time in the window is first determined, and the difference between the two is calculated. Then the time difference between the last acquisition time and the first acquisition time in the window is calculated, and the strain data difference is divided by the time difference to obtain the average change rate of the strain data in the sliding window. The average change rate reflects the average change speed of the strain data in the window, with the unit of strain unit per time unit.

[0068] Step S1424: The start time, end time and corresponding average change rate of each sliding window are recorded to form an average change rate sequence, which is arranged in the order of the sliding windows.

[0069] In this embodiment, the start time, end time and corresponding average change rate of each sliding window are recorded, the start time and end time are stored in the form of time stamp, and the average change rate is stored in the form of numerical value. The records of all sliding windows are arranged in the order of the sliding windows to form an average change rate sequence, which is organized in the order of the sliding windows.

[0070] Step S1425: The numerical distribution of the average change rate sequence is analyzed. If more than a preset proportion of the average change rates in the average change rate sequence are positive and the next average change rate is greater than the previous average change rate, it is determined that the overall change trend of the strain data of the distributed fiber modified strain time sequence is a strain increasing trend.

[0071] In this embodiment, the numerical distribution of the average change rate sequence is analyzed. First, a preset proportion is determined according to the accuracy requirement of trend extraction. If more than a preset proportion of the average change rates in the average change rate sequence are positive and the next average change rate is greater than the previous average change rate, it indicates that the change speed of the strain data is gradually increasing, and it is determined as a strain increasing trend.

[0072] Step S1426: If more than a preset proportion of the average change rates in the average change rate sequence are negative and the absolute value of the next average change rate is greater than the absolute value of the previous average change rate, it is determined that the overall change trend of the strain data of the distributed fiber modified strain time sequence is a strain decreasing trend.

[0073] In the embodiment, if the average change rate exceeding the preset proportion in the average change rate sequence is negative, and the absolute value of the next average change rate is greater than the absolute value of the previous average change rate, it indicates that the change speed of the strain data gradually increases and the direction is decreasing, and it is determined as the strain decreasing trend.

[0074] Step S1427: If the absolute value of the average change rate exceeding the preset proportion in the average change rate sequence is less than the preset threshold value and the value changes in the preset fluctuation range, the preset fluctuation range is determined based on the historical fluctuation range of the strain data in the normal use state of the bridge, it is determined that the overall change trend of the strain data of the distributed optical fiber corrected strain time sequence is the strain stable trend.

[0075] In the embodiment, if the absolute value of the average change rate exceeding the preset proportion in the average change rate sequence is less than the preset threshold value, and the value changes in the preset fluctuation range, the preset threshold value and the preset fluctuation range are determined based on the historical fluctuation range of the strain data in the normal use state of the bridge, it indicates that the change speed of the strain data is slow and stable, and it is determined as the strain stable trend.

[0076] Step S1428: The average change rate of each sliding window, the corresponding window time period and the overall change trend of the strain data are stored in association to form the trend analysis result of the distributed optical fiber corrected strain time sequence.

[0077] In the embodiment, the average change rate of each sliding window, the corresponding window time period and the overall change trend of the strain data are stored in association, the association process adopts the data merging algorithm to ensure the consistency and integrity of the data. The formed trend analysis result is organized in the dimension of the distributed optical fiber layout position, which is convenient for subsequent analysis and application.

[0078] Step S143: From the environmental corrected spatial strain deviation feature set, the statistical results of the numerical difference of each time grouping, the maximum value, the minimum value, the average value and the variance of the numerical difference, and the associated collection frequency are extracted, sorted in the order of collection time, and a spatial strain deviation time sequence is formed.

[0079] In the embodiment, from the environmental corrected spatial strain deviation feature set, the statistical results of the numerical difference of each time grouping, the maximum value, the minimum value, the average value and the variance of the numerical difference, and the associated collection frequency are extracted, sorted in the order of collection time, and a spatial strain deviation time sequence is formed. The time sequence is organized in the dimension of the collection time, and each time sequence includes the collection time, the numerical difference, the maximum value, the minimum value, the average value, the variance and the collection frequency.

[0080] Step S144: Extract the response sensitivity curve of each distributed optical fiber from the distributed optical fiber and bridge structure stress response correlation mapping table, determine the sensitivity coefficient of each distributed optical fiber in the corresponding strain range, normalize the sensitivity coefficients of all distributed optical fibers, and obtain the relative weight value of each distributed optical fiber. The relative weight value is dimensionless, and the higher the sensitivity coefficient, the greater the relative weight value.

[0081] In this embodiment, the response sensitivity curve of each distributed optical fiber is extracted from the distributed optical fiber and bridge structure stress response correlation mapping table. According to the range of the strain data after environmental correction, the sensitivity coefficient of each distributed optical fiber in the corresponding strain range is determined. The sensitivity coefficient is the slope of the response sensitivity curve in the corresponding strain range. The sensitivity coefficients of all distributed optical fibers are normalized. The normalization adopts the minimum-maximum normalization algorithm, converts the sensitivity coefficients into dimensionless relative weight values, and the value range of the relative weight value is 0 to 1. The higher the sensitivity coefficient, the greater the relative weight value.

[0082] Step S145: Establish a spatial strain deviation and strain change trend correlation model, correlate the numerical difference in the spatial strain deviation time series with the overall change trend of the strain data in the distributed optical fiber corrected strain time series, and calculate the correlation degree of the numerical difference grouped at each time and the overall change trend of the strain data at the corresponding time.

[0083] Step S1451: Extract the numerical difference average value of each time grouping from the spatial strain deviation time series, and extract the average change rate of the sliding window in which the corresponding time grouping is located from the trend analysis result of the distributed optical fiber corrected strain time series.

[0084] In this embodiment, the numerical difference average value of each time grouping is extracted from the spatial strain deviation time series. The numerical difference average value is the average value of all numerical differences in the time grouping. The average change rate of the sliding window in which the corresponding time grouping is located is extracted from the trend analysis result of the distributed optical fiber corrected strain time series. The time stamp matching algorithm is used to ensure that the extracted average change rate is consistent with the time of the time grouping.

[0085] Step S1452: Take the numerical difference average value of each time grouping and the average change rate of the corresponding sliding window as a group of correlation data pairs, arrange all time grouping correlation data pairs, and form a correlation data pair sequence.

[0086] In this embodiment, the numerical difference average value of each time grouping and the average change rate of the corresponding sliding window are taken as a group of correlation data pairs. Each group of correlation data pairs contains a numerical difference average value and an average change rate. All time grouping correlation data pairs are arranged in chronological order to form a correlation data pair sequence.

[0087] Step S1453: Correlation analysis is performed on the sequence of the associated data pairs, and a correlation coefficient between the average value of the numerical difference and the average change rate is calculated. The correlation coefficient has a value range of -1 to 1, and the greater the absolute value of the correlation coefficient, the stronger the correlation between the numerical difference and the overall change trend of the strain data.

