A method, apparatus and system for inverting the stress state of long-span bridges

Abstract

This application relates to the field of bridge stress state detection technology, specifically to a method, device, and system for inverting the stress state of long-span bridges. The method includes: real-time acquisition of stress, vibration, and temperature data at various monitoring points on the bridge deck; dividing the entire data acquisition time into multiple detection periods; obtaining the vehicle load influence coefficient for each monitoring point in each detection period based on the abnormal characteristics of the stress and vibration data at each monitoring point, as well as the degree of difference between the abnormal stress and vibration characteristics; and combining this with the waveform similarity of all vehicle load influence coefficients and temperature data for each monitoring point on that day to obtain the optimized step size factor for each monitoring point on that day, thereby filtering the stress and vibration data of each monitoring point on that day. This application improves the accuracy of bridge stress state assessment by adaptively obtaining the optimized step size factor for each monitoring point on that day.

Description

Technical Field

[0001] This application relates to the field of bridge stress state detection technology, specifically to a method, device, and system for inverting the stress state of long-span bridges. Background Technology

[0002] Long-span bridges are crucial components of modern transportation infrastructure, and their safe operation is of paramount importance. During long-term service, bridge structures inevitably suffer from environmental erosion, material aging, fatigue damage, and the effects of increasing traffic loads and extreme weather events, leading to a gradual degradation of their load-bearing capacity and safety performance. Therefore, inverse analysis of the stress state of bridges is an important measure for preventative maintenance.

[0003] Traditional bridge inspection methods, such as periodic load tests, are not only time-consuming, labor-intensive, and costly, but also struggle to achieve real-time, continuous condition monitoring, often resulting in delayed responses to sudden damage or rapid performance deterioration. Currently, real-time data analysis techniques are commonly used for monitoring the stress state of bridges. However, in actual inspection processes, due to the uncertainty of dynamic loads on bridges and the influence of temperature changes on the stress state, it is difficult to accurately obtain more precise real-time monitoring data, thus affecting the accurate assessment of the internal stress conditions of the bridge and leading to inaccurate stress state inversion results for long-span bridges. Summary of the Invention

[0004] To address the aforementioned technical problems, the purpose of this application is to provide a method, apparatus, and system for inverting the stress state of long-span bridges. The specific technical solution adopted is as follows:

[0005] In a first aspect, embodiments of this application provide a method for inverting the stress state of a long-span bridge, the method comprising the following steps:

[0006] Real-time acquisition of stress, vibration, and temperature data at various monitoring points on the bridge deck;

[0007] The entire data acquisition time is divided into multiple detection periods; based on the steepness of all peaks in the stress data of each monitoring point in each detection period, and the difference between the stress data fluctuation amplitude of each monitoring point and its adjacent monitoring points in each detection period, the first outlier value of each monitoring point in each detection period is obtained.

[0008] Based on the dispersion and high-frequency characteristics of the vibration data of each monitoring point in each detection period, and the differences between the frequencies of all modal components of the vibration data of each monitoring point and its neighboring monitoring points in each detection period, the second outlier of each monitoring point in each detection period is obtained; based on the degree of difference between the first outlier and the second outlier of each monitoring point in each detection period and its neighboring period, the asynchronous coefficient of each monitoring point in each detection period is obtained; and combined with the first outlier and the second outlier of each monitoring point in each detection period, the vehicle load influence coefficient of each monitoring point in each detection period is obtained.

[0009] Based on the waveform similarity of all vehicle load influence coefficients and temperature data at each monitoring point on that day, as well as the average level of all vehicle load influence coefficients, the comprehensive influence coefficient of each monitoring point on that day is obtained. Then, the optimization step size factor of each monitoring point on that day is obtained, and the stress and vibration data of each monitoring point on that day are filtered.

[0010] Preferably, the method for obtaining the first outlier of each monitoring point in each detection period is as follows: calculate the sum of the kurtosis of the peaks where all maximum points are located in the fitted curve of the stress data of each monitoring point in each detection period; calculate the mean of the absolute difference of the fluctuation significance of each monitoring point and its two adjacent monitoring points in each detection period; and record the product of the sum and the mean as the first outlier of each monitoring point in each detection period; wherein, the fluctuation significance of each monitoring point in each detection period refers to the mean of all maximum values ​​in the fitted curve of the stress data of each monitoring point in each detection period.

