Bridge deflection measuring device and method

Abstract

This application relates to the field of bridge deflection measurement technology, specifically to a bridge deflection measurement device and method, which includes: dynamically identifying vehicle impact, static load, and environmental interference data based on clustering and tilt angle change characteristics; dividing the tilt angle data into dynamic tilt angle data, static interference data, and static load data; integrating tilt angle distribution symmetry and temperature drift compensation to construct the goodness of each tilt angle data point; constructing a tilt angle drift-temperature curve through static interference data points; constructing a load influence degree by combining wind speed correlation and static load characteristic degree; and constraining the goodness of each static load data point by combining goodness and wind speed; constructing the goodness of each dynamic tilt angle data point based on the similarity between neighborhood data distributions to improve the symmetry accuracy of the deflection curve when vehicles cross the bridge; and performing tilt angle integration with goodness as weight to construct the deflection curve, thereby improving the measurement accuracy and efficiency of the bridge deflection curve.

Description

Technical Field

[0001] This application relates to the field of bridge deflection measurement technology, specifically to a bridge deflection measuring device and measurement method. Background Technology

[0002] In the scenario of deflection measurement of simply supported beam bridges, the deflection signal is affected by the superposition of quasi-static and impact effects from vehicle dynamic loads, static loads such as pedestrians / accumulated objects, and natural disturbances such as wind speed and temperature. Existing methods are difficult to effectively separate these factors. The symmetry disruption and lack of drift compensation during the integration process of tilt angle data further exacerbate the distortion of the deflection curve, failing to meet the requirements of high-precision monitoring. Summary of the Invention

[0003] To address the aforementioned technical problems, the purpose of this application is to provide a bridge deflection measuring device and method, the specific technical solution of which is as follows:

[0004] In a first aspect, embodiments of this application provide a method for measuring bridge deflection, the method comprising the following steps:

[0005] Collect tilt angle data and temperature data of the bridge at each preset location at each time, as well as wind speed data on the bridge; construct a tilt angle data sequence at each location; and record the sequence of tilt angle data at the same time as the tilt angle sequence at the same time.

[0006] Based on the data values ​​and trends of each data point in each dip angle data sequence, dynamic dip angle data and static dip angle data in each dip angle data sequence are obtained; based on the data distribution characteristics in the dip angle sequence at the same moment where each static dip angle data point is located, combined with wind speed data, the goodness of each static dip angle data point is calculated.

[0007] In each tilt angle data sequence, static load data points and static disturbance data points are classified based on the goodness of the static tilt angle data points. Based on the tilt angle drift and corresponding temperature of all static disturbance data points, tilt angle drift-temperature curves are constructed. Combining the tilt angle data values ​​and corresponding temperatures of each tilt angle data point, the drift influence of each tilt angle data point is constructed.

[0008] Based on the similarity between the wind speed data and the fluctuation of static load data points in each tilt angle data sequence, and the difference between the data distribution of static load data points and static disturbance data points in each tilt angle data sequence, combined with the drift influence, goodness and corresponding wind speed of each static load data point, the constrained goodness of each static load data point is constructed.

[0009] Based on the symmetry of the distribution of data measurement locations on the bridge, the symmetrical tilt angle data sequence of each tilt angle data sequence is determined; the goodness of each dynamic tilt angle data point is constructed based on the similarity between the distribution of the neighborhood data of each dynamic tilt angle data point and the corresponding symmetrical tilt angle data sequence.

[0010] Based on the original deflection calculation method, and combined with the goodness of various tilt angle data points, the new bridge deflection at each time and location is calculated; based on the new bridge deflection at all locations at each time, the bridge deflection curve at each time is constructed.

[0011] In one embodiment, the process of acquiring the dynamic tilt angle data and the static tilt angle data is as follows:

[0012] The slope of the fitted curve for each dip angle data point in each dip angle data sequence is calculated as the dip angle change rate of each dip angle data point in each dip angle data sequence. The product of the difference in the dip angle change rate and the difference in the dip angle data value between any two dip angle data points in each dip angle data sequence is used as the distance between any two dip angle data points in the clustering algorithm. All dip angle data points in each dip angle data sequence are clustered. The data in the cluster with the largest mean dip angle data in the cluster are used as dynamic dip angle data, and the data in the remaining clusters are used as static dip angle data.