[0088] In this embodiment, the correlation analysis is performed on the sequence of the associated data pairs, and the Pearson correlation coefficient algorithm is used to calculate the correlation coefficient between the average value of the numerical difference and the average change rate. The correlation coefficient has a value range of -1 to 1, and the greater the absolute value of the correlation coefficient, the stronger the correlation between the numerical difference and the overall change trend of the strain data. When the correlation coefficient is positive, it indicates that the two are positively correlated. When the correlation coefficient is negative, it indicates that the two are negatively correlated.

[0089] Step S1454: According to the correlation coefficient and in combination with the relative weight value of the distributed optical fiber corresponding to the time grouping, the correlation degree of each time grouping is calculated. The calculation method of the correlation degree is that the correlation coefficient is multiplied by the average value of the relative weight value of the corresponding distributed optical fiber.

[0090] In this embodiment, the average value of the relative weight value of the distributed optical fiber corresponding to the time grouping is first calculated. The average value of the relative weight value is the average value of the relative weight values of all the distributed optical fibers in the time grouping. Then, the correlation coefficient is multiplied by the average value of the relative weight value to obtain the correlation degree of each time grouping. The correlation degree reflects the correlation degree between the numerical difference and the overall change trend of the strain data and the influence of the fiber weight.

[0091] Step S1455: If the correlation degree of any one time grouping is greater than a preset correlation degree threshold, it is determined that the spatial strain deviation of the time grouping and the overall change trend of the strain data have strong correlation.

[0092] In this embodiment, a preset correlation degree threshold is set. The preset value is determined according to the accuracy requirement of the correlation analysis. If the correlation degree of any one time grouping is greater than the preset correlation degree threshold, it is determined that the spatial strain deviation of the time grouping and the overall change trend of the strain data have strong correlation, which indicates that the change trends of the two are consistent.

[0093] Step S1456: If the correlation degree of any one time grouping is less than or equal to the preset correlation degree threshold, the correlation degree state of the corresponding time data needs to be marked in the subsequent generation of the space-time strain evolution characteristics with sensitivity weight. The correlation degree state directly marks the correlation degree value of the time grouping.

[0094] In this embodiment, if the correlation degree of any time grouping is less than or equal to the preset correlation degree threshold, it is determined that the spatial strain deviation of the time grouping is weakly correlated with the overall trend of the strain data, and the correlation degree state of the corresponding time data needs to be marked in the subsequent generation of the space-time strain evolution feature with sensitivity weight. The correlation degree state directly marks the correlation degree value of the time grouping, which is convenient for subsequent analysis.

[0095] Step S1457: record the correlation degree, correlation coefficient, weight value average and correlation degree state of each time grouping to form a correlation analysis result, and associate the correlation analysis result with the corresponding time grouping information, numerical difference and average change rate.

[0096] In this embodiment, the correlation degree, correlation coefficient, weight value average and correlation degree state of each time grouping are recorded to form a correlation analysis result. The correlation analysis result is associated with the corresponding time grouping information, numerical difference and average change rate, and a data merging algorithm is used in the association process to ensure the consistency and integrity of the data.

[0097] Step S146: in combination with the relative weight value of each distributed optical fiber, the spatial strain deviation data in the correlation model of the spatial strain deviation and the strain trend and the overall trend data of the strain data are weighted processed, the numerical difference is multiplied by the relative weight value of the corresponding distributed optical fiber, and the correlation degree of the overall trend of the strain data is multiplied by the relative weight value of the corresponding distributed optical fiber to obtain the weighted spatial strain deviation data and the weighted strain trend correlation degree.

[0098] In this embodiment, in combination with the relative weight value of each distributed optical fiber, the spatial strain deviation data in the correlation model of the spatial strain deviation and the strain trend and the overall trend data of the strain data are weighted processed. The numerical difference is multiplied by the relative weight value of the corresponding distributed optical fiber to obtain the weighted spatial strain deviation data, and the correlation degree of the overall trend of the strain data is multiplied by the relative weight value of the corresponding distributed optical fiber to obtain the weighted strain trend correlation degree. The weighted processing uses a weighted average algorithm to ensure that the data of the distributed optical fiber with a higher weight value has a greater influence on the result.

[0099] Step S147: taking time as the horizontal axis, taking the weighted spatial strain deviation data as the vertical axis, and taking the weighted strain trend correlation degree as the auxiliary vertical axis, a space-time strain evolution feature with sensitivity weight is constructed. Each time node in the space-time strain evolution feature with sensitivity weight simultaneously presents the weighted spatial strain deviation data, the weighted strain trend correlation degree, the corresponding acquisition frequency and the distributed optical fiber layout position information of the time node, and the evolution curves of different key stress regions of the bridge in the space-time strain evolution feature with sensitivity weight are distinguished by different identifiers.

[0100] In this embodiment, the time is taken as the horizontal axis, the weighted spatial strain deviation data is taken as the vertical axis, and the weighted strain change trend correlation degree is taken as the auxiliary vertical axis to construct the time-space strain evolution feature with sensitivity weight. The time-space strain evolution feature is presented in the form of a curve by using a visualization algorithm. Each time node presents the weighted spatial strain deviation data, the weighted strain change trend correlation degree, the corresponding acquisition frequency, and the distributed optical fiber layout position information at the same time. The evolution curves of different bridge key stress areas are distinguished by different colors, line types, or markers, which facilitates intuitive observation of the strain evolution law of different areas.

[0101] Step S150: Based on the time-space strain evolution feature with sensitivity weight, the abnormal evolution section meeting the bridge pre-fracture feature is marked by a multi-fiber cooperative verification mechanism, and a bridge fracture monitoring result containing verification results is generated. The bridge fracture monitoring result containing verification results includes the bridge key stress area corresponding to the abnormal evolution section, the strain deviation change details of the bridge key stress area, and the verification consistency information of the multi-fiber data.

[0102] Step S151: Extract the weighted spatial strain deviation data, the weighted strain change trend correlation degree, the corresponding bridge key stress area, the distributed optical fiber layout position, and the acquisition frequency of all time nodes from the time-space strain evolution feature with sensitivity weight, and arrange them in chronological order to form a target data sequence.

[0103] In this embodiment, the weighted spatial strain deviation data, the weighted strain change trend correlation degree, the corresponding bridge key stress area, the distributed optical fiber layout position, and the acquisition frequency of all time nodes are extracted from the time-space strain evolution feature with sensitivity weight, and arranged in chronological order to form a target data sequence. The target data sequence is organized in time dimension, and each data point contains the weighted spatial strain deviation data, the weighted strain change trend correlation degree, the bridge key stress area, the distributed optical fiber layout position, and the acquisition frequency.

