Bridge load test detection system based on wireless transmission

Abstract

The application belongs to the technical field of intelligence, and discloses a bridge load test detection system based on wireless transmission; the system comprises the following steps: uniformly distributing a bridge into M sections, and calculating the double-field coupling weight coefficient of each section; dynamically selecting a test point according to the double-field coupling weight coefficient of each section; monitoring the test point by using a movable wireless sensor node, and collecting the structural response data of the test point; constructing a sensitivity matrix by using initial excitation parameters and the structural response data, obtaining a modified objective function, and optimizing and modifying the objective function through iteration to obtain optimal excitation parameters; predicting the maximum dynamic load force and the actual dynamic load force; judging whether to perform early warning according to the predicted maximum dynamic load force and the predicted actual dynamic load force; and the accuracy and efficiency of the bridge load test are significantly improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent technology, specifically to a bridge load testing and detection system based on wireless transmission. Background Technology

[0002] Existing bridge load testing systems mostly rely on fixed sensor deployment at typical locations recommended by design specifications (such as mid-span, bearings, bridge deck, bridge abutments, and bridge tail). This approach has drawbacks, as it mainly depends on engineers selecting key points based on experience or design standards. This method cannot reflect the dynamic response characteristics of the entire bridge structure. Furthermore, most existing testing methods use known static loads or standard vehicle loads for excitation, but these loads cannot fully represent the complex dynamic load conditions faced by bridges in actual operation.

[0003] In view of this, the present invention proposes a bridge load testing and detection system based on wireless transmission to solve the above problems. Summary of the Invention

[0004] To overcome the aforementioned deficiencies of the prior art and to achieve the above objectives, the present invention provides the following technical solution: a bridge load testing and detection system based on wireless transmission, comprising:

[0005] Test point selection unit: The bridge is uniformly distributed into M sections, and the dual-field coupling weight coefficient of each section is calculated; the test points are dynamically selected based on the dual-field coupling weight coefficient of each section.

[0006] The dual-field coupling weight coefficients are obtained by: based on the strain field data measured in real time by the distributed strain sensor array, constructing the strain field tensor by calculating the partial derivatives of the strain field in the horizontal and vertical directions, and using the Frobenius norm to calculate the magnitude of the strain field gradient.

[0007] Cross-sectional vibration acceleration data are collected using a mobile excitation device. The vibration acceleration response is decomposed into a superposition of various modes using modal analysis technology. The energy contribution of each mode is separated using the QR decomposition method. The spatially distributed dual-field coupling weight coefficient is defined by combining the amplitude of the strain field gradient with the total mode contribution.

[0008] Wireless sensor node deployment unit: Uses mobile wireless sensor nodes to monitor test points and collect structural response data of the test points;

[0009] Dynamic excitation optimization unit: Constructs a sensitivity matrix using initial excitation parameters and structural response data, establishes an objective function, corrects the objective function based on the sensitivity matrix, and iteratively optimizes and corrects the objective function to finally determine the optimal excitation parameters;

[0010] Maximum dynamic load prediction unit: Based on the optimal excitation parameters of the test point, the corresponding structural response data, and the material feedback data of the section, the maximum dynamic load is predicted;

[0011] Actual dynamic load prediction unit: The actual dynamic load predicted based on the excitation parameters at the test points, structural response data, and environmental data;

[0012] Bridge load early warning unit: It determines whether to issue an early warning based on the predicted maximum dynamic load and the predicted actual dynamic load.

[0013] Furthermore, the specific method for uniformly distributing the bridge into M sections and calculating the dual-field coupling weighting coefficient for each section includes:

[0014] Strain field data measured in real time by a distributed strain sensor array;

[0015] The strain field tensor is obtained by calculating the partial derivatives of the strain field data in the horizontal and vertical directions.

[0016] The magnitude of the strain field gradient is calculated using the Frobenius norm for the strain field tensor.

[0017] Cross-sectional vibration acceleration data collected by a mobile excitation device; the vibration acceleration response data is decomposed into a superposition of N mode shapes using modal analysis technology, including the i-th mode shape vector and the modal coordinates of the i-th mode shape function;

[0018] The QR decomposition method is used to perform modal separation on the vibration acceleration data. The energy contribution of the Nth mode is calculated by the transpose of the i-th mode vector, the symmetry matrix, and the cross-sectional vibration acceleration data.

[0019] The total mode contribution is obtained by combining the energy contribution of the Nth mode and the Nth mode vector;

[0020] The amplitude of the strain field gradient and the contribution of the total vibration mode are fused to obtain the dual-field coupling weighting coefficient.