[0013] In one embodiment, the process of obtaining the goodness of each static tilt angle data point is as follows:

[0014] In the simultaneous dip angle sequence, dip angle data points whose dip angle values ​​are smaller than the previous dip angle data value and larger than the next dip angle data value are taken as ideal feature data points; the midpoint within the simultaneous dip angle sequence is taken as the symmetry zero point, and the two dip angle data points closest to the symmetry zero point are taken as a pair of corresponding dip angle data points; the goodness of the static dip angle data point q is calculated. , The expression is: In the formula, N represents the number of elements in the dip angle sequence at the same time; n represents the number of ideal feature data points in the dip angle sequence at the same time for static dip angle data point q. represents the absolute value of the difference between the absolute values ​​of the two dip angle data points in the i-th group of the dip angle data point pair at the same instant in the dip angle sequence of the static dip angle data point q; b represents the number of corresponding dip angle data point pairs in the dip angle sequence of the static dip angle data point q. This represents the inclination data value of the symmetrical zero point within the inclination sequence at the same instantaneous moment of the static inclination data point q. This represents the wind speed at the moment the static tilt angle data point q was collected; It is a preset minimum positive number.

[0015] In one embodiment, the process of obtaining the drift influence is as follows:

[0016] In each tilt angle data sequence, the static tilt angle data point with the highest goodness is taken as the reference point, the average goodness of all static tilt angle data points is taken as the first threshold, the data points with goodness greater than or equal to the first threshold are taken as static disturbance data points, and the remaining data points are taken as static load data points.

[0017] The difference between the tilt angle data values ​​of each static interference data point and the corresponding reference point is taken as the tilt angle drift value of each static interference data point; curve fitting is performed based on the temperature and tilt angle drift values ​​corresponding to all static interference data points, and the resulting fitted curve is denoted as the tilt angle drift-temperature curve of each tilt angle data sequence.

[0018] The tilt drift value of each tilt data point in the tilt drift-temperature curve is determined by the temperature corresponding to each tilt data point; the difference between the tilt data value and the tilt drift value of each tilt data point is divided by the calculated tilt data value to obtain the drift influence degree of each tilt data point.

[0019] In one embodiment, the process of obtaining the goodness of each static load data point after constraint is as follows:

[0020] The similarity between the fitted curve of all wind speed data and the fitted curve of all static load data points in each tilt angle data sequence is used as the correlation between tilt angle data and wind speed in each tilt angle data sequence.

[0021] In each tilt angle data sequence, the sequence consisting of all static load data points is denoted as the first sequence; the sequence consisting of static disturbance data points with the same temperature as all data points in the first sequence is denoted as the second sequence; the sum of the differences in tilt angle data values ​​between the corresponding data points in the first sequence and the second sequence is used as the static load characteristic of each tilt angle data sequence.

[0022] Based on the wind speed corresponding to each static load data point, and combined with the correlation of the tilt angle data sequence and the static load characteristic degree, the load influence degree of each static load data point is calculated.

[0023] Calculate the goodness of static load data point k after constraints , The expression is:

[0024] ,in, This represents the goodness of static load data point k before constraints. The load influence degree of static load data point k. This represents the degree of drift influence of static load data point k.

[0025] In one embodiment, the expression for the load influence degree is:

[0026] In the formula, The normalized value of the wind speed corresponding to static load data point k is represented; GL and Z represent the correlation and the static load characteristic degree of the tilt angle data sequence where static load data point k is located, respectively.

[0027] In one embodiment, the process of obtaining the symmetrical dip angle data sequence of each dip angle data sequence is as follows:

[0028] Using the bridge's preset location as a symmetrical point, the inclination data sequences of two acquisition locations that are symmetrical about the symmetrical point are used as symmetrical inclination data sequences to obtain the symmetrical inclination data sequences of each inclination data sequence.

[0029] In one embodiment, the process of obtaining the excellence of each dynamic tilt angle data point is as follows:

[0030] In each dip angle data series, each dynamic dip angle data point The sequence of all neighborhood data of a point is denoted as the neighborhood data sequence of each dynamic tilt angle data point. The reciprocal of the metric distance between a point and the neighborhood data sequence of each dynamic dip point in its symmetric dip data sequence is used as... The approximation degree between a point and each dynamic dip angle data point in the symmetrical dip angle data sequence is used, with the maximum value of the approximation degree in the symmetrical dip angle data sequence being taken as the dynamic dip angle data point. The quality of the point.

[0031] In one embodiment, the process of obtaining the new bridge deflection at each time and location is as follows:

[0032] The product of the inclination values ​​and the goodness of each inclination angle is used to replace the inclination values ​​of each inclination angle in the original bridge deflection calculation formula. The resulting calculation is used as the new bridge deflection at each time and location.

[0033] Secondly, embodiments of this application also provide a bridge deflection measuring device, including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement the steps of any of the methods described above.