[0104] Step S152: Set the determination standard of the bridge pre-fracture feature, which includes that the weighted spatial strain deviation data exceeds the preset deviation threshold, the weighted strain change trend correlation degree exceeds the preset correlation threshold, and the corresponding state lasts at least two consecutive acquisition time points.

[0105] In this embodiment, the determination standard of the bridge pre-failure feature is set, the preset deviation threshold and the preset correlation threshold are first determined, and the preset values are determined according to the design specification and historical monitoring data of the bridge structure. The determination standard of the bridge pre-failure feature includes that the weighted spatial strain deviation data exceeds the preset deviation threshold, the weighted strain change trend correlation exceeds the preset correlation threshold, and the corresponding state lasts at least two consecutive collection time points. The number of consecutive collection time points is determined according to the development speed of the bridge fracture, so as to ensure that the abnormal evolution section in the pre-failure stage can be accurately identified.

[0106] Step S153: The target data sequence is traversed to identify whether each time node meets the determination standard of the bridge pre-failure feature. If any one time node meets the determination standard of the bridge pre-failure feature, the time node is marked as a suspected abnormal node.

[0107] In this embodiment, the target data sequence is traversed, and for each time node, it is checked whether the determination standard of the bridge pre-failure feature is met. If the weighted spatial strain deviation data exceeds the preset deviation threshold, the weighted strain change trend correlation exceeds the preset correlation threshold, and the corresponding state lasts at least two consecutive collection time points, the time node is marked as a suspected abnormal node. The marking process adopts a conditional judgment algorithm to ensure that the determination result of each time node is accurate.

[0108] Step S154: For the marked suspected abnormal node, a multi-fiber cooperative verification mechanism is started, and the weighted spatial strain deviation data and the weighted strain change trend correlation data of all distributed optical fibers in the key stress area of the bridge corresponding to the suspected abnormal node are extracted.

[0109] In this embodiment, for the marked suspected abnormal node, a multi-fiber cooperative verification mechanism is started. First, the weighted spatial strain deviation data and the weighted strain change trend correlation data of all distributed optical fibers in the key stress area of the bridge corresponding to the suspected abnormal node are extracted, and the extraction process adopts a distributed data query algorithm to ensure the efficiency and accuracy of data extraction.

[0110] Step S155: The consistency proportion of the weighted spatial strain deviation data of all distributed optical fibers in the key stress area of the bridge is calculated.

[0111] Step S1551: From the space-time strain evolution feature data with sensitivity weight, the identification information of all distributed optical fibers in the key stress area of the bridge corresponding to the current suspected abnormal node and the weighted spatial strain deviation data and the weighted strain change trend correlation data corresponding to each distributed optical fiber are extracted.

[0112] In this embodiment, the identification information of all distributed optical fibers in the key stress area of the bridge corresponding to the current suspected abnormal node is extracted from the spatial and temporal strain evolution feature data with sensitivity weight, and the identification information includes the unique number and layout position of the distributed optical fiber. At the same time, the spatial strain deviation data with weight and the strain change trend correlation data with weight corresponding to each distributed optical fiber are extracted, and the data is stored in the form of time stamp to ensure consistency with the suspected abnormal node.

[0113] Step S1552: According to the determination standard of the bridge pre-fracture feature, whether the spatial strain deviation data with weight of each distributed optical fiber exceeds the preset deviation threshold value and whether the strain change trend correlation data with weight exceeds the preset correlation threshold value are checked one by one.

[0114] In this embodiment, according to the determination standard of the bridge pre-fracture feature, whether the spatial strain deviation data with weight of each distributed optical fiber exceeds the preset deviation threshold value and whether the strain change trend correlation data with weight exceeds the preset correlation threshold value are checked one by one. The preset deviation threshold value and the preset correlation threshold value are determined according to the design specification and historical monitoring data of the bridge structure.

[0115] Step S1553: The number of distributed optical fibers satisfying the set condition is counted, which is recorded as the number of fibers satisfying the condition, and the set condition is that the spatial strain deviation data with weight exceeds the preset deviation threshold value and the strain change trend correlation data with weight exceeds the preset correlation threshold value.

[0116] In this embodiment, the number of distributed optical fibers satisfying the set condition is counted, and the set condition is that the spatial strain deviation data with weight exceeds the preset deviation threshold value and the strain change trend correlation data with weight exceeds the preset correlation threshold value. The number of fibers satisfying the condition reflects the number of distributed optical fibers in the key stress area of the bridge that meet the bridge pre-fracture feature.

[0117] Step S1554: The total number of distributed optical fibers participating in data collection in the key stress area of the bridge is counted, and the total number of distributed optical fibers is the number of all distributed optical fibers laid in the key stress area of the bridge, and the distributed optical fibers not participating in the collection at this moment due to failure or data loss are excluded.

[0118] In this embodiment, the total number of distributed optical fibers participating in data collection in the key stress area of the bridge is counted, and the total number of distributed optical fibers is the number of all distributed optical fibers laid in the key stress area of the bridge, and the distributed optical fibers not participating in the collection at this moment due to failure or data loss are excluded to ensure the accuracy of the statistics.

[0119] Step S1555: The ratio of the number of fibers satisfying the condition to the total number is calculated, and the ratio is determined as the consistency proportion of the spatial strain deviation data with weight of the distributed optical fibers in the key stress area of the bridge.

[0120] In this embodiment, the ratio of the number of fibers meeting the condition to the total number is calculated, and the value of the ratio ranges from 0 to 1. The ratio is determined as the consistency ratio of the weighted spatial strain deviation data of the distributed optical fiber in the key stress area of the bridge. The higher the consistency ratio, the better the consistency of the monitoring data of the distributed optical fiber in the key stress area of the bridge.

[0121] Step S1556: If the collection frequency of some distributed optical fibers in the key stress area of the bridge is different from that of other distributed optical fibers, frequency normalization processing is required for the distributed optical fiber data with different collection frequencies when calculating the consistency ratio, so that the judgment criteria of the pre-bridge fracture characteristics of different frequency data are consistent.

[0122] In this embodiment, if the collection frequency of some distributed optical fibers in the key stress area of the bridge is different from that of other distributed optical fibers, frequency normalization processing is required for the distributed optical fiber data with different collection frequencies when calculating the consistency ratio. The frequency normalization processing uses an interpolation algorithm to convert the data with different collection frequencies to the same collection frequency, so that the judgment criteria of the pre-bridge fracture characteristics of different frequency data are consistent, and the accuracy of the consistency ratio calculation is ensured.

[0123] Step S1557: Record the calculated consistency ratio, the number of fibers meeting the condition, the total number, and the frequency normalization processing. The data of the calculated consistency ratio, the number of fibers meeting the condition, the total number, and the frequency normalization processing are used as the result content of the multi-fiber collaborative verification result.