[0011] Preferably, the formula for calculating the second outlier value of each monitoring point in each detection period is as follows: In the formula, This represents the second outlier at the i-th monitoring point during the j-th monitoring period. The range of vibration data at the i-th monitoring point during the j-th monitoring period is given by [the range of vibration data at the i-th monitoring point during the j-th monitoring period]. Let be the significant value of high-frequency vibration at the i-th monitoring point during the j-th monitoring period. Let be the vibration mode difference coefficient of the i-th monitoring point during the j-th monitoring period; where, The method for obtaining the data is as follows: decompose the vibration data of the i-th monitoring point in the j-th detection period into multiple modal components, and obtain the spectrum sequence of each modal component. The maximum value of the frequency corresponding to the centroid of the spectrum sequence is recorded as the high-frequency vibration significance value of the i-th monitoring point in the j-th detection period. The method for obtaining the frequency of the spectral centroid corresponding to all modal components of the vibration data of each monitoring point in the j-th detection period is as follows: the frequencies of all modal components of the vibration data of each monitoring point in the j-th detection period are arranged in ascending order to obtain the spectral centroid frequency sequence of each monitoring point in the j-th detection period. The mean value of the Euclidean distance between the spectral centroid frequency sequences of the i-th monitoring point and its two adjacent monitoring points in the j-th detection period is taken as the vibration modal difference coefficient of the i-th monitoring point in the j-th detection period.

[0012] Preferably, the asynchronous coefficient of each monitoring point in each detection period refers to the DTW distance between all first outliers and all second outliers of each monitoring point in each detection period and its nearest neighboring period.

[0013] Preferably, the vehicle load influence coefficient of each monitoring point in each detection period refers to the ratio of the average of the first and second outliers of each monitoring point in each detection period to the asynchronous coefficient.

[0014] Preferably, the method for obtaining the comprehensive influence coefficient of each monitoring point on the same day is as follows: the vehicle load influence coefficients of each monitoring point during all detection periods on the same day are arranged in ascending order of time as the first sequence of each monitoring point on the same day; the average temperature of each monitoring point during all detection periods on the same day is arranged in ascending order of time as the second sequence of each monitoring point on the same day; the SBD distance between the first sequence and the second sequence of each monitoring point on the same day is taken as the coupling correlation coefficient of each monitoring point on the same day; and the ratio of the mean of the vehicle load influence coefficients of each monitoring point during all detection periods on the same day to the coupling correlation coefficient is recorded as the comprehensive influence coefficient of each monitoring point on the same day.

[0015] Preferably, the formula for calculating the optimization step size factor for each monitoring point on that day is: In the formula, Let i be the optimization step size factor for the i-th monitoring point on that day. and These represent the preset minimum and maximum values ​​of the step size factor, respectively. Let be the comprehensive impact coefficient of the i-th monitoring point on that day. It is the hyperbolic tangent function.

[0016] Preferably, during the process of filtering the stress and vibration data of each monitoring point on the same day, the step size factor of the LMS filtering algorithm is set to the optimized step size factor of each monitoring point on the same day.

[0017] Secondly, embodiments of this application provide a stress state inversion device for long-span bridges, the stress state inversion device comprising:

[0018] The data acquisition module is used to acquire stress, vibration and temperature data at various monitoring points on the bridge deck in real time.

[0019] The load influence analysis module is used to obtain the vehicle load influence coefficient of each monitoring point in each detection period based on the anomaly characteristics of stress data and vibration data at each monitoring point, as well as the asynchronous characteristics between stress anomalies and vibration anomalies.

[0020] The stress state inversion module is used to obtain the comprehensive influence coefficient of each monitoring point on the day based on the waveform similarity of the vehicle load influence and temperature data of each monitoring point on the day, as well as the average level of the vehicle load influence coefficient of each monitoring point on the day. Then, the step size factor in the filtering algorithm is optimized to obtain the filtered monitoring data of each monitoring point on the day, and then the stress state of each monitoring point on the day is inverted.

[0021] Thirdly, this application also provides a system for inverting the stress state of a long-span bridge. The system includes a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of any of the above-described methods for inverting the stress state of a long-span bridge.