[0034] The embodiments of this application have at least the following beneficial effects:

[0035] This application first uses a clustering algorithm and tilt angle change characteristics to dynamically identify vehicle impact, static load, and environmental disturbance data, dividing the tilt angle data into dynamic tilt angle data, static disturbance data, and static load data. It then integrates tilt angle distribution symmetry, temperature drift compensation, and wind speed to quantify data reliability and provide a basis for weighted integration. A tilt angle drift-temperature curve is constructed using static disturbance data points to correct temperature drift errors in real time. The drift influence degree is introduced to dynamically adjust the weights and suppress cumulative integration errors. Combining wind speed correlation and static load characteristics, a load influence degree is constructed to constrain the goodness weights of static load points, avoiding confusion between wind loads and static loads. Based on the similarity between each dynamic tilt angle data point and the neighborhood data distribution of each dynamic tilt angle data point in the corresponding symmetrical tilt angle data sequence, the goodness of each dynamic tilt angle data point is constructed, improving the symmetry accuracy of the deflection curve when vehicles cross the bridge. Tilt angle integration is performed using the goodness as weights to construct the deflection curve, improving the measurement accuracy and efficiency of the bridge deflection curve. Attached Figure Description

[0036] 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.

[0037] Figure 1 A flowchart illustrating the steps of a bridge deflection measurement method according to one embodiment of this application;

[0038] Figure 2 This is a schematic diagram illustrating the process of acquiring dynamic and static tilt angle data. Detailed Implementation

[0039] 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 bridge deflection measuring device and method proposed 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.

[0040] Unless otherwise defined, 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.

[0041] The following description, in conjunction with the accompanying drawings, details the specific scheme of the bridge deflection measuring device and method provided in this application.

[0042] Please see Figure 1 The diagram illustrates a flowchart of a bridge deflection measurement method according to an embodiment of this application, which includes the following steps:

[0043] Step S1: Collect tilt angle data and temperature data of the bridge at each preset location at each time, as well as wind speed data on the bridge; construct a tilt angle data sequence for each location; and record the sequence of tilt angle data at the same time as the tilt angle sequence at the same time.

[0044] The bridge deflection measurement device mainly consists of an inclination sensor array, a data transmission system, and a data processing and analysis system. The inclination sensor array comprises multiple high-precision, low-noise MEMS inclinometers. Inclination sensors are installed at pre-selected key section locations along the longitudinal axis of the bridge's main beam to measure the bridge's bending deformation in the vertical plane. In this embodiment, inclination sensors are installed at nine locations: the left support (0 span), 1 / 8 span, 1 / 4 span…1 span (right support). In other embodiments of this application, the implementer can adjust the settings according to actual conditions. The number of inclination sensors can be adjusted to be every 5m-12m along the bridge. The number of inclination sensors is related to the bridge length; the longer the bridge, the more inclination sensors are required.

[0045] Temperature sensors are also installed at each tilt sensor mounting location to measure temperature changes. All sensors sample synchronously at a consistent frequency; in this embodiment, the data acquisition frequency is set to 100Hz. In other embodiments of this application, the implementer can set the data acquisition frequency according to actual conditions.

[0046] Inclination data is collected at various locations using tilt sensors, temperature data is collected at various locations using temperature sensors, and wind speed data is collected at the half-span of the bridge using wind speed sensors. All collected tilt, temperature, and wind speed data are transmitted to a server for analysis to facilitate subsequent deflection measurements.

[0047] The sequence formed by arranging all tilt angle data collected by each tilt sensor in ascending chronological order is denoted as the tilt angle data sequence of each tilt sensor. The sequence formed by arranging the tilt angle data collected by all tilt sensors at the same moment in the order from the left support to the first span is denoted as the tilt angle sequence at the same moment.

[0048] Step S2: Classify the data points in each dip angle data sequence based on their data values ​​and trends to obtain dynamic dip angle data and static dip angle data in each dip angle data sequence; calculate the goodness of each static dip angle data point based on the data distribution characteristics in the dip angle sequence at the same moment where each static dip angle data point is located.

[0049] The above steps completed the acquisition of tilt angle data at different locations on the bridge. The tilt angle data acquired by the tilt angle sensor is usually used to calculate the deflection by integration. The accuracy of the tilt angle data acquisition directly affects the deflection calculation. However, the tilt angle sensor may experience noise interference such as zero-point drift and temperature drift during the tilt angle monitoring of the bridge, which will lead to the accumulation of low-frequency errors during the integration process, resulting in low accuracy of the deflection calculation results.

[0050] Based on the above steps, tilt angle data and temperature data were collected at each tilt sensor. Since bridge structures are typically subjected to multiple loads during service, the deflection monitoring signal contains various loads and effects. When vehicles cross the bridge, the deflection is related to three types of loads; when vehicles do not cross the bridge, the deflection is related to two types of loads (excluding vehicle impact as described below). When vehicles cross the bridge, the dynamic deflection time history curve of the bridge is formed by superimposing the ultra-low frequency curve of the quasi-static effect of vehicle loads with the dynamic effects of vehicle impacts and noise at different frequencies. That is, the dynamic deflection can be composed of three parts: first, the deflection caused by static loads, which is a low-frequency, high-amplitude vibration; second, the bridge vibration caused by vehicle vibrations and bridge surface unevenness, which is a medium-frequency, low-amplitude vibration; and third, the vibration caused by natural disturbances such as wind loads, which is a high-frequency, low-amplitude vibration.