[0124] In this embodiment, the calculated consistency ratio, the number of fibers meeting the condition, the total number, and the frequency normalization processing are recorded. The frequency normalization processing includes whether frequency normalization processing is performed and the processing method. The above data are used as the result content of the multi-fiber collaborative verification result, which is convenient for subsequent analysis and verification.

[0125] Step S156: If the consistency ratio exceeds the preset consistency threshold, it is determined that the suspected abnormal node passes the multi-fiber collaborative verification mechanism verification and is determined as an effective abnormal node. If the consistency ratio does not exceed the preset consistency threshold, the suspected abnormal node is determined as an invalid abnormal node and is excluded from the subsequent abnormal evolution section marking.

[0126] In this embodiment, a preset consistency threshold is set, and the preset value is determined according to the monitoring accuracy requirement of the bridge structure. If the consistency ratio exceeds the preset consistency threshold, it is determined that the suspected abnormal node is verified by the multi-fiber cooperative verification mechanism, and is determined as an effective abnormal node. If the consistency ratio does not exceed the preset consistency threshold, the suspected abnormal node is determined as an invalid abnormal node, which is excluded from the subsequent abnormal evolution section marking. The determination process adopts a conditional judgment algorithm to ensure that the verification result of each suspected abnormal node is accurate.

[0127] Step S157: Mark the time interval composed of continuous effective abnormal nodes as an abnormal evolution section conforming to the pre-bridge fracture feature, and record the starting time, ending time, number of effective abnormal nodes contained, and corresponding bridge key stress area of each abnormal evolution section conforming to the pre-bridge fracture feature.

[0128] For example, step S1571: Sort all effective abnormal nodes verified by the multi-fiber cooperative verification mechanism in the order of collection time to form an ordered effective abnormal node sequence.

[0129] In this embodiment, all effective abnormal nodes verified by the multi-fiber cooperative verification mechanism are sorted in the order of collection time, and the collection time is stored in the form of a time stamp to ensure the accuracy of the sorting. The ordered effective abnormal node sequence is organized in time dimension to facilitate subsequent analysis.

[0130] Step S1572: Traverse the ordered effective abnormal node sequence to check whether the collection time of adjacent two effective abnormal nodes is continuous, and the continuity is that the time interval between the collection time of the latter effective abnormal node and the collection time of the former effective abnormal node is equal to the time interval of the corresponding collection frequency.

[0131] In this embodiment, the ordered effective abnormal node sequence is traversed to check whether the collection time of adjacent two effective abnormal nodes is continuous. The judgment standard for continuity is that the time interval between the collection time of the latter effective abnormal node and the collection time of the former effective abnormal node is equal to the time interval of the corresponding collection frequency, which ensures the continuity of the abnormal evolution section.

[0132] Step S1573: If the collection time of adjacent two effective abnormal nodes is continuous, the two effective abnormal nodes are classified into the same abnormal evolution section candidate group; if the collection time of adjacent two effective abnormal nodes is not continuous, the time interval is greater than the time interval of the collection frequency, the abnormal evolution section candidate group in which the former effective abnormal node is located is determined as a complete abnormal evolution section conforming to the pre-bridge fracture feature, and a new abnormal evolution section candidate group is re-established with the current effective abnormal node as the starting node.

[0133] In this embodiment, if the collection time of the two adjacent effective abnormal nodes is continuous, the two effective abnormal nodes are classified into the same abnormal evolution segment candidate group; if the collection time of the two adjacent effective abnormal nodes is discontinuous, the abnormal evolution segment candidate group in which the previous effective abnormal node is located is determined as a complete abnormal evolution segment meeting the characteristics of the pre-bridge fracture period, and a new abnormal evolution segment candidate group is re-established with the current effective abnormal node as the starting node. The continuity and accuracy of the abnormal evolution segment are ensured.

[0134] Step S1574: In the process of establishing the abnormal evolution segment candidate group, if the number of effective abnormal nodes contained in any one of the abnormal evolution segment candidate groups is less than the preset node number threshold, it is determined that the abnormal evolution segment candidate group is a short-term anomaly, and the abnormal evolution segment candidate group is not marked as an abnormal evolution segment meeting the characteristics of the pre-bridge fracture period; if the number of effective abnormal nodes contained in the abnormal evolution segment candidate group reaches or exceeds the preset node number threshold, the abnormal evolution segment candidate group is marked as an abnormal evolution segment meeting the characteristics of the pre-bridge fracture period.

[0135] In this embodiment, the preset node number threshold is set, and the preset value is determined according to the duration requirement of the pre-bridge fracture period characteristics. If the number of effective abnormal nodes contained in any one of the abnormal evolution segment candidate groups is less than the preset node number threshold, it is determined that the abnormal evolution segment candidate group is a short-term anomaly, and the abnormal evolution segment candidate group is not marked as an abnormal evolution segment meeting the characteristics of the pre-bridge fracture period; if the number of effective abnormal nodes reaches or exceeds the preset node number threshold, the abnormal evolution segment candidate group is marked as an abnormal evolution segment meeting the characteristics of the pre-bridge fracture period.

[0136] Step S1575: The starting time of each abnormal evolution segment meeting the characteristics of the pre-bridge fracture period is determined as the collection time of the first effective abnormal node in the abnormal evolution segment meeting the characteristics of the pre-bridge fracture period, the ending time is determined as the collection time of the last effective abnormal node in the abnormal evolution segment meeting the characteristics of the pre-bridge fracture period, and the number of effective abnormal nodes contained in each abnormal evolution segment meeting the characteristics of the pre-bridge fracture period is counted.

[0137] In this embodiment, the starting time of each abnormal evolution segment meeting the characteristics of the pre-bridge fracture period is determined as the collection time of the first effective abnormal node in the abnormal evolution segment, and the ending time is determined as the collection time of the last effective abnormal node in the abnormal evolution segment. The number of effective abnormal nodes contained in each abnormal evolution segment is counted, and the number reflects the duration of the abnormal evolution segment.

[0138] Step S1576: The information of the key stress area of the bridge corresponding to each abnormal evolution segment meeting the characteristics of the pre-bridge fracture period is extracted, and the information includes the name of the key stress area of the bridge, the number of distributed optical fibers participating in verification in the key stress area of the bridge, and the distribution of the layout position.

[0139] In this embodiment, the information of the key stress area of the bridge corresponding to each abnormal evolution section meeting the characteristics of the early stage of bridge fracture is extracted, and the information includes the name of the key stress area of the bridge, the number of distributed optical fibers participating in verification, and the distribution of the layout position. The distribution of the layout position is stored in the form of coordinates, which facilitates intuitive understanding of the spatial position of the abnormal evolution section.

[0140] Step S1577: The start time, end time, number of effective abnormal nodes, corresponding bridge key stress area information, and optical fiber distribution information of each abnormal evolution section meeting the characteristics of the early stage of bridge fracture are associated and stored to form an abnormal evolution section basic information table.