[0021] Furthermore, the specific methods for using mobile wireless sensor nodes to monitor test points and collect structural response data at the test points include:

[0022] The obtained dual-field coupling weight coefficients of the M cross sections are sorted in descending order, and the cross sections corresponding to the top 10% of the dual-field coupling weight coefficients are set as mandatory test points.

[0023] For the required test points, mobile wireless sensor nodes are used to monitor the test points and collect structural response data.

[0024] Furthermore, the structural response data includes deflection data, vertical acceleration data, and material feedback data;

[0025] Material feedback data includes material damage value, material strength, material corrosivity, material elastic modulus, and material fracture toughness.

[0026] Furthermore, the deflection data is acquired through the following methods:

[0027] Step a1: Selection and marking of target feature points on the bridge; the mobile wireless sensor nodes on the bridge are used as target feature points;

[0028] Step a2: Using a high-resolution camera, fix the camera to a stable reference point outside the bridge;

[0029] Step a3: Record the motion video of the bridge target feature points in real time;

[0030] Step a4: Denoise and enhance the video images in the acquired running video to obtain preprocessed video image data;

[0031] Step a5: For the first frame of preprocessed video image data, the Hough transform detection method is used to detect the marked target feature points, and the target tracking algorithm is used to obtain the motion trajectory of the feature points from the first frame to the i-th frame;

[0032] Step a6: Convert the two-dimensional pixel coordinates of the target feature points in the preprocessed video image into three-dimensional coordinates using a perspective projection model;

[0033] Step a7: Based on the three-dimensional coordinates of feature points on the motion trajectory, calculate the difference between the ordinate of each preprocessed video image frame and the ordinate of the first preprocessed video image data, which is the deflection data.

[0034] Furthermore, the specific method for constructing a sensitivity matrix using initial excitation parameters and structural response data, establishing an objective function, refining the objective function based on the sensitivity matrix, and ultimately determining the optimal excitation parameters through iterative optimization includes:

[0035] Step b1: Set initial excitation parameters; where the excitation parameters include excitation frequency and excitation mass, where the excitation frequency and excitation mass are the driving speed and load of the heavy vehicle;

[0036] Step b2: Retrieve structural response data from the test points;

[0037] Step b3: Using the structural response data from the test points, establish an initial sensitivity matrix between the initial excitation parameters and the structural response data;

[0038] Step b4, construct the objective function; where the objective function is represented as the sum of two parts:

[0039] Part 1: Error Term, which calculates the error between the structural response data at the test points and the standard structural response data;

[0040] Part Two: Penalty Items, which use adjustment coefficients to regulate the magnitude of material damage values;

[0041] Step b5: Based on the transpose of the initial sensitivity matrix, modify the objective function using the structural response data, standard structural response data, and material damage values;

[0042] Step b6: Repeat the iteration for the modified objective function until the minimum modified objective function is obtained, and then output the optimal excitation parameters.

[0043] Furthermore, the method for obtaining the material damage value includes:

[0044] Step c1: Establishing the reference temperature field: Obtain the bridge height at the test point; obtain the standard parameters of the materials, including density, specific heat capacity, and thermal conductivity; obtain solar radiation; obtain the ambient temperature and wind speed;

[0045] The temperature field coefficient outside the bridge is calculated based on solar radiation, wind speed, and ambient temperature, using the formula: T sur (x,y,t)=F(T amb Q sun ,h), where T sur T represents the temperature field coefficient outside the bridge. amb Q represents the ambient temperature. sun The standard solar radiation is given by h, the convective heat transfer coefficient, a constant calculated from wind speed and surface roughness, and F, a linear function, with the formula: h = C * v. n Where C represents surface roughness, v represents wind speed, and n is a constant obtained through experiments, with a value range of [0,1].

[0046] The temperature field coefficient inside the bridge is calculated based on the material's thermal conductivity, specific heat capacity, and density. The formula is: T int (z,t)=T sur *F(md,br,H,k), where md represents density, br represents specific heat capacity, H represents the height of the bridge, k represents thermal conductivity, and T... int This represents the temperature field coefficient inside the bridge, and z is the specific depth location within the bridge height.

[0047] The reference temperature field is obtained by using a weighted average method to calculate the temperature field coefficients outside and inside the bridge.

[0048] Step c2: Measured temperature field coefficients: The actual temperature field coefficients of the bridge cross section are collected using infrared thermal imaging technology;

[0049] Step c3: Calculate the difference temperature field between the standard temperature field coefficient and the measured temperature field coefficient;

[0050] Step c4: Correct the material standard parameters based on the temperature difference field to obtain the material damage value.