[0051] (1) First, the data in the tilt angle data sequence is classified into static tilt angle data and dynamic tilt angle data. The dynamic tilt angle data is the tilt angle data affected by the vehicle passing the tilt angle sensor, while the static tilt angle data is the tilt angle data affected by the static load or natural disturbances such as noise and wind load on the bridge. According to the instantaneous rotation response characteristics under the impact of the vehicle, when the vehicle travels and affects a certain tilt angle sensor, it will cause an impact that results in obvious and rapid fluctuations or oscillations in the tilt angle data. The tilt angle data will show a waveform containing peaks and troughs.

[0052] Therefore, this application uses K-means clustering to classify the dip angle data in each dip angle data sequence. The number of clusters is two, and initial cluster centers are randomly selected. The distance between two dip angle data points is the product of the absolute difference between the rates of change of dip angle and the absolute difference between the dip angle data values. The larger the distance metric, the lower the probability that the two dip angle data points belong to the same dip angle category. Thus, the dip angle data points in the dip angle data sequence can be divided into two categories, denoted as the first category and the second category. K-means clustering is a well-known technique, and its specific process will not be elaborated further.

[0053] It should be noted that this application only provides one clustering method for clustering tilt angle data. There are many existing clustering methods, and implementers can also use other clustering algorithms to cluster tilt angle data. This application does not impose any specific restrictions.

[0054] The rate of change of dip angle at each dip angle data point in each dip angle data sequence is as follows: Using each dip angle data sequence as input to the least squares method, curve fitting is performed to obtain the fitted curve for each dip angle data sequence. The slope of this fitted curve at each dip angle data point is taken as the rate of change of dip angle at each data point. The least squares method is a well-known technique, and its specific process will not be elaborated further.

[0055] It should be noted that, for curve fitting of tilt angle data sequences, this application only provides one fitting method. There are many existing fitting methods, and implementers may also use other fitting algorithms to perform curve fitting of tilt angle data sequences. This application does not impose any specific limitations.

[0056] Calculate the average value of the tilt data points in each cluster; take the data in the cluster with the largest average value as the dynamic tilt data, and take the data in the other cluster as the static tilt data.

[0057] (2) Based on the above steps, the data points in the tilt angle data sequence were classified. For static tilt angle data, the tilt angle data points that are not disturbed were first obtained. Here, taking the static tilt angle data point q in any tilt angle data sequence as an example, since the data sampling frequency of different tilt angle sensors is consistent and synchronous sampling is performed, the static tilt angle data point q has tilt angle data points with the same sampling time in the other tilt angle data sequences. According to prior knowledge, in an ideal situation, that is, without external interference, the tilt angle data at different locations of the bridge should show a monotonic change from positive to negative. The left half-span is positive but gradually decreases, the right half-span is negative and the absolute value gradually increases, the mid-span is the zero point of tilt angle, and the absolute value of tilt angle is the largest at the support, symmetrical to the zero point of tilt angle. For bridge deflection, the deflection should show a monotonic increase from the support to the mid-span, reaching the maximum at the mid-span, and then decreasing symmetrically. The deflection curve is a smooth parabola, symmetrical to the mid-span.

[0058] The goodness of the static dip angle data point q is constructed based on the distribution of dip angle data values ​​in the dip angle sequence at the same instant. The acquisition process is as follows:

[0059] The number of dip angle data points whose data value distribution in the dip angle sequence at the same moment conforms to the ideal characteristics is analyzed. Specifically, since the nine dip angle data points from the left support to the first span should ideally show a decreasing change, in the dip angle sequence at the same moment, if the value of each dip angle data point is smaller than the value of the previous dip angle data point and larger than the value of the next dip angle data point, then the dip angle data point is regarded as the dip angle data point whose data value distribution conforms to the ideal characteristics and is recorded as the ideal characteristic data point.

[0060] Using the tilt data collected by the tilt sensor at 1 / 2 span as the zero point of symmetry, that is, the midpoint of the tilt sequence at the same moment is taken as the zero point of symmetry, and the two tilt data points closest to the zero point of symmetry are taken as a pair of corresponding tilt data points. Then there are a total of 4 pairs of corresponding tilt data points in a tilt sequence at the same moment.