[0022] As can be seen from the above embodiments, the method, apparatus, and system for inverting the stress state of a long-span bridge provided in this application have at least the following beneficial effects:

[0023] This application proposes a method, device, and system for inverting the stress state of long-span bridges. By deeply analyzing the anomaly characteristics of the stress state and vibration state of the bridge structure under the influence of vehicle loads, and further considering the synchronous relationship between these anomalies, a vehicle load influence coefficient is constructed for each monitoring point during each monitoring period. This accurately assesses the degree of influence of vehicle loads on bridge monitoring data. By analyzing the coupled influence of temperature and vehicle loads on the bridge structure at each monitoring point, a coupling correlation coefficient is calculated. Combined with the vehicle load influence coefficient, a comprehensive influence coefficient for each monitoring point on the same day is constructed, accurately characterizing the degree of interference to the monitoring data at each monitoring point. Based on the comprehensive influence coefficient, the monitoring data of each monitoring point on the bridge is filtered and optimized, which reduces data errors caused by load and temperature during data acquisition. The filtered monitoring data is then input into a finite element analysis model for stress state inversion, which helps improve the accuracy of bridge stress state assessment. Attached Figure Description

[0024] To more clearly illustrate the technical solutions and advantages in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 A flowchart illustrating the steps of a method for inverting the stress state of a long-span bridge according to an embodiment of this application;

[0026] Figure 2 A flowchart illustrating the acquisition of the comprehensive influence coefficient of each monitoring point on a given day, as provided in one embodiment of this application;

[0027] Figure 3 This is a schematic diagram of the structure of a stress state inversion device for a long-span bridge provided in one embodiment of this application. Detailed Implementation

[0028] To further illustrate the technical means and effects adopted by this application to achieve the intended purpose of the invention, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a method, apparatus, and system for inverting the stress state of a long-span bridge according to this application. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0029] Unless otherwise specified and limited, terms such as “comprising,” “including,” or any other variations thereof are intended to cover a non-exclusive inclusion, such that a circuit structure, article, or device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such an article or device. Without further limitation, an element defined by the phrase “comprising one…” does not exclude the presence of other identical elements in the article or device that includes said element. Furthermore, the term “and / or” as used herein includes any and all combinations of one or more of the associated listed items. All technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0030] The following description, in conjunction with the accompanying drawings, details the specific scheme of the stress state inversion method, device, and system for long-span bridges provided in this application.

[0031] Please see Figure 1 The diagram illustrates a flowchart of a method for inverting the stress state of a long-span bridge according to an embodiment of this application. The method includes the following steps:

[0032] Step 1: Real-time acquisition of stress, vibration, and temperature data at various monitoring points on the bridge deck.

[0033] Long-span bridges are subjected to a complex service environment and are affected by a variety of factors, resulting in dynamic changes in their stress state and structural performance. Vehicle loads and temperature are two of the most significant factors influencing the structure during daily operation. Therefore, this application sets up monitoring points at predetermined intervals along the central axis of the bridge's foundation slab, and installs strain gauges, accelerometers, and temperature sensors at each monitoring point to collect stress, vibration, and temperature data in real time. In this embodiment, the time interval for collecting stress and vibration data is 0.01 seconds, and the time interval for collecting temperature data is 5 seconds, with a predetermined distance of 8 meters. This yields stress, vibration, and temperature data for all monitoring points.

[0034] Step 2: Divide the entire data acquisition time into multiple detection periods; based on the steepness of all peaks in the stress data of each monitoring point in each detection period, and the difference in the fluctuation amplitude of stress data between each monitoring point and its adjacent monitoring points in each detection period, obtain the first outlier value of each monitoring point in each detection period.

[0035] Bridge stress state inversion involves monitoring the response data of the bridge structure during actual use, such as strain and vibration, and combining this data with the bridge's physical model and mechanical principles. Mathematical optimization methods, such as finite element analysis, are then used to infer the internal stress distribution and load conditions of the bridge. However, the load-bearing capacity of a bridge is significantly influenced by a combination of factors, including its overall structural stiffness, dead load, live load, and environmental conditions. Under the combined effect of these factors, real-time stress data may contain varying degrees of error, thus affecting the accuracy and reliability of the bridge stress state inversion. Since vehicle load and temperature are major factors affecting bridge structures, this application analyzes the characteristics of vehicle load and temperature changes to correct stress data and compensate for errors. This improves the accuracy of the inversion results, ensures an accurate assessment of the bridge's stress state, and provides strong support for bridge safety management and maintenance decisions.