[0051] Furthermore, the specific method for predicting the maximum dynamic load force based on the optimal excitation parameters at the test point, the corresponding structural response data, and the material feedback data of the cross section includes:

[0052] The optimal excitation parameters are applied over a period of time across the cross section, rather than at a single point in time.

[0053] Based on the optimal excitation parameters, retrieve the structural response data and material feedback data of the test point within the time period corresponding to the optimal excitation parameters.

[0054] Time-domain feature extraction, frequency-domain feature extraction, and Lyapunov exponent extraction were performed on the structural response data and material feedback data at the test point during the time period.

[0055] The structural response data and material feedback data of the corresponding time period, as well as the extracted time-domain feature data, frequency-domain feature data, and Lyapunov exponent, are input into the SVR model to predict the maximum dynamic load of the bridge in the time period corresponding to the optimal excitation parameters.

[0056] Furthermore, the specific methods for predicting the actual dynamic load force based on the excitation parameters at the test points, structural response data, and environmental data include:

[0057] The excitation parameters, structural response data, and environmental data from the test points are retrieved and timestamp aligned. The aligned actual dataset is then used as input to the trained SVR model to predict the actual dynamic load.

[0058] Furthermore, the specific methods for determining whether to issue an early warning based on the predicted maximum dynamic load force and the predicted actual dynamic load force include:

[0059] The predicted dynamic load capacity is compared with the maximum dynamic load capacity of the bridge. If the dynamic load capacity is greater than the maximum dynamic load capacity of the bridge, an early warning is issued to control the number of vehicles passing through the bridge.

[0060] The technical effects and advantages of the bridge load testing and detection system based on wireless transmission of the present invention are as follows:

[0061] This invention automatically identifies key test points by using the dual-field coupling weighting coefficients of strain field and vibration modal field, avoiding the one-sided point selection problem caused by relying on manual experience in traditional methods, and achieving a more comprehensive reflection of the overall structural state of the bridge.

[0062] The introduction of mobile wireless sensor nodes eliminates the need for extensive wiring, adapts to different bridge structure types, offers greater deployment flexibility, and reduces maintenance costs.

[0063] By using high-resolution cameras and computer vision techniques (such as Hough transform and target tracking algorithms), non-contact measurement of bridge deflection data was achieved. Compared to traditional contact displacement sensors, this method does not require direct contact with the bridge surface, reducing interference and damage to the bridge and avoiding wear and tear issues caused by prolonged use.

[0064] By iteratively optimizing the structural response data and the objective function, the optimal solution of the excitation load parameters (such as vehicle speed and mass) is achieved, which improves the effectiveness and relevance of bridge response testing.

[0065] By comparing the reference temperature field with the measured temperature field, the material damage value can be calculated, breaking through the limitations of traditional manual inspection and destructive testing. It is non-contact, highly efficient, and highly accurate.

[0066] By integrating multi-dimensional features such as time domain, frequency domain, and Lyapunov exponent, and combining them with the SVR intelligent model, accurate prediction of maximum dynamic load force and actual dynamic load force can be achieved.

[0067] By comparing the actual load with the maximum dynamic load, timely warning instructions are issued to regulate traffic flow, thus providing active safety protection. Attached Figure Description

[0068] Figure 1 This is a schematic diagram of a bridge load testing and detection system based on wireless transmission according to the present invention;

[0069] Figure 2 This is a schematic diagram of the test point selection method of the present invention;

[0070] Figure 3 This is a schematic diagram of a bridge load testing method based on wireless transmission according to the present invention. Detailed Implementation

[0071] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0072] Example 1

[0073] Please see Figure 1 and Figure 2 As shown in the figure, this embodiment of a bridge load testing and detection system based on wireless transmission includes:

[0074] Test point selection unit: The bridge is uniformly distributed into M sections, and the dual-field coupling weight coefficient of each section is calculated; the test points are dynamically selected based on the dual-field coupling weight coefficient of each section.

[0075] The dual-field coupling weight coefficients are obtained by: based on the strain field data measured in real time by the distributed strain sensor array, constructing the strain field tensor by calculating the partial derivatives of the strain field in the horizontal and vertical directions, and using the Frobenius norm to calculate the magnitude of the strain field gradient.

[0076] Cross-sectional vibration acceleration data are collected using a mobile excitation device. The vibration acceleration response is decomposed into a superposition of various modes using modal analysis technology. The energy contribution of each mode is separated using the QR decomposition method. The spatially distributed dual-field coupling weight coefficient is defined by combining the amplitude of the strain field gradient with the total mode contribution.