[0061] The goodness of point q in the static tilt data is calculated using the following expression:

[0062]

[0063] In the formula, The value represents the goodness of the static dip angle data point q; N represents the number of elements in the dip angle sequence at the same time; n represents the number of ideal feature data points in the dip angle sequence at the same time for the static dip angle data point q. represents the absolute value of the difference between the absolute values ​​of the data points of the two dip angle data points in the i-th pair of corresponding dip angle data points in the dip angle sequence at the same instant of the static dip angle data point q; b represents the number of corresponding dip angle data point pairs in the dip angle sequence at the same instant of the static dip angle data point q. In this embodiment, the value of b is 4. This represents the inclination data value of the symmetrical zero point within the inclination sequence at the same instantaneous moment of the static inclination data point q. This indicates the wind speed at the moment the static tilt angle data point q was collected; The value is a preset, extremely small positive number, intended to prevent the numerator or denominator from being zero. In the embodiments of this application, The value is set to 0.001. In other embodiments of this application, the implementer may adjust it according to the actual situation.

[0064] The smaller the value, the more symmetrical the distribution of the tilt angle data, and the more it conforms to the ideal situation. The smaller the absolute value and the closer it is to 0, the more it conforms to the characteristics of a symmetrical zero point, and the greater its goodness. The lower the wind speed, the less likely the static tilt angle data point is to be affected by wind load.

[0065] The goodness of each static tilt angle data point in the tilt angle data sequence can be obtained by following the above steps. The greater the goodness, the less the static tilt angle data point is affected by the load.

[0066] Step S3: In each tilt angle data sequence, static load data points and static disturbance data points are classified based on the goodness of the static tilt angle data points; tilt angle drift-temperature curves are constructed based on the tilt angle drift of all static disturbance data points and their corresponding temperatures; and the drift influence of each tilt angle data point is constructed by combining the tilt angle data value and corresponding temperature of each tilt angle data point.

[0067] The static tilt angle data point with the highest goodness in each tilt angle data sequence is used as the reference point; and the average goodness of the static tilt angle data points in each tilt angle data sequence is used as the first threshold. The static tilt angle data points are further classified according to the first threshold to obtain static load data points and static disturbance data points, wherein the goodness of the static disturbance data points is greater than or equal to the first threshold.

[0068] The absolute value of the difference between the tilt data value of each static disturbance data point and the corresponding reference point in each tilt data sequence is taken as the tilt drift value of each static disturbance data point in each tilt data sequence. The coordinate points corresponding to each static disturbance data point are obtained by taking the temperature at the same acquisition time and location as the x-axis and the tilt drift value as the y-axis. The coordinate points corresponding to all static disturbance data points in each tilt data sequence are used as the input of the least squares method to perform curve fitting. The fitted curve is recorded as the tilt drift-temperature curve of each tilt data sequence.

[0069] The tilt drift-temperature curve can be obtained by following the above steps. The tilt drift value can then be adaptively obtained based on the temperature values ​​of the tilt data points. For each tilt data sequence, the tilt drift value of each tilt data point in the corresponding tilt drift-temperature curve is taken as the tilt drift value of that data point. The absolute value of the difference between the tilt data value and the tilt drift value is divided by the calculated tilt data value to determine the drift influence of each data point. A larger drift influence indicates a more severe impact of temperature drift on the tilt data point, resulting in a smaller weighting effect in subsequent calculations.

[0070] Step S4: Based on the similarity between the wind speed data and the fluctuation of the static load data points in each tilt angle data sequence, and the difference between the data distribution of the static load data points and the static disturbance data points in each tilt angle data sequence, combined with the drift influence, goodness and corresponding wind speed of each static load data point, construct the constrained goodness of each static load data point.

[0071] Based on the characteristics of static load data points, it is known that they may be affected by wind loads and other static loads. The application of static loads disrupts the symmetry of the corresponding data sequence of tilt angle data points, resulting in a lower goodness of performance. Therefore, directly using goodness of performance to reflect the interference status of static load data points is ineffective. This application further adaptively constructs the load influence degree of static load data points based on the characteristics of wind loads and static loads, as well as the influence of temperature drift. Static loads typically include stationary pedestrians, vehicles, or piled-up objects. The tilt angle should be maximum at the location where the static load is applied, attenuating towards both sides of the application point. This application obtains the correlation between tilt angle data and wind speed, as well as the static load characteristic degree, based on the changes in tilt angle data collected by the tilt angle sensor and the distribution characteristics of tilt angle data values ​​at different locations on the bridge under static load application. The specific process is as follows:

[0072] First, all static load data points in each tilt angle data sequence are fitted using a curve fitting algorithm to obtain the tilt angle data curves that change with time. Simultaneously, all wind speed data are fitted using the same algorithm to obtain wind speed data curves that change with time. Furthermore, the absolute value of the Pearson correlation coefficient between the tilt angle data curves and wind speed data curves for each tilt angle data sequence is obtained as the correlation between the tilt angle data and wind speed of the corresponding tilt angle sensor for each tilt angle data sequence. The Pearson correlation coefficient is a well-known technique, and its specific process will not be elaborated upon. This application uses the least squares method as the curve fitting algorithm, but other curve fitting algorithms may be used, and this application does not impose specific limitations.