[0036] During vehicle movement, the load on the bridge structure at a single monitoring point changes dynamically. Generally, the greater the vehicle load, the greater the stress on the structure. Consequently, when the bridge structure is in good condition, the stress data fluctuates smoothly with changes in vehicle load, and the stress data changes at adjacent locations are relatively similar when the same vehicle passes by. However, if an anomaly occurs at a hinge joint or beam connection in the bridge, the stress value at that location may be several times higher than normal, and the stress data fluctuations will be more drastic, with a significantly increased rate of change.

[0037] Therefore, this application defines a preset duration for each data acquisition period as a detection time interval, which is 5 seconds in this embodiment. Taking the j-th detection time interval of the i-th monitoring point as an example, the following analysis is performed: when a vehicle passes near each monitoring point, the stress increases significantly, and the stress decreases rapidly after the vehicle leaves. Therefore, a quadratic polynomial fitting technique is used to obtain the fitting curve of the stress data of the i-th monitoring point in the j-th detection time interval; the sum of the kurtosis of the peaks where all the maximum points of the fitted curve of the stress data of the i-th monitoring point in the j-th detection time interval is denoted as... The result The larger the value, the more drastic the stress data change at that location. Then, the mean of all maxima in the fitted curve of the stress data at the i-th monitoring point during the j-th detection period is recorded as the fluctuation significance of the i-th monitoring point during the j-th detection period; the mean of the absolute differences in the fluctuation significance between the i-th monitoring point and its two adjacent monitoring points during the j-th detection period is recorded as... The result This reflects the difference in the fluctuation range of stress data between the i-th monitoring point and its adjacent monitoring points during the j-th detection period.

[0038] As a preferred embodiment, based on the steepness of all peaks in the stress data of each monitoring point in each detection period, and the difference between the stress data fluctuation amplitude of each monitoring point and its adjacent monitoring points in each detection period, the first outlier value of each monitoring point in each detection period is obtained, which is used to characterize the degree of anomaly of the stress data of each monitoring point in each detection period.

[0039] In this embodiment, the first outlier of the i-th monitoring point in the j-th detection period is denoted as... Its specific expression is: In the formula, This represents the first outlier at the i-th monitoring point during the j-th monitoring period. This is the sum of the kurtosis of all peaks containing the maximum values ​​in the fitted curve of the stress data at the i-th monitoring point during the j-th monitoring period. It is the mean of the absolute differences in the fluctuation significance between the i-th monitoring point and its two adjacent monitoring points within the j-th detection period.

[0040] income It reflects the stress anomaly characteristics caused by vehicle load at the i-th monitoring point during the j-th detection period. The larger the value, the greater the degree of anomaly in the stress data of the i-th monitoring point during the j-th detection period.

[0041] Step 3: Based on the dispersion and high-frequency characteristics of the vibration data of each monitoring point in each detection period, and the differences in the frequencies of all modal components of the vibration data of each monitoring point and its neighboring monitoring points in each detection period, obtain the second outlier value of each monitoring point in each detection period; based on the degree of difference between the first outlier value and the second outlier value of each monitoring point in each detection period and its neighboring periods, obtain the asynchronous coefficient of each monitoring point in each detection period; and combine the first outlier value and the second outlier value of each monitoring point in each detection period to obtain the vehicle load influence coefficient of each monitoring point in each detection period.