[0077] Wireless sensor node deployment unit: Uses mobile wireless sensor nodes to monitor test points and collect structural response data of the test points;

[0078] Dynamic excitation optimization unit: Constructs a sensitivity matrix using initial excitation parameters and structural response data, establishes an objective function, corrects the objective function based on the sensitivity matrix, and iteratively optimizes and corrects the objective function to finally determine the optimal excitation parameters;

[0079] Maximum dynamic load prediction unit: Based on the optimal excitation parameters of the test point, the corresponding structural response data, and the material feedback data of the section, the maximum dynamic load is predicted;

[0080] Actual dynamic load prediction unit: The actual dynamic load predicted based on the excitation parameters at the test points, structural response data, and environmental data;

[0081] Bridge load early warning unit: It determines whether to issue an early warning based on the predicted maximum dynamic load and the predicted actual dynamic load.

[0082] The strain field data of a bridge refers to the strain generated inside the structure of a bridge when it is subjected to external loads. The spatial distribution of these strains forms the strain field.

[0083] The specific methods for uniformly distributing the bridge into M sections and calculating the two-field coupling weighting coefficients for each section include:

[0084] Strain field data measured in real time by a distributed strain sensor array;

[0085] The strain field tensor is obtained by calculating the partial derivatives of the strain field data along the horizontal and vertical axes, as shown in the formula: Where, ∈ x and ∈ y These are the components of the strain field data in the x and y directions, respectively. x =x 2 +y 2 ,∈ y =2xy, This represents the strain gradient along the x-axis. This represents the rate of change of the strain field data in the x-axis direction along the y-axis. This represents the strain gradient along the y-axis. This represents the rate of change of the strain field data in the y-axis direction in the x-axis direction;

[0086] The strain field tensor is calculated using the Frobenius norm, and the magnitude of the strain field gradient is obtained using the following formula: The magnitude of the strain field gradient is an important physical quantity describing the degree of strain change and is used to assess the degree of local deformation of a structure.

[0087] Cross-sectional vibration acceleration data collected by a mobile excitation device; the vibration acceleration response data is decomposed into a superposition of N mode shapes using modal analysis technology, as shown in the formula: Where Φ i Let q represent the vector of the i-th mode shape. i u(t) represents the modal coordinates of the i-th order mode shape function, u(t) represents the vibration acceleration data, t represents the time point, N represents the total order, and i represents the order index.

[0088] The QR decomposition method was used to perform modal separation on the vibration acceleration data, and the energy contribution of the Nth mode was calculated using the following formula: Where, η i This represents the energy contribution of the i-th mode shape. Let represent the transpose of the i-th mode shape vector, M be a symmetric matrix, and u(t) represent the acceleration response data at time t; Let represent the norm of the inner product of the i-th mode shape vector, the symmetry matrix, and the vibration acceleration data. A larger norm indicates a greater contribution of the i-th mode shape vector to the vibration acceleration data. This represents the sum of contributions from all mode shapes, where j is the order index;

[0089] The total mode contribution is obtained by combining the energy contribution of the mode shape and the mode shape vector, using the following formula: Where TC represents the total mode shape contribution;

[0090] The amplitude of the strain field gradient and the contribution of the total vibration mode are combined to obtain the dual-field coupling weighting coefficient, as shown in the formula: Where α and β are the magnitudes of the strain field gradient and the weights of the total mode shape contribution, respectively, and ρ is the weighting coefficient of the two-field coupling.

[0091] Among them, the distributed strain sensor array and the mobile excitation device are mobile and can traverse every section of the bridge;

[0092] In bridge load tests, it is necessary to monitor the stress and vibration characteristics of the bridge in order to determine whether it is safe and where it is prone to problems. To do this, sensors are placed on the bridge to collect data and analyze it. However, the problem is that it is impossible to install sensors everywhere on the bridge, as that would be too costly. Therefore, sensors cannot be placed randomly; they must be placed in the most critical locations, such as the places most prone to problems or the places that have the greatest impact on the safety of the bridge.

[0093] The amplitude of the strain field gradient and the contribution of the total vibration mode are combined because the safety of a bridge is related to both its static stress and its dynamic vibration. If only one is considered, some key information will be ignored. Therefore, these two pieces of information are combined to form a more comprehensive indicator.

[0094] The specific methods for using mobile wireless sensor nodes to monitor test points and collect structural response data at the test points include:

[0095] The obtained dual-field coupling weight coefficients of the M cross sections are sorted in descending order, and the cross sections corresponding to the top 10% of the dual-field coupling weight coefficients are set as mandatory test points; mobile wireless sensor nodes (such as magnetic installation method) are used to monitor the test points and collect the structural response data of the test points;

[0096] Among them, the cross sections corresponding to the top 10% of the dual-field coupling weight coefficients are selected as mandatory test points because the dual-field coupling weight coefficients may fluctuate due to local noise or error. Selecting the top 10% can avoid the random interference of a single peak.