[0073] Then, in each tilt angle data sequence, the temperature data collected at the same acquisition time and location as each tilt angle data point is taken as the temperature of each tilt angle data point; a sequence composed of all static load data points is obtained, denoted as the first sequence; a sequence composed of static interference data points with the same temperature corresponding to all data points in the first sequence is obtained, denoted as the second sequence; the sum of the absolute values ​​of the differences in tilt angle data values ​​between corresponding data points in the first sequence and the second sequence is taken as the static load characteristic of the tilt angle data sequence. If there are no static interference data points with the same temperature, the static interference data point with the closest corresponding temperature is obtained.

[0074] The greater the static load characteristic, the more severely the original symmetry is destroyed, and the more the static load data points conform to the characteristics of static load.

[0075] Furthermore, the load influence degree can be adaptively constructed, expressed as:

[0076] ,in, The load influence degree of static load data point k; GL represents the normalized value of the wind speed at the time of data acquisition for static load data point k; Z represents the correlation between the tilt angle data and the wind speed in the tilt angle data sequence containing static load data point k; and Z represents the static load characteristic degree of the tilt angle data sequence containing static load data point k. This application obtains the normalized value of each wind speed data by normalizing all wind speeds using the minimization method.

[0077] The goodness of static load data points can then be constrained by the load influence degree, resulting in the constrained goodness, expressed as:

[0078] ,in, The goodness of the static load data point k after constraints; This indicates the goodness of static load data point k before constraints. The load influence degree of static load data point k. This represents the drift influence of static load data point k. A larger drift influence indicates a smaller actual performance level. Conversely, a larger load influence indicates a higher actual performance level for the static load data point.

[0079] Step S5: Determine the symmetrical tilt data sequence of each tilt data sequence based on the symmetrical distribution of data measurement locations on the bridge; construct the goodness of each dynamic tilt data point based on the similarity between each dynamic tilt data point and the neighborhood data distribution of each dynamic tilt data point in the corresponding symmetrical tilt data sequence; calculate the new bridge deflection at each time and location based on the original deflection calculation method and the goodness of various tilt data points; construct the bridge deflection curve at each time based on the new bridge deflection at all locations at each time.

[0080] Based on the above steps, the goodness and actual goodness of static disturbance data points and static load data points in the static tilt angle data points were constructed. This application further obtains the goodness of dynamic tilt angle data points. As described above, dynamic tilt angle data points usually also exhibit symmetry. For any dynamic tilt angle data point in a tilt angle data sequence Q... In short, first obtain In this embodiment of the application, the neighborhood data points of a point are the distances from the dynamic tilt angle data points. The 10 most recent dip angle data points are used as dynamic dip angle data points. The neighborhood data points of the point, and The sequence of all neighboring data points of a point is denoted as . The neighborhood data sequence of a point. In other embodiments of this application, the implementer can adjust the neighborhood size according to the actual situation. The more neighborhood data points there are, the more accurate the calculation will be, but the longer it will take.

[0081] Furthermore, obtain the tilt data sequence that is symmetrical to the tilt data sequence Q. Specifically, taking the 1 / 2 span of the bridge as the symmetrical point, the two tilt data sequences collected by the two tilt sensors whose sampling times are consistent about the symmetrical point are used as symmetrical tilt data sequences.

[0082] Get The DTW distance between the neighborhood data sequence of a point and the neighborhood data sequence of each dynamic dip data point in its symmetric dip data sequence is used as the reciprocal of the DTW distance. The approximation degree between a point and each dynamic dip angle data point in its symmetrical dip angle data sequence is used, with the dynamic dip angle data point having the largest approximation in the symmetrical dip angle data sequence being selected as the approximation. The corresponding point of the point, and The approximation between a point and its corresponding point is used as the dynamic tilt angle data point. The dominance of the points is then adaptively constructed based on the above steps. The calculation of the DTW distance is a well-known technique, and its specific process will not be elaborated further. It should be noted that this application only provides one distance measurement method for measuring the distance between neighboring data sequences. Many existing distance measurement methods exist, and implementers may also use other distance measurement algorithms to measure the distance between neighboring data sequences; this application does not impose any specific limitations.

[0083] Furthermore, when obtaining bridge deflection through tilt angle data, it is usually obtained by integration. This integration involves tilt angle data points collected by multiple tilt angle sensors, which may include dynamic tilt angle data points, static interference data points, static load data points, and other types of tilt angle data points. When integrating, the corresponding goodness is used as a constant and multiplied by the corresponding tilt angle data value before integration, which effectively reduces the deflection error caused by environmental interference such as temperature drift.