[0141] In this embodiment, the start time, end time, number of effective abnormal nodes, corresponding bridge key stress area information, and optical fiber distribution information of each abnormal evolution section meeting the characteristics of the early stage of bridge fracture are associated and stored, and a data merging algorithm is used in the association process to ensure the consistency and integrity of the data. The abnormal evolution section basic information table formed is organized in the dimension of the key stress area of the bridge, which facilitates subsequent analysis and early warning.

[0142] Step S158: The weighted spatial strain deviation data change details, weighted strain change trend correlation degree change details, corresponding distributed optical fiber layout position, and collection frequency of all effective abnormal nodes in each abnormal evolution section meeting the characteristics of the early stage of bridge fracture are extracted, and the multi-fiber collaborative verification result of the abnormal evolution section meeting the characteristics of the early stage of bridge fracture is recorded, which includes the consistency ratio and the number of optical fibers participating in verification.

[0143] In this embodiment, the weighted spatial strain deviation data change details, weighted strain change trend correlation degree change details, corresponding distributed optical fiber layout position, and collection frequency of all effective abnormal nodes in each abnormal evolution section meeting the characteristics of the early stage of bridge fracture are extracted, and the multi-fiber collaborative verification result of the abnormal evolution section meeting the characteristics of the early stage of bridge fracture is recorded, which includes the consistency ratio and the number of optical fibers participating in verification.

[0144] Step S159: The information of all abnormal evolution sections meeting the characteristics of the early stage of bridge fracture is summarized to form a bridge fracture monitoring result containing verification results, and each abnormal evolution section meeting the characteristics of the early stage of bridge fracture corresponds to an independent entry in the bridge fracture monitoring result containing verification results. Each entry includes the time interval of the abnormal evolution section meeting the characteristics of the early stage of bridge fracture, the corresponding bridge key stress area, the strain deviation change details, the correlation degree change details, the collection frequency information, and the multi-fiber collaborative verification result.

[0145] In this embodiment, the information of all abnormal evolution sections meeting the characteristics of the early stage of bridge fracture is aggregated, and a data aggregation algorithm is used to aggregate all the information into a bridge fracture monitoring result containing verification results. Each abnormal evolution section meeting the characteristics of the early stage of bridge fracture in the monitoring result corresponds to an independent entry, and each entry contains a time interval, a corresponding key stress region of the bridge, strain deviation change details, correlation degree change details, acquisition frequency information, and multi-fiber cooperative verification results. The monitoring result is stored in a structured data format, facilitating subsequent analysis and early warning.

[0146] Based on the same inventive concept, please refer to Figure 2 , which shows a structural schematic block diagram of a distributed optical fiber-based bridge fracture monitoring system 100 provided by an embodiment of the present application for executing the above-mentioned distributed optical fiber-based bridge fracture monitoring method. The distributed optical fiber-based bridge fracture monitoring system 100 can include a communication unit 110, a machine-readable storage medium 120, and a processor 130.

[0147] In this embodiment, the machine-readable storage medium 120 and the processor 130 are both located in the distributed optical fiber-based bridge fracture monitoring system 100 and are separately arranged. However, it should be understood that the machine-readable storage medium 120 can also be independent of the distributed optical fiber-based bridge fracture monitoring system 100 and can be accessed by the processor 130 through a bus interface. Alternatively, the machine-readable storage medium 120 can also be integrated into the processor 130 and can communicate and interact with external systems through the communication unit 110.

[0148] The processor 130 is the control center of the distributed optical fiber-based bridge fracture monitoring system 100, connects all parts of the distributed optical fiber-based bridge fracture monitoring system 100 through various interfaces and lines, executes the software programs and / or modules stored in the machine-readable storage medium 120, and calls the data stored in the machine-readable storage medium 120, executes various functions of the distributed optical fiber-based bridge fracture monitoring system 100 and processes data, thereby overall monitoring the distributed optical fiber-based bridge fracture monitoring system 100. Optionally, the processor 130 can include one or more processing cores; for example, the processor 130 can integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, user interface, and application program, etc., and the modem processor mainly processes wireless communication. It can be understood that the above-mentioned modem processor can also not be integrated into the processor. Among them, the machine-readable storage medium 120 is used to store machine-executable instructions for executing the scheme of the present application, and the processor 130 is used to execute the machine-executable instructions stored in the machine-readable storage medium 120 to realize the distributed optical fiber-based bridge fracture monitoring method provided by the foregoing method embodiment.

[0149] It should be noted that the foregoing description of embodiments of the application has been presented for the purposes of illustration and description. It is not intended to be exhaustive or to limit the application to the precise form described, and many modifications, variations, and alternatives are possible.

Claims

1. A bridge fracture monitoring method based on distributed optical fiber, characterized in that, The method includes: A stress response correlation mapping table between distributed optical fibers and bridge structures is constructed. The table records the correspondence between the deployment location of distributed optical fibers and the key stress-bearing areas of the bridge, the sensing range parameters of distributed optical fibers at each deployment location, and the response sensitivity curves of distributed optical fibers under different stresses. The sensing range parameters include the spatial coverage range, strain measurement range, and measurement accuracy of the optical fiber. Based on the stress response correlation mapping table between the distributed optical fiber and the bridge structure, the acquisition frequency of the distributed optical fiber is dynamically adjusted for different usage states of the bridge. The distributed optical fiber strain data and synchronous environmental influencing factor data under the corresponding states are collected to generate a strain and environment correlation dataset with acquisition frequency identification. The environmental influencing factor data includes the rate of change of ambient temperature and wind speed in the wind data. Environmental interference removal processing is performed on the strain and environment-related dataset with acquisition frequency identifier. Based on the environmental impact factor data, the strain data of different distributed optical fibers in the key stress area of ​​the same bridge are corrected. The differences in the corrected strain data are extracted to form a spatial strain deviation feature set after environmental correction. The spatial strain deviation feature set after environmental correction is correlated with the strain data change trend in the time dimension. Combined with the response sensitivity curve in the stress response correlation mapping table between the distributed optical fiber and the bridge structure, a spatiotemporal strain evolution feature with sensitivity weight is generated. The spatiotemporal strain evolution feature with sensitivity weight presents the change law of strain deviation in different key stress areas of the bridge over time and the corresponding response weight of the optical fiber. Based on the spatiotemporal strain evolution characteristics with sensitivity weights, anomaly evolution segments that conform to the early-stage characteristics of bridge fracture are marked through a multi-fiber collaborative verification mechanism, generating bridge fracture monitoring results containing verification results. The bridge fracture monitoring results containing verification results include the key stress area of ​​the bridge corresponding to the anomaly evolution segment, details of strain deviation changes in the key stress area of ​​the bridge, and verification consistency information of multi-fiber data.