[0097] The existing method is to measure key parts of the entire bridge, such as the middle connection, the connection between the bridgehead and the bridge tail, and install sensors at these key parts. The deployment of sensors requires a lot of manpower and equipment to install and fix the sensors, and the wiring is complicated. Regular maintenance of the sensors requires closing the bridge, which affects normal traffic.

[0098] Mobile wireless sensor nodes eliminate the need for complex wiring, reducing installation and maintenance costs, and will not disrupt normal traffic on the bridge when maintenance is required.

[0099] Structural response data includes deflection data, vertical acceleration data, and material feedback data; among which, the vertical acceleration data is obtained by measuring a piezoelectric accelerometer.

[0100] Material feedback data includes material damage value, material strength, material corrosivity, material elastic modulus, and material fracture toughness;

[0101] Among them, material strength is acquired through force sensors, material corrosion is acquired through electrochemical sensors, material elastic modulus is acquired through fiber optic gratings, and material fracture toughness is acquired through acoustic emission sensors; all of these sensors are deployed on mobile wireless sensor nodes.

[0102] Existing deflection data is acquired using displacement sensors, which are typically contact sensors. This can lead to some interference, such as friction or vibration generated when the sensor is in contact with the surface, which can affect measurement accuracy. Furthermore, the sensor is affected by structural deformation, especially on bridges under heavy stress, where the sensor itself may undergo slight deformation, affecting the measurement results.

[0103] Methods for obtaining deflection data include:

[0104] Step a1: Selection and marking of target feature points on the bridge; the mobile wireless sensor nodes on the bridge are used as target feature points;

[0105] Step a2: Using a high-resolution camera, fix the camera to a stable reference point outside the bridge;

[0106] Step a3: Record the motion video of the bridge target feature points in real time;

[0107] Step a4: Denoise and enhance the video images in the acquired running video to obtain preprocessed video image data;

[0108] Step a5: For the first frame of preprocessed video image data, the Hough transform detection method is used to detect the marked target feature points, and the target tracking algorithm is used to obtain the motion trajectory of the feature points from the first frame to the i-th frame;

[0109] Step a6: Convert the two-dimensional pixel coordinates of the target feature points in the preprocessed video image into three-dimensional coordinates using a perspective projection model;

[0110] Step a7: Based on the three-dimensional coordinates of feature points on the motion trajectory, calculate the difference between the ordinate on each frame of the preprocessed video image and the ordinate in the first frame of the preprocessed video image data, which is the deflection data;

[0111] By monitoring the movement of target feature points on the bridge in real time using high-resolution cameras, there is no need to install any physical sensors on the bridge structure to measure deflection data; this avoids problems such as the installation, maintenance, and force interference of contact sensors.

[0112] Furthermore, video monitoring technology does not affect the structure of the bridge itself, avoiding the errors and structural damage that may be caused by traditional sensors.

[0113] The specific methods for constructing a sensitivity matrix using initial excitation parameters and structural response data, establishing an objective function, refining the objective function based on the sensitivity matrix, and iteratively optimizing the objective function to finally determine the optimal excitation parameters include:

[0114] Step b1: Set initial excitation parameters; where the excitation parameters include excitation frequency and excitation mass, which are the driving speed and load of the heavy vehicle; obtain the effective range of excitation parameters according to relevant materials; the initial excitation parameters are a combination of parameters randomly selected within the effective range of the excitation parameters using a random method.

[0115] Step b2: Retrieve structural response data from the test points; monitor the structural response data of the bridge test points under different initial excitation parameters over a long period of time and store it in the database;

[0116] Step b3: Using the structural response data from the test points, establish an initial sensitivity matrix between the initial excitation parameters and the structural response data. Each row represents the sensitivity of a structural response data point to an excitation parameter at the same time point. A higher sensitivity indicates a more significant impact of the excitation parameter on the structural response data. S is the sensitivity matrix of the excitation parameter and the structural response data; the sensitivity matrix is ​​calculated using the finite difference method to obtain the partial derivatives of the initial excitation parameter and the structural response data.