[0084] The above steps effectively obtain deflection data points at different locations on the bridge at each time step. Furthermore, the least squares method is used to perform curve fitting on the deflection data points at all locations at each time step to obtain the bridge's deflection curve for each time step. Since the bridge involved in this application is a simply supported beam, a quadratic polynomial can be used. The deflection curve is fitted using this as the fitting equation. This represents the output deflection curve of the bridge; U represents the span of the bridge, specifically the horizontal distance between the two supports, with zero deflection at the left support; a represents the coefficient to be fitted; and x represents the position on the bridge. This allows us to obtain the deflection curve of the bridge at each time point.

[0085] The process of acquiring dynamic and static tilt data is illustrated in the diagram below. Figure 2 As shown.

[0086] Based on the same inventive concept as the above method, this application also provides a bridge deflection measuring device, 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 bridge deflection measuring methods.

[0087] In summary, this application provides a bridge deflection measurement method. First, it dynamically identifies vehicle impact, static load, and environmental disturbance data based on clustering and tilt angle change characteristics, dividing the tilt angle data into dynamic tilt angle data, static disturbance data, and static load data. It then integrates tilt angle distribution symmetry, temperature drift compensation, and wind speed to quantify data reliability and provide a basis for weighted integration. A tilt angle drift-temperature curve is constructed using static disturbance data points to correct temperature drift errors in real time, and the drift influence degree is introduced to dynamically adjust the weights and suppress cumulative integration errors. Combining wind speed correlation and static load characteristics, a load influence degree is constructed to constrain the goodness weights of static load points, avoiding confusion between wind loads and static loads. Based on the similarity between each dynamic tilt angle data point and the neighborhood data distribution of each dynamic tilt angle data point in the corresponding symmetrical tilt angle data sequence, the goodness of each dynamic tilt angle data point is constructed, improving the symmetry accuracy of the deflection curve when vehicles cross the bridge. Tilt angle integration is performed with goodness as the weight, thereby constructing the deflection curve and improving the measurement accuracy and efficiency of the bridge deflection curve.

[0088] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this application. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0089] 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.

[0090] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the principles of this application should be included within the protection scope of this application.

Claims

1. A method for measuring bridge deflection, characterized in that, The method includes the following steps: Collect tilt angle data and temperature data of the bridge at each preset location at each time, as well as wind speed data on the bridge; construct a tilt angle data sequence at each location; and record the sequence of tilt angle data at the same time as the tilt angle sequence at the same time. Based on the data values ​​and trends of each data point in each dip angle data sequence, dynamic dip angle data and static dip angle data in each dip angle data sequence are obtained; based on the data distribution characteristics in the dip angle sequence at the same moment where each static dip angle data point is located, combined with wind speed data, the goodness of each static dip angle data point is calculated. In each tilt angle data sequence, static load data points and static disturbance data points are classified based on the goodness of the static tilt angle data points. Based on the tilt angle drift and corresponding temperature of all static disturbance data points, tilt angle drift-temperature curves are constructed. Combining the tilt angle data values ​​and corresponding temperatures of each tilt angle data point, the drift influence of each tilt angle data point is constructed. Based on the similarity between the wind speed data and the fluctuation of static load data points in each tilt angle data sequence, and the difference between the data distribution of static load data points and static disturbance data points in each tilt angle data sequence, combined with the drift influence, goodness and corresponding wind speed of each static load data point, the constrained goodness of each static load data point is constructed. Based on the symmetry of the distribution of data measurement locations on the bridge, a symmetrical inclination data sequence is determined for each inclination data sequence; the goodness of each dynamic inclination data point is constructed based on the similarity between the distribution of each dynamic inclination data point and the neighborhood data distribution of each dynamic inclination data point in the corresponding symmetrical inclination data sequence; based on the original deflection calculation method and combined with the goodness of various inclination data points, the new bridge deflection at each time and location is calculated; and the bridge deflection curve at each time and location is constructed based on the new bridge deflection at all locations at each time.

2. The bridge deflection measurement method as described in claim 1, characterized in that, The process of acquiring the dynamic tilt angle data and the static tilt angle data is as follows: The slope of the fitted curve for each dip angle data point in each dip angle data sequence is calculated as the dip angle change rate of each dip angle data point in each dip angle data sequence. The product of the difference in the dip angle change rate and the difference in the dip angle data value between any two dip angle data points in each dip angle data sequence is used as the distance between any two dip angle data points in the clustering algorithm. All dip angle data points in each dip angle data sequence are clustered. The data in the cluster with the largest mean dip angle data in the cluster are used as dynamic dip angle data, and the data in the remaining clusters are used as static dip angle data.