2. The bridge fracture monitoring method based on distributed optical fiber according to claim 1, characterized in that, The construction of the stress response correlation mapping table between distributed optical fibers and bridge structures includes: Obtain the structural design documents and material mechanical parameters of the bridge, and extract the key stress area information of the bridge based on the structural design documents. The material mechanical parameters include the elastic modulus, Poisson's ratio and tensile strength parameters of the main structural material of the bridge. The key stress area information of the bridge includes the main beam support point area, the mid-span stress area, the pier connection area and the bearing stress area of ​​the bridge. A three-dimensional stress distribution simulation analysis was performed on each critical stress area of ​​the bridge. The stress field model of the critical stress area was constructed using the finite element analysis method. The stress concentration sub-regions within the critical stress area of ​​each bridge were determined by the stress field model of the critical stress area. The stress concentration sub-regions within the critical stress area of ​​each bridge are local continuous areas where the stress value distribution within the critical stress area of ​​the bridge is higher than that of the surrounding area. For each stress concentration sub-region, the deployment path direction of the distributed optical fiber in the stress concentration sub-region is determined based on the stress distribution gradient output by the stress field model of the key stress region. The deployment path direction of the distributed optical fiber in the stress concentration sub-region must be consistent with the direction of maximum change of stress distribution gradient, and the deployment path must cover the entire spatial range of the stress concentration sub-region. Multiple test points are selected along the deployment path, and stress response tests are performed on the distributed optical fiber at each test point. By applying standard stresses of different gradients, the strain output value of the distributed optical fiber under each standard stress is recorded. Based on the correspondence between the strain output value and the standard stress, the response sensitivity curve of the distributed optical fiber at each test point is fitted and generated. The average curve of the response sensitivity curves of all test points on each deployment path is calculated, and the average curve is used as the distributed optical fiber response sensitivity curve corresponding to that deployment path. At the same time, the deployment position, deployment path direction, sensing range parameters and response sensitivity curve of the distributed optical fiber corresponding to each key stress area of ​​the bridge are recorded to form a stress response correlation mapping table between distributed optical fiber and bridge structure. In the stress response correlation mapping table between distributed optical fiber and bridge structure, each key stress area of ​​the bridge is associated with at least three deployment positions of distributed optical fibers, and the deployment positions of the three distributed optical fibers are located in different stress gradient regions of the stress concentration sub-region, so as to realize the acquisition of multi-gradient strain data within the same key stress area of ​​the bridge.

3. The bridge fracture monitoring method based on distributed optical fiber according to claim 1, characterized in that, The stress response correlation mapping table based on the distributed optical fiber and the bridge structure dynamically adjusts the acquisition frequency of the distributed optical fiber for different bridge usage states, collects distributed optical fiber strain data and synchronous environmental influencing factor data under the corresponding states, and generates a strain and environment correlation dataset with acquisition frequency identifiers, including: The deployment location of each distributed optical fiber, the corresponding key stress area of ​​the bridge, and the response sensitivity curve are extracted from the stress response correlation mapping table between the distributed optical fiber and the bridge structure. The stress change characteristics of the key stress area under different bridge usage conditions are analyzed. The different bridge usage conditions include bridge unloaded state, bridge fully loaded state, bridge bearing temporary load state, bridge encountering wind load state in environmental load, and bridge encountering temperature load state in environmental load. For each bridge usage condition, the reference value of the strain data acquisition frequency is determined according to the stress change characteristics of the corresponding key stress area of ​​the bridge. The stress change is gradual under the unloaded state of the bridge, so the first type of acquisition frequency reference value is set. The stress change is drastic under the full load state of the bridge and under the temporary load state of the bridge, so the second type of acquisition frequency reference value is set. The time interval of the first type of acquisition frequency reference value is longer than the time interval of the second type of acquisition frequency reference value. Establish adjustment rules for environmental impact factors and collection frequency. When the rate of change of ambient temperature in the environmental impact factor data exceeds the preset rate of change, or the wind speed in the wind data exceeds the preset wind speed, the collection frequency of the corresponding distributed optical fiber is increased based on the adjustment rules for environmental impact factors and collection frequency. Specifically, the increase is determined according to the difference between the rate of change of ambient temperature and the preset rate of change, and the difference between the wind speed and the preset wind speed. The larger the difference, the greater the increase in collection frequency. When the bridge is in any usage state and the acquisition triggering conditions are met, the distributed optical fiber at the corresponding deployment location is triggered to acquire strain data according to the adjusted acquisition frequency. During the acquisition process, the acquisition time, acquisition frequency and the deployment location of the corresponding distributed optical fiber are recorded for each strain data acquisition, forming a distributed optical fiber strain sequence with acquisition frequency identification in this usage state. While collecting distributed optical fiber strain sequences with acquisition frequency identifiers, environmental impact factor data at corresponding times are collected by environmental monitoring equipment deployed around the bridge. The environmental impact factor data includes environmental temperature data, environmental humidity data, wind force data, and precipitation data. The acquisition time of the environmental impact factor data is synchronized with the acquisition time of the strain data. Specifically, this is achieved by associating the time base of the environmental monitoring equipment with the distributed optical fiber acquisition system. The distributed optical fiber strain sequence with acquisition frequency identifier under the same usage state is associated and bound with the environmental influencing factor data at the corresponding time to generate a strain-environment association subset with acquisition frequency identifier under the usage state. The strain-environment correlation subsets with acquisition frequency identifiers for all bridge usage states are aggregated to form a complete strain-environment correlation dataset with acquisition frequency identifiers. Each data entry in the strain-environment correlation dataset with acquisition frequency identifiers includes the location of the distributed optical fiber deployment, strain data, acquisition time, acquisition frequency, environmental influencing factor data, and the corresponding bridge usage state identifier.

4. The bridge fracture monitoring method based on distributed optical fiber according to claim 1, characterized in that, The process involves removing environmental interference from the strain-environment correlation dataset with the acquisition frequency identifier, correcting the strain data of different distributed optical fibers within the same bridge's critical stress area based on environmental influencing factor data, extracting the differences in the corrected strain data, and forming a spatial strain deviation feature set after environmental correction, including: From the strain and environment correlation dataset with acquisition frequency identifier, all distributed optical fiber strain data and associated environmental influencing factor data corresponding to the same key stress area of ​​the bridge are selected. The selection criteria are the correspondence between the key stress area of ​​the bridge and the location of the distributed optical fiber in the stress response correlation mapping table between the distributed optical fiber and the bridge structure. For each set of strain data and environmental influencing factor data in the selected key stress area of ​​the same bridge, an interference correlation model between environmental influencing factors and strain data is established. By analyzing the correlation between historical environmental influencing factor data and strain data, the interference coefficients of environmental temperature, environmental humidity, and wind force on strain data are determined. Based on the interference correlation model and interference coefficient between the environmental influencing factors and strain data, environmental interference stripping is performed on each set of strain data. The strain interference value caused by the environmental influencing factors is calculated. The strain interference value is subtracted from the original strain data to obtain the environmentally corrected strain data. The environmentally corrected strain data only reflects the strain generated by the stress on the bridge structure. All distributed fiber strain data after environmental correction within the critical stress area of ​​the same bridge are grouped according to the same acquisition time to obtain multiple time groups. Each time group contains the environmental correction strain data of all distributed fibers within the critical stress area of ​​the bridge at that time. Calculate the numerical difference between the environmentally corrected strain data of any two distributed optical fibers within each time group, and record all numerical differences and the corresponding deployment locations and acquisition frequencies of the two distributed optical fibers. The distribution characteristics of numerical differences within each time group are statistically analyzed, and the maximum, minimum, average, and variance of numerical differences within each time group are determined. The statistical results of the maximum, minimum, average, and variance of numerical differences are then correlated with the environmental impact factor data and collection frequency corresponding to each time group. The statistical results of the numerical differences, corresponding distributed optical fiber deployment locations, acquisition frequencies, maximum, minimum, average and variance of the numerical differences, and related environmental influencing factors are summarized for all time groups to form an environmentally corrected spatial strain deviation feature set. Each entry in the environmentally corrected spatial strain deviation feature set corresponds to the spatial strain deviation related information of a time group, and each entry contains an identifier for environmental interference stripping processing.