[0117] For example, at a certain time, the initial excitation parameters of the test point are (ja, jb), and the structural response data of the test point at this time are (j1, j2, j3, j4, j5, j6, j7);

[0118] Step b4, construct the objective function; where the objective function is represented as the sum of two parts:

[0119] Part 1: Error Term, which calculates the error between the structural response data at the test points and the standard structural response data;

[0120] Part Two: Penalty Term, which uses an adjustment coefficient to regulate the magnitude of material damage value; typically, when the material damage value is too large, the penalty term will increase, thereby increasing the value of the objective function;

[0121] Step b5: Based on the transpose of the initial sensitivity matrix, the objective function is modified using the structural response data, standard structural response data, and material damage values. The formula is: F new =2*S T *|Y cur -Y str |+μ*ss, where F new For the corrected objective function, S T To initialize the transpose of the sensitivity matrix, Y cur Represents structural response data, Y str This represents standard structural response data (obtained from the Bridge Construction Manual), where ss is the material damage value and v is the adjustment coefficient;

[0122] Step b6: Repeat the iteration for the modified objective function until the minimum modified objective function is obtained, and then output the optimal excitation parameters; the calculation formula for the iteration is: Where, P(f,m) k Let f represent the activation parameters for the k-th iteration, m represent the activation quality, and τ represent the learning rate. To correct the gradient of the objective function with respect to the excitation parameters, P(f,m) k+1 This represents the stimulus parameter for the (k+1)th fall, where k is the iteration index.

[0123] Methods for obtaining material damage values ​​include:

[0124] Step c1: Establishing the reference temperature field: Obtain the bridge height at the test point; obtain material parameters, including density, specific heat capacity, and thermal conductivity; obtain solar radiation; obtain ambient temperature and wind speed;

[0125] The height of the bridge refers to the distance between the surface of the bridge road and the area below the road surface, which can be obtained directly from the bridge design drawings; density, specific heat capacity, and thermal conductivity refer to the density, specific heat capacity, and thermal conductivity of the material used to form the bridge road surface, which can be obtained from the material handbook; solar radiation is collected by a radiation sensor; ambient temperature and wind speed are obtained by a temperature sensor and anemometer; the radiation sensor, temperature sensor, and anemometer can be installed on a mobile wireless sensor node.

[0126] The temperature field coefficient outside the bridge is calculated based on solar radiation, wind speed, and ambient temperature, using the formula: T sur (x,y,t)=F(T amb Q sun ,h), where T sur T represents the temperature field coefficient outside the bridge. amb Q represents the ambient temperature. sunThe standard solar radiation is given by h, the convective heat transfer coefficient, a constant calculated from wind speed and surface roughness, and F, a linear function, with the formula: h = C * v. n Where C represents surface roughness, obtained from a material handbook, v represents wind speed, and n is a constant obtained through experiments, with a value range of [0,1].

[0127] The temperature field coefficient inside the bridge is calculated based on the material's thermal conductivity, specific heat capacity, and density. The formula is: T int (z,t)=T sur *F(md,br,H,k), where md represents density, br represents specific heat capacity, H represents the height of the bridge, k represents thermal conductivity, and T... int This represents the temperature field coefficient inside the bridge, and z is the specific depth location within the bridge height.

[0128] The reference temperature field is obtained by using a weighted average method to calculate the temperature field coefficients outside and inside the bridge.

[0129] Step c2: Measured temperature field coefficients: The actual temperature field coefficients of the bridge cross section are collected using infrared thermal imaging technology;

[0130] Step c3: Calculate the difference temperature field between the standard temperature field coefficient and the measured temperature field coefficient;

[0131] Step c4: Correct the material standard parameters based on the temperature difference field to obtain the material damage value;

[0132] Temperature changes are a crucial factor affecting material properties and structural health, especially for structures like bridges during long-term use and aging. Temperature not only affects the surface of bridges but also causes changes in the internal temperature of the structure. Therefore, calculations based on temperature fields can more comprehensively reflect the impact of temperature on materials.

[0133] Differences in temperature fields can reflect signs of material aging, degradation, or localized damage. Calculating material damage values ​​can more accurately assess changes in bridge materials during actual use. For example, thermal expansion and contraction of materials in high or low temperature environments can even lead to the appearance of microcracks, all of which can be reflected through differences in temperature fields.

[0134] This method allows for dynamic correction of bridge material parameters, adjusting material properties under different environments and time periods to reflect the structural health status in real time and identify potential material damage values ​​in advance.

[0135] One of the biggest drawbacks of using the maximum dynamic load as a fixed value in existing methods is that it does not take into account the changes in the actual condition of the bridge. As the bridge is used for a longer period of time, it may experience fatigue, cracks, corrosion, material aging and other phenomena, all of which affect the load-bearing capacity of the bridge. If calculations are based solely on standard data, it is impossible to reflect the actual condition of the bridge materials and structure and their changes over time.

[0136] The specific methods for predicting the maximum dynamic load based on the optimal excitation parameters at the test point, the corresponding structural response data, and the material feedback data of the cross section include:

[0137] The optimal excitation parameters pass through a cross section corresponding to the test point over a period of time, not just in one second.

[0138] Based on the optimal excitation parameters, retrieve the structural response data and material feedback data of the test point within the time period corresponding to the optimal excitation parameters.