3. The bridge deflection measurement method as described in claim 1, characterized in that, The process of obtaining the excellence of each static tilt angle data point is as follows: In the dip angle sequence at the same moment, the dip angle data point whose dip angle data value is smaller than the previous dip angle data value and larger than the next dip angle data value is taken as the ideal feature data point; the midpoint in the dip angle sequence at the same moment is taken as the symmetry zero point, and the two dip angle data points that are closest to the symmetry zero point are taken as a pair of corresponding dip angle data points; Calculate the goodness of point q in the static tilt angle data. , The expression is: In the formula, N represents the number of elements in the dip angle sequence at the same time; n represents the number of ideal feature data points in the dip angle sequence at the same time for static dip angle data point q. represents the absolute value of the difference between the absolute values ​​of the two dip angle data points in the i-th group of the dip angle data point pair at the same instant in the dip angle sequence of the static dip angle data point q; b represents the number of corresponding dip angle data point pairs in the dip angle sequence of the static dip angle data point q. This represents the inclination data value of the symmetrical zero point within the inclination sequence at the same instantaneous moment of the static inclination data point q. This represents the wind speed at the moment the static tilt angle data point q was collected; It is a preset minimum positive number.

4. The bridge deflection measurement method as described in claim 1, characterized in that, The process for obtaining the drift influence is as follows: In each tilt angle data sequence, the static tilt angle data point with the highest goodness is taken as the reference point, the average goodness of all static tilt angle data points is taken as the first threshold, the data points with goodness greater than or equal to the first threshold are taken as static disturbance data points, and the remaining data points are taken as static load data points. The difference between the tilt angle data values ​​of each static interference data point and the corresponding reference point is taken as the tilt angle drift value of each static interference data point; curve fitting is performed based on the temperature and tilt angle drift values ​​corresponding to all static interference data points, and the resulting fitted curve is denoted as the tilt angle drift-temperature curve of each tilt angle data sequence. The tilt drift value of each tilt data point in the tilt drift-temperature curve is determined by the temperature corresponding to each tilt data point; the difference between the tilt data value and the tilt drift value of each tilt data point is divided by the calculated tilt data value to obtain the drift influence degree of each tilt data point.

5. The bridge deflection measurement method as described in claim 1, characterized in that, The process of obtaining the goodness of each static load data point after constraint is as follows: The similarity between the fitted curve of all wind speed data and the fitted curve of all static load data points in each tilt angle data sequence is used as the correlation between tilt angle data and wind speed in each tilt angle data sequence. In each tilt angle data sequence, the sequence consisting of all static load data points is denoted as the first sequence; the sequence consisting of static disturbance data points with the same temperature as all data points in the first sequence is denoted as the second sequence; the sum of the differences in tilt angle data values ​​between the corresponding data points in the first sequence and the second sequence is used as the static load characteristic of each tilt angle data sequence. Based on the wind speed corresponding to each static load data point, and combined with the correlation of the tilt angle data sequence and the static load characteristic degree, the load influence degree of each static load data point is calculated. Calculate the goodness of static load data point k after constraints , The expression is: ,in, This represents the goodness of static load data point k before constraints. The load influence degree of static load data point k. This represents the degree of drift influence of static load data point k.

6. The bridge deflection measurement method as described in claim 5, characterized in that, The expression for the load influence degree is: In the formula, The normalized value of the wind speed corresponding to static load data point k is represented; GL and Z represent the correlation and the static load characteristic degree of the tilt angle data sequence where static load data point k is located, respectively.

7. The bridge deflection measurement method as described in claim 1, characterized in that, The process for obtaining the symmetrical dip angle data sequence of each dip angle data sequence is as follows: Using the bridge's preset location as a symmetrical point, the inclination data sequences of two acquisition locations that are symmetrical about the symmetrical point are used as symmetrical inclination data sequences to obtain the symmetrical inclination data sequences of each inclination data sequence.

8. A method for measuring bridge deflection as described in claim 1, characterized in that, The process of obtaining the excellence of each dynamic tilt angle data point is as follows: In each dip angle data series, each dynamic dip angle data point The sequence of all neighborhood data of a point is denoted as the neighborhood data sequence of each dynamic tilt angle data point. The reciprocal of the metric distance between a point and the neighborhood data sequence of each dynamic dip point in its symmetric dip data sequence is used as... The approximation degree between a point and each dynamic dip angle data point in the symmetrical dip angle data sequence is used, with the maximum value of the approximation degree in the symmetrical dip angle data sequence being taken as the dynamic dip angle data point. The quality of the point.

9. A method for measuring bridge deflection as described in claim 1, characterized in that, The process of obtaining the new bridge deflection at each time and location is as follows: The product of the inclination values ​​and the goodness of each inclination angle is used to replace the inclination values ​​of each inclination angle in the original bridge deflection calculation formula. The resulting calculation is used as the new bridge deflection at each time and location.

10. A bridge deflection measuring device, 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 the steps of the method as described in any one of claims 1-9.

Images