5. The bridge fracture monitoring method based on distributed optical fiber according to claim 1, characterized in that, The process involves modeling the correlation between the environmentally corrected spatial strain deviation feature set and the temporal strain data change trend, and combining this with the response sensitivity curve in the stress response correlation mapping table between the distributed optical fiber and the bridge structure to generate spatiotemporal strain evolution features with sensitivity weights, including: Extract the time sequence of environmentally corrected strain data for each distributed optical fiber from the strain and environment correlation dataset with acquisition frequency identifier, and arrange them in chronological order of acquisition time to form a distributed optical fiber corrected strain time series. The distributed optical fiber corrected strain time series is subjected to trend extraction processing. The average rate of change of strain data in each time window is calculated using the sliding window method. The overall trend of strain data is determined based on the average rate of change. The overall trend of strain data includes strain increasing trend, strain decreasing trend and strain stabilization trend. The statistical results of the numerical differences, maximum values, minimum values, average values ​​and variances of each time group and the associated acquisition frequencies are extracted from the environmentally corrected spatial strain deviation feature set. The results are then arranged in chronological order of acquisition time to form a spatial strain deviation time series. The response sensitivity curve of each distributed optical fiber is extracted from the stress response correlation mapping table between the distributed optical fiber and the bridge structure. The sensitivity coefficient of each distributed optical fiber in the corresponding strain range is determined. The sensitivity coefficients of all distributed optical fibers are normalized to obtain the relative weight value of each distributed optical fiber. The relative weight value is dimensionless, and the higher the sensitivity coefficient, the larger the relative weight value. A correlation model between spatial strain deviation and strain change trend is established, and the numerical differences in the spatial strain deviation time series are correlated with the overall change trend of strain data in the distributed optical fiber corrected strain time series. The correlation degree between the numerical differences of each time group and the overall change trend of the strain data at the corresponding time is calculated. By combining the relative weight value of each distributed optical fiber, the spatial strain deviation data and the overall trend data of strain data in the correlation model between spatial strain deviation and strain change trend are weighted. The numerical difference is multiplied by the relative weight value of the corresponding distributed optical fiber, and the correlation degree of the overall trend of strain data is multiplied by the relative weight value of the corresponding distributed optical fiber, so as to obtain the weighted spatial strain deviation data and the weighted correlation degree of strain change trend. Using time as the horizontal axis, weighted spatial strain deviation data as the vertical axis, and weighted strain change trend correlation as the auxiliary vertical axis, a spatiotemporal strain evolution feature with sensitivity weights is constructed. In the spatiotemporal strain evolution feature with sensitivity weights, each time node simultaneously presents the weighted spatial strain deviation data, weighted strain change trend correlation, corresponding acquisition frequency, and distributed optical fiber deployment location information for that time node. Furthermore, the evolution curves of different key stress areas of the bridge in the spatiotemporal strain evolution feature with sensitivity weights are distinguished by different identifiers.

6. The bridge fracture monitoring method based on distributed optical fiber according to claim 5, characterized in that, The process of trend extraction processing for the distributed optical fiber corrected strain time series, employing the sliding window method to calculate the average rate of change of strain data within each time window, and determining the overall trend of strain data based on the average rate of change, includes: The time length of the sliding window is determined based on the acquisition frequency and data length of the distributed optical fiber corrected strain time series, and the time length of the sliding window needs to cover at least three consecutive acquisition times. Starting from the initial acquisition time of the distributed fiber corrected strain time series, the window is slid by a preset window movement step size, where the window movement step size corresponds to an acquisition time, and each window contains all distributed fiber corrected strain data within the corresponding time period. For the distributed fiber-corrected strain data within each sliding window, the difference between the strain data at the first and last acquisition times within the window is calculated. The difference is then divided by the time interval within the window, which is the time difference between the last and first acquisition times within the window, to obtain the average rate of change of the strain data within the sliding window. Record the start time, end time, and corresponding average rate of change for each sliding window to form an average rate of change sequence, which is arranged in the order of the sliding windows. Analyze the numerical distribution of the average rate of change sequence. If the average rate of change exceeding a preset proportion in the average rate of change sequence is positive and the subsequent average rate of change is greater than the previous average rate of change, then the overall trend of strain data in the distributed optical fiber corrected strain time series is determined to be an increasing strain trend. If the average rate of change exceeding a preset proportion in the average rate of change sequence is negative and the absolute value of the next average rate of change is greater than the absolute value of the previous average rate of change, then the overall trend of the strain data of the distributed optical fiber corrected strain time series is determined to be a strain decreasing trend. If the absolute value of the average rate of change exceeding a preset proportion in the average rate of change sequence is less than a preset threshold and the value changes within a preset fluctuation range, the preset fluctuation range is determined based on the historical fluctuation range of strain data under normal bridge use conditions, then the overall trend of strain data in the distributed optical fiber corrected strain time series is determined to be a strain stable trend. The average rate of change of each sliding window, the corresponding window time period, and the overall trend of strain data are correlated and stored to form the trend analysis results of the distributed fiber-optic corrected strain time series.