[0139] Time-domain feature extraction, frequency-domain feature extraction, and Lyapunov exponent extraction were performed on the structural response data and material feedback data at the test point during the time period.

[0140] The structural response data and material feedback data of the corresponding time period, as well as the extracted time-domain feature data, frequency-domain feature data and Lyapunov exponent, are input into the SVR model to predict the maximum bridge dynamic load in the time period corresponding to the optimal excitation parameters.

[0141] The time period corresponding to the optimal excitation parameters takes into account the dynamic characteristics of the bridge response, rather than a single instantaneous value. The structural response and material properties of the bridge are affected by a variety of factors, especially changes in environmental factors. Therefore, selecting a time period for data analysis can better reflect the true dynamic response of the structure in a variable environment.

[0142] Time-domain features can reflect direct information about how a signal changes over time, such as maximum displacement, root mean square value, and pulse duration; these features can capture nonlinear changes in structural response, especially the influence of dynamic excitation on structural response.

[0143] Frequency domain characteristics can reveal the frequency characteristics of the structural response, such as natural frequency and resonance peak, which helps to analyze the response of the bridge under specific frequency excitation and is of great significance for predicting the timing and intensity of the maximum dynamic load.

[0144] The Lyapunov index is an indicator for measuring the stability and chaos of a system and can effectively reflect the stability state of a bridge. Introducing the Lyapunov index can help identify chaotic vibration phenomena that may occur in bridges under high dynamic loads, thus taking these extreme dynamic responses into account when predicting the maximum dynamic load.

[0145] The specific methods for predicting the actual dynamic load force based on test point excitation parameters, structural response data, and environmental data include:

[0146] Retrieve excitation parameters, structural response data, and environmental data from the test points; align the timestamps; obtain the aligned actual dataset; use the actual dataset as input to the trained SVR model to predict the actual dynamic load force;

[0147] Among them, the excitation parameters, structural response data, and environmental data at the test points are real-time excitation parameters, structural response data, and environmental data collected in real time.

[0148] Specific methods for determining whether to issue an early warning based on the predicted maximum dynamic load and the predicted actual dynamic load include:

[0149] The predicted dynamic load capacity is compared with the maximum dynamic load capacity of the bridge. If the dynamic load capacity is greater than the maximum dynamic load capacity of the bridge, an early warning is issued to control the number of vehicles passing through the bridge.

[0150] In this embodiment, key test points are automatically identified by the dual-field coupling weighting coefficient of strain field and vibration modal field, avoiding the one-sided point selection problem caused by relying on human experience in traditional methods, and realizing a more comprehensive reflection of the overall structural state of the bridge.

[0151] The introduction of mobile wireless sensor nodes eliminates the need for extensive wiring, adapts to different bridge structure types, offers greater deployment flexibility, and reduces maintenance costs.

[0152] By using high-resolution cameras and computer vision techniques (such as Hough transform and target tracking algorithms), non-contact measurement of bridge deflection data was achieved. Compared to traditional contact displacement sensors, this method does not require direct contact with the bridge surface, reducing interference and damage to the bridge and avoiding wear and tear issues caused by prolonged use.

[0153] By iteratively optimizing the structural response data and the objective function, the optimal solution of the excitation load parameters (such as vehicle speed and mass) is achieved, which improves the effectiveness and relevance of bridge response testing.

[0154] By comparing the reference temperature field with the measured temperature field, the material damage value can be calculated, breaking through the limitations of traditional manual inspection and destructive testing. It is non-contact, highly efficient, and highly accurate.

[0155] By integrating multi-dimensional features such as time domain, frequency domain, and Lyapunov exponent, and combining them with the SVR intelligent model, accurate prediction of maximum dynamic load force and actual dynamic load force can be achieved.

[0156] By comparing the actual load with the maximum dynamic load, timely warning instructions are issued to regulate traffic flow, thus providing active safety protection.

[0157] Example 2

[0158] Please see Figure 3 As shown, parts not described in detail in this embodiment are described in Embodiment 1. A bridge load testing method based on wireless transmission is provided, including:

[0159] Step SS1: Divide the bridge into M sections evenly and calculate the dual-field coupling weighting coefficient for each section; dynamically select test points based on the dual-field coupling weighting coefficient for each section.

[0160] Step SS2: Use mobile wireless sensor nodes to monitor the test points and collect structural response data of the test points;

[0161] Step SS3: Construct a sensitivity matrix using the initial excitation parameters and structural response data, establish an objective function, modify the objective function based on the sensitivity matrix, and iteratively optimize the modified objective function to finally determine the optimal excitation parameters;

[0162] Step SS4: Based on the optimal excitation parameters at the test point, the corresponding structural response data, and the material feedback data of the cross section, predict the maximum dynamic load force.