[0042] Furthermore, when a bridge has local defects, the structural stiffness near the defect location decreases, making it easier for vehicle loads to induce larger vibration responses with significantly increased amplitude. While high-frequency components are usually not prominent in the vibration response at normal bridge locations, local discontinuities in defect areas, such as cracks or loose connections, can excite high-frequency vibrations and cause differences in vibration between different monitoring points. Therefore, this application first obtains the range of vibration data at the i-th monitoring point during the j-th detection period, denoted as . The range reflects the amplitude difference under the influence of vehicle load. The larger the value, the greater the influence of vehicle load on the vibration data of the i-th monitoring point in the j-th detection period. Then, the vibration data of the i-th monitoring point in the j-th detection period is decomposed using the empirical mode decomposition algorithm. The number of output modal components is set to N, and modal components with different frequency characteristics are obtained. In this embodiment, N is 6. Then, the spectrum sequence of each modal component is obtained using fast discrete Fourier transform, and the frequency corresponding to the spectral centroid of each spectrum sequence is obtained. The maximum value among the frequencies corresponding to the spectral centroids of all spectrum sequences is recorded as the high-frequency vibration significance value of the i-th monitoring point in the j-th detection period, denoted as . The larger the value, the more significant the high-frequency components in the vibration data of the i-th monitoring point under the vehicle load during the j-th detection period. Furthermore, to consider the differences in vibration modes among different monitoring points, the frequencies of the spectral centroids corresponding to all modal components of the vibration data of each monitoring point during the j-th detection period are arranged from smallest to largest, obtaining the spectral centroid frequency sequence of each monitoring point during the j-th detection period. The Euclidean distance between the spectral centroid frequency sequences of the i-th monitoring point and its two adjacent monitoring points during the j-th detection period is calculated. The average of the obtained Euclidean distances is used as the vibration mode difference coefficient of the i-th monitoring point during the j-th detection period, denoted as [equation missing]. The result It reflects the differences in vibration modes between adjacent monitoring points.

[0043] In a preferred embodiment, based on the dispersion and high-frequency characteristics of the vibration data of each monitoring point in each detection period, and the difference between the frequencies of all modal components of the vibration data of each monitoring point and its adjacent monitoring points in each detection period, a second outlier value of each monitoring point in each detection period is obtained, which is used to characterize the degree of anomaly of the vibration data of each monitoring point in each detection period.

[0044] In this embodiment, the second outlier of the i-th monitoring point in the j-th detection period is denoted as... Its formula is: In the formula, This represents the second outlier at the i-th monitoring point during the j-th monitoring period. The range of vibration data at the i-th monitoring point during the j-th monitoring period is given by [the range of vibration data at the i-th monitoring point during the j-th monitoring period]. Let be the significant value of high-frequency vibration at the i-th monitoring point during the j-th monitoring period. Let be the vibration mode difference coefficient of the i-th monitoring point during the j-th monitoring period. The obtained... The larger the value, the more obvious the abnormal characteristics of the vibration data of the i-th monitoring point in the j-th detection period.

[0045] Furthermore, at locations where the bridge structure is damaged, if a vehicle passes by, the stress anomaly characteristics experienced by the bridge structure at these locations are typically greater, and the vibration anomaly characteristics are more pronounced. Therefore, the stress state anomaly characteristics and vibration state anomaly characteristics induced by vehicle loads are more consistent. In view of this, this application uses the M preceding detection periods of the j-th detection period as the nearest neighbor periods of the j-th detection period, where M is an integer within the range [12, 15], and in this embodiment, M is 13. Through the above steps, the first and second anomaly values ​​of each monitoring point in each detection period can be obtained. The DTW distance between all first and second anomaly values ​​of the i-th monitoring point in the j-th detection period and its nearest neighbor periods is calculated as the asynchronous coefficient of the i-th monitoring point in the j-th detection period. The obtained asynchronous coefficient reflects the asynchronous characteristics of stress anomaly and vibration anomaly of the i-th monitoring point of the bridge in the j-th detection period. The smaller the value, the more synchronized the stress anomaly characteristics and vibration anomaly characteristics of the i-th monitoring point of the bridge are in the j-th detection period, and the greater the influence of vehicle loads on the monitoring data at that monitoring point.

[0046] Furthermore, the ratio of the mean of the first and second outliers at the monitoring point during the j-th detection period to the asynchronous coefficient is used as the vehicle load influence coefficient of the i-th monitoring point during the j-th detection period. This vehicle load influence coefficient reflects the degree to which the stress state of the i-th monitoring point on the bridge is affected by vehicle load during the j-th detection period; the larger the value, the greater the influence of vehicle load on the stress state of the i-th monitoring point on the bridge during the j-th detection period.