7. The bridge fracture monitoring method based on distributed optical fiber according to claim 5, characterized in that, The establishment of a correlation model between spatial strain deviation and strain change trend involves correlating the numerical differences in the spatial strain deviation time series with the overall change trend of strain data in the distributed fiber optic corrected strain time series, and calculating the correlation degree between the numerical differences of each time-group and the overall change trend of the strain data at the corresponding time, including: The average numerical difference of each time group is extracted from the spatial strain deviation time series, and the average rate of change of the sliding window corresponding to the time group is extracted from the trend analysis results of the distributed optical fiber corrected strain time series. The average difference of the numerical values ​​of each time group and the average rate of change of the corresponding sliding window are taken as a set of associated data pairs. All the associated data pairs of time groups are sorted out to form a sequence of associated data pairs. Correlation analysis is performed on the associated data series to calculate the correlation coefficient between the average numerical difference and the average rate of change. The correlation coefficient ranges from -1 to 1. The larger the absolute value of the correlation coefficient, the stronger the correlation between the numerical difference and the overall trend of change of the strain data. Based on the correlation coefficient and the relative weight value of the distributed optical fiber corresponding to the group at that time, the correlation degree of each group at that time is calculated. The correlation degree is calculated by multiplying the correlation coefficient by the average value of the relative weight value of the corresponding distributed optical fiber. If the correlation of any time group is greater than the preset correlation threshold, it is determined that the spatial strain deviation of that time group is strongly correlated with the overall trend of strain data. If the correlation degree of any time group is less than or equal to the preset correlation degree threshold, the correlation degree status of the corresponding time data needs to be marked in the subsequent generation of spatiotemporal strain evolution features with sensitivity weight. The correlation degree status directly marks the correlation degree value of the time group. Record the correlation degree, correlation coefficient, average weight value, and correlation state of each time group to form the correlation analysis results. Then, correlate the correlation analysis results with the corresponding time group information, numerical differences, and average rate of change.

8. The bridge fracture monitoring method based on distributed optical fiber according to claim 1, characterized in that, The spatiotemporal strain evolution characteristics based on the sensitivity weights are used to mark abnormal evolution segments that conform to the early-stage characteristics of bridge fracture through a multi-fiber collaborative verification mechanism, generating bridge fracture monitoring results containing verification results, including: From the spatiotemporal strain evolution characteristics with sensitivity weights, extract the weighted spatial strain deviation data, weighted strain change trend correlation, corresponding key stress areas of the bridge, distributed optical fiber deployment locations and acquisition frequencies for all time nodes, and organize them in chronological order to form a target data sequence. The criteria for determining the early characteristics of bridge fracture are set, which include weighted spatial strain deviation data exceeding a preset deviation threshold, weighted strain change trend correlation exceeding a preset correlation threshold, and the corresponding state lasting for at least two consecutive acquisition times. Traverse the target data sequence and identify whether each time node meets the criteria for determining the early characteristics of bridge fracture. If any time node meets the criteria for determining the early characteristics of bridge fracture, then mark the time node as a suspected abnormal node. For marked suspected abnormal nodes, a multi-fiber collaborative verification mechanism is initiated to extract weighted spatial strain deviation data and weighted strain change trend correlation data of all distributed optical fibers in the critical stress area of ​​the bridge corresponding to the suspected abnormal node. Calculate the consistency ratio of weighted spatial strain deviation data of all distributed optical fibers in the critical stress area of ​​the bridge; If the consistency ratio exceeds the preset consistency threshold, the suspected abnormal node is determined to be a valid abnormal node after being verified by the multi-fiber collaborative verification mechanism; if the consistency ratio does not exceed the preset consistency threshold, the suspected abnormal node is determined to be an invalid abnormal node and is excluded from subsequent abnormal evolution segment marking. The time intervals consisting of consecutive valid anomaly nodes are marked as anomaly evolution segments that conform to the characteristics of the early stage of bridge fracture. The start time, end time, number of valid anomaly nodes, and corresponding key stress areas of the bridge are recorded for each anomaly evolution segment that conforms to the characteristics of the early stage of bridge fracture. Extract the weighted spatial strain deviation data change details, weighted strain change trend correlation details, corresponding distributed optical fiber deployment location and acquisition frequency of all effective abnormal nodes in each abnormal evolution segment that conforms to the characteristics of the early stage of bridge fracture. At the same time, record the multi-fiber collaborative verification results of the abnormal evolution segment that conforms to the characteristics of the early stage of bridge fracture. The multi-fiber collaborative verification results include the consistency ratio and the number of optical fibers participating in the verification. Information on all abnormal evolution segments that conform to the early characteristics of bridge fracture is summarized to form bridge fracture monitoring results including verification results. Each abnormal evolution segment that conforms to the early characteristics of bridge fracture corresponds to an independent entry in the bridge fracture monitoring results including the time interval of the abnormal evolution segment that conforms to the early characteristics of bridge fracture, the corresponding key stress area of ​​the bridge, details of strain deviation changes, details of correlation changes, acquisition frequency information, and multi-fiber collaborative verification results.

9. The bridge fracture monitoring method based on distributed optical fiber according to claim 8, characterized in that, The calculation of the consistency ratio of weighted spatial strain deviation data of all distributed optical fibers in the critical stress area of ​​the bridge includes: From the spatiotemporal strain evolution feature data with sensitivity weights, extract the identification information of all distributed optical fibers in the key stress area of ​​the bridge corresponding to the current suspected abnormal node, as well as the weighted spatial strain deviation data and weighted strain change trend correlation data corresponding to each distributed optical fiber. Based on the criteria for determining the early characteristics of bridge fracture, check one by one whether the weighted spatial strain deviation data of each distributed optical fiber exceeds the preset deviation threshold and whether the weighted strain change trend correlation exceeds the preset correlation threshold. The number of distributed optical fibers that meet the set conditions is recorded as the number of optical fibers that meet the conditions. The set conditions are that the weighted spatial strain deviation data exceeds a preset deviation threshold and the weighted strain change trend correlation exceeds a preset correlation threshold. The total number of distributed optical fibers participating in data acquisition within the critical stress area of ​​the bridge is counted. The total number of distributed optical fibers refers to all distributed optical fibers deployed within the critical stress area of ​​the bridge, excluding distributed optical fibers that did not participate in the acquisition at that moment due to faults or missing data. Calculate the ratio of the number of optical fibers that meet the conditions to the total number, and determine this ratio as the consistency ratio of the weighted spatial strain deviation data of distributed optical fibers in the critical stress area of ​​the bridge. If the sampling frequency of some distributed optical fibers in the critical stress area of ​​the bridge is different from that of other distributed optical fibers, frequency normalization processing is required for the distributed optical fiber data with different sampling frequencies when calculating the consistency ratio, so that the judgment criteria for the early characteristics of bridge fracture are consistent for different frequency data. Record the calculated consistency ratio, the number of fibers meeting the conditions, the total number of fibers, and the frequency normalization process. Use the calculated consistency ratio, the number of fibers meeting the conditions, the total number of fibers, and the frequency normalization process as the results of the multi-fiber collaborative verification.

10. A bridge fracture monitoring system based on distributed optical fiber, characterized in that, include: processor; A machine-readable storage medium for storing machine-executable instructions of the processor; The processor is configured to execute the bridge fracture monitoring method based on any one of claims 1 to 9 by executing the machine-executable instructions.

Images