[0047] Step 4: Based on the waveform similarity of all vehicle load influence coefficients and temperature data at each monitoring point on that day, as well as the average level of all vehicle load influence coefficients, obtain the comprehensive influence coefficient of each monitoring point on that day, and then obtain the optimization step size factor of each monitoring point on that day, and then filter the stress and vibration data of each monitoring point on that day.

[0048] Furthermore, besides vehicle loads, the temperature difference caused by solar radiation also significantly impacts the structural performance of bridges. When temperature changes, the stress state of a bridge undergoes multifaceted changes, including but not limited to structural deformation due to thermal expansion, changes in stress distribution, changes in material properties, changes in vibration characteristics, and changes in structural stability. These changes may adversely affect the safety and performance of the bridge. Therefore, this application deeply considers the correlation between temperature effects and the stress state of bridges. Taking the i-th monitoring point as an example, since the temperature change of the bridge structure is highly periodic with sunrise and sunset, the vehicle load influence coefficients of the i-th monitoring point during all detection periods within the day are arranged in ascending order of time, serving as the first sequence for the i-th monitoring point on that day; the average temperature of the i-th monitoring point during all detection periods within the day is arranged in ascending order of time, serving as the second sequence for the i-th monitoring point on that day. Because the effect of temperature on the structural performance of bridges has a certain time delay, when the surface temperature of the bridge changes, heat needs a certain amount of time to be transferred and distributed into the material. Therefore, this application calculates the SBD (Shape-Based Distance) distance between the first and second sequences of the i-th monitoring point on that day, as the coupling correlation coefficient of the i-th monitoring point on that day. The SBD distance is used to compare the waveform similarity of the two sequences, which can weaken the time delay effect on the temporal differences between the two sequences, making the calculated coupling correlation coefficient more accurately reflect the degree of coupling influence of temperature and vehicle load on the i-th monitoring point of the bridge on that day. The smaller the coupling correlation coefficient of the i-th monitoring point on that day, the greater the coupling influence of temperature and measurement load on the bridge structure at the i-th monitoring point. During real-time monitoring, the data collected at the corresponding monitoring point is more easily affected by multiple factors, leading to larger errors.

[0049] Furthermore, the ratio of the mean of the vehicle load influence coefficients of the i-th monitoring point across all detection periods on that day to the coupling correlation coefficient is denoted as the comprehensive influence coefficient of the i-th monitoring point on that day. Similarly, the comprehensive influence coefficients of all monitoring points on that day can be calculated. The flowchart for obtaining the comprehensive influence coefficient of each monitoring point on that day is shown below. Figure 2 As shown, the higher the comprehensive influence coefficient of each monitoring point on that day, the greater the interference from environmental factors on the bridge location where each monitoring point is located, and the more likely the monitoring data collected at the corresponding monitoring point will have a large error.

[0050] Therefore, this application employs the LMS filtering algorithm to process the bridge-related data collected in real time from each monitoring point. Based on the coupling correlation coefficient of each monitoring point on the same day, the step size factor in the LMS filter is optimized to obtain more accurate monitoring data. Specifically, when the comprehensive influence coefficient is larger, the interference on the collected data is more severe, so setting a larger step size factor can help the filter quickly track these changes; conversely, a smaller step size factor can be set to improve the stability of the filter. Since the step size factor is usually no higher than 0.1, the optimized step size factor for the i-th monitoring point on the same day is calculated as follows: In the formula, Let i be the optimization step size factor for the i-th monitoring point on that day. and These represent the preset minimum and maximum values ​​of the step size factor, respectively, in this embodiment. and They are 0.01 and 0.1 respectively. Let be the comprehensive impact coefficient of the i-th monitoring point on that day. This is the hyperbolic tangent function, used to normalize data.

[0051] Similarly, the optimal step size factor for all monitoring points on that day can be obtained.

[0052] The various monitoring data from each monitoring point on that day are used as inputs to the LMS filtering algorithm. The calculated optimized step size factor for each monitoring point on that day is used as the step size factor. The various monitoring data from each monitoring point on that day are then filtered to obtain the filtered monitoring data for each monitoring point. The various monitoring data refer to stress data and vibration data. The LMS filtering algorithm is a well-known technique, and its specific process will not be described in detail.

[0053] Finally, the filtered monitoring data from each monitoring point for the day are input into the finite element analysis model for stress state inversion, which helps to accurately assess the stress state of the bridge. The use of the finite element analysis model to invert the stress state is a well-known technique, and the specific process will not be elaborated further.

[0054] Please see Figure 3 , Figure 3 This is a schematic diagram of a device for inverting the stress state of a long-span bridge according to an embodiment of this application. In this embodiment, the terminal includes units used to execute the steps in an embodiment corresponding to a method for inverting the stress state of a long-span bridge. See also... Figure 3 The bridge stress state inversion device includes: a data acquisition module, a load influence analysis module, and a stress state inversion module.

[0055] The data acquisition module is used to acquire stress, vibration and temperature data at various monitoring points on the bridge deck in real time.

[0056] The load influence analysis module is used to obtain the vehicle load influence coefficient of each monitoring point in each detection period based on the anomaly characteristics of stress data and vibration data at each monitoring point, as well as the asynchronous characteristics between stress anomalies and vibration anomalies.

[0057] The stress state inversion module is used to obtain the comprehensive influence coefficient of each monitoring point on the day based on the waveform similarity of the vehicle load influence and temperature data of each monitoring point on the day, as well as the average level of the vehicle load influence coefficient of each monitoring point on the day. Then, the step size factor in the filtering algorithm is optimized to obtain the filtered monitoring data of each monitoring point on the day, and then the stress state of each monitoring point on the day is inverted.

[0058] Based on the same inventive concept as the above method, this application embodiment also provides a long-span bridge stress state inversion system, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of any one of the above-described long-span bridge stress state inversion methods.

[0059] The various embodiments in this application are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0060] It should be noted that, unless otherwise specified and limited, terms such as “comprising,” “including,” or any other variations thereof are intended to cover non-exclusive inclusion, such that a circuit structure, article, or device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such article or device. Without further limitations, an element defined by the phrase “comprising one…” does not exclude the presence of other identical elements in the article or device that includes said element. Furthermore, the term “and / or” as used herein includes any and all combinations of one or more of the associated listed items.

[0061] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not invented in this application.

[0062] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope.

Claims

1. A method for inverting the stress state of a long-span bridge, characterized in that, The method includes the following steps: Real-time acquisition of stress, vibration, and temperature data at various monitoring points on the bridge deck; The entire data acquisition time is divided into multiple detection periods; based on the steepness of all peaks in the stress data of each monitoring point in each detection period, and the difference between the stress data fluctuation amplitude of each monitoring point and its adjacent monitoring points in each detection period, the first outlier value of each monitoring point in each detection period is obtained. Based on the dispersion and high-frequency characteristics of the vibration data of each monitoring point in each detection period, and the differences between the frequencies of all modal components of the vibration data of each monitoring point and its neighboring monitoring points in each detection period, the second outlier of each monitoring point in each detection period is obtained; based on the degree of difference between the first outlier and the second outlier of each monitoring point in each detection period and its neighboring period, the asynchronous coefficient of each monitoring point in each detection period is obtained; and combined with the first outlier and the second outlier of each monitoring point in each detection period, the vehicle load influence coefficient of each monitoring point in each detection period is obtained. Based on the waveform similarity of all vehicle load influence coefficients and temperature data at each monitoring point on that day, as well as the average level of all vehicle load influence coefficients, the comprehensive influence coefficient of each monitoring point on that day is obtained, and then the optimization step size factor of each monitoring point on that day is obtained, and then the stress and vibration data of each monitoring point on that day are filtered. The formula for calculating the second outlier value of each monitoring point in each detection period is as follows: In the formula, This represents the second outlier at the i-th monitoring point during the j-th monitoring period. The range of vibration data at the i-th monitoring point during the j-th monitoring period is given by [the range of vibration data at the i-th monitoring point during the j-th monitoring period]. Let be the significant value of high-frequency vibration at the i-th monitoring point during the j-th monitoring period. Let be the vibration mode difference coefficient of the i-th monitoring point during the j-th detection period; The vehicle load influence coefficient of each monitoring point in each detection period refers to the ratio of the average of the first and second outliers of each monitoring point in each detection period to the asynchronous coefficient. The method for obtaining the comprehensive influence coefficient of each monitoring point on the same day is as follows: the vehicle load influence coefficients of each monitoring point during all detection periods on the same day are arranged in ascending order of time, which is taken as the first sequence of each monitoring point on the same day; the average temperature of each monitoring point during all detection periods on the same day is arranged in ascending order of time, which is taken as the second sequence of each monitoring point on the same day; the SBD distance between the first sequence and the second sequence of each monitoring point on the same day is taken as the coupling correlation coefficient of each monitoring point on the same day; the ratio of the mean of the vehicle load influence coefficients of each monitoring point during all detection periods on the same day to the coupling correlation coefficient is recorded as the comprehensive influence coefficient of each monitoring point on the same day.

2. The method for inverting the stress state of a long-span bridge as described in claim 1, characterized in that, The method for obtaining the first outlier of each monitoring point in each detection period is as follows: calculate the sum of the kurtosis of the peaks where all maximum points are located in the fitted curve of the stress data of each monitoring point in each detection period; Calculate the mean of the absolute differences in the significance of fluctuations between each monitoring point and its two adjacent monitoring points within each detection period; The product of the sum and the mean is recorded as the first outlier of each monitoring point in each detection period; wherein, the significance of the fluctuation of each monitoring point in each detection period refers to the mean of all maxima in the fitted curve of the stress data of each monitoring point in each detection period.

3. The method for inverting the stress state of a long-span bridge as described in claim 1, characterized in that, The method for obtaining the significant value of high-frequency vibration is as follows: the vibration data of the i-th monitoring point in the j-th detection period is decomposed into multiple modal components, and the spectrum sequence of each modal component is obtained. The maximum value of the frequency corresponding to the centroid of the spectrum sequence is recorded as the significant value of high-frequency vibration of the i-th monitoring point in the j-th detection period. The method for obtaining the vibration modal difference coefficient is as follows: the frequencies of the centroids corresponding to all modal components of the vibration data of each monitoring point in the j-th detection period are arranged from small to large to obtain the centroid frequency sequence of each monitoring point in the j-th detection period. The mean value of the Euclidean distance between the centroid frequency sequences of the i-th monitoring point and its two adjacent monitoring points in the j-th detection period is taken as the vibration modal difference coefficient of the i-th monitoring point in the j-th detection period.

4. The method for inverting the stress state of a long-span bridge as described in claim 1, characterized in that, The asynchronous coefficient of each monitoring point in each detection period refers to the DTW distance between all first outliers and all second outliers of each monitoring point in each detection period and its nearest neighboring period.

5. The method for inverting the stress state of a long-span bridge as described in claim 1, characterized in that, The formula for calculating the optimization step size factor for each monitoring point on that day is as follows: In the formula, Let i be the optimization step size factor for the i-th monitoring point on that day. and These represent the preset minimum and maximum values ​​of the step size factor, respectively. Let be the comprehensive impact coefficient of the i-th monitoring point on that day. It is the hyperbolic tangent function.

6. The method for inverting the stress state of a long-span bridge as described in claim 1, characterized in that, In the process of filtering the stress and vibration data of each monitoring point on the same day, the step size factor of the LMS filtering algorithm is set to the optimized step size factor of each monitoring point on the same day.

7. A device for inverting the stress state of a long-span bridge, characterized in that, The method for inverting the stress state of a long-span bridge as described in any one of claims 1-6, wherein the bridge stress state inversion device comprises: The data acquisition module is used to acquire stress, vibration and temperature data at various monitoring points on the bridge deck in real time. The load influence analysis module is used to obtain the vehicle load influence coefficient of each monitoring point in each detection period based on the anomaly characteristics of stress data and vibration data at each monitoring point, as well as the asynchronous characteristics between stress anomalies and vibration anomalies. The stress state inversion module is used to obtain the comprehensive influence coefficient of each monitoring point on the day based on the waveform similarity of the vehicle load influence and temperature data of each monitoring point on the day, as well as the average level of the vehicle load influence coefficient of each monitoring point on the day. Then, the step size factor in the filtering algorithm is optimized to obtain the filtered monitoring data of each monitoring point on the day, and then the stress state of each monitoring point on the day is inverted.

8. A stress state inversion system for long-span bridges, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements a method for inverting the stress state of a long-span bridge as described in any one of claims 1-6.

Images