[0163] Step SS5: Based on the excitation parameters at the test points, structural response data, and environmental data, predict the actual dynamic load force;

[0164] Step SS6: Determine whether to issue an early warning based on the predicted maximum dynamic load and the predicted actual dynamic load.

[0165] Example 3

[0166] This embodiment discloses an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the operation mode of the bridge load testing and detection system based on wireless transmission provided above.

[0167] Since the electronic device described in this embodiment is the electronic device used in implementing the wireless transmission-based bridge load testing system described in this application, those skilled in the art can understand the specific implementation methods and various variations of the electronic device in this embodiment based on the wireless transmission-based bridge load testing system described in this application. Therefore, how the electronic device implements the method in this application will not be described in detail here. Any electronic device used by those skilled in the art in implementing the wireless transmission-based bridge load testing system described in this application falls within the scope of protection of this application.

[0168] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters and thresholds in the formulas are set by those skilled in the art according to the actual situation.

[0169] The above description is merely a preferred embodiment of the present invention, and the scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for users of ordinary technical skills, any improvements and modifications made without departing from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A bridge load testing and detection system based on wireless transmission, characterized in that, include: Test point selection unit: The bridge is uniformly distributed into M sections, and the dual-field coupling weighting coefficient of each section is calculated; The test points are dynamically selected based on the dual-field coupling weighting coefficients of each cross section; The dual-field coupling weight coefficients are obtained by: based on the strain field data measured in real time by the distributed strain sensor array, constructing the strain field tensor by calculating the partial derivatives of the strain field in the horizontal and vertical directions, and using the Frobenius norm to calculate the magnitude of the strain field gradient. Cross-sectional vibration acceleration data are collected using a mobile excitation device. The vibration acceleration response is decomposed into a superposition of various modes using modal analysis technology. The energy contribution of each mode is separated using the QR decomposition method. The spatially distributed dual-field coupling weight coefficient is defined by combining the amplitude of the strain field gradient with the total mode contribution. Wireless sensor node deployment unit: Uses mobile wireless sensor nodes to monitor test points and collect structural response data of the test points; Dynamic excitation optimization unit: Constructs a sensitivity matrix using initial excitation parameters and structural response data, establishes an objective function, corrects the objective function based on the sensitivity matrix, and iteratively optimizes and corrects the objective function to finally determine the optimal excitation parameters; Maximum dynamic load prediction unit: Based on the optimal excitation parameters of the test point, the corresponding structural response data, and the material feedback data of the section, the maximum dynamic load is predicted; Actual dynamic load prediction unit: The actual dynamic load predicted based on the excitation parameters at the test points, structural response data, and environmental data; Bridge load early warning unit: Determines whether to issue an early warning based on the predicted maximum dynamic load and the predicted actual dynamic load; The specific method for uniformly distributing the bridge into M sections and calculating the dual-field coupling weighting coefficient for each section includes: Strain field data measured in real time by a distributed strain sensor array; The strain field tensor is obtained by calculating the partial derivatives of the strain field data in the horizontal and vertical directions. The magnitude of the strain field gradient is calculated using the Frobenius norm for the strain field tensor. Cross-sectional vibration acceleration data collected by a mobile excitation device; the vibration acceleration response data is decomposed into a superposition of N mode shapes using modal analysis technology, including the i-th mode shape vector and the modal coordinates of the i-th mode shape function; The QR decomposition method is used to perform modal separation on the vibration acceleration data. The energy contribution of the Nth mode is calculated by the transpose of the i-th mode vector, the symmetry matrix, and the cross-sectional vibration acceleration data. The total mode contribution is obtained by combining the energy contribution of the Nth mode and the Nth mode vector; The amplitude of the strain field gradient and the contribution of the total vibration mode are fused to obtain the dual-field coupling weighting coefficient.

2. The bridge load testing and detection system based on wireless transmission according to claim 1, characterized in that, The specific methods for using mobile wireless sensor nodes to monitor test points and collect structural response data at the test points include: The obtained dual-field coupling weight coefficients of the M cross sections are sorted in descending order, and the cross sections corresponding to the top 10% of the dual-field coupling weight coefficients are set as mandatory test points. For the required test points, mobile wireless sensor nodes are used to monitor the test points and collect structural response data.

3. The bridge load testing and detection system based on wireless transmission according to claim 2, characterized in that, The structural response data includes deflection data, vertical acceleration data, and material feedback data; Material feedback data includes material damage value, material strength, material corrosivity, material elastic modulus, and material fracture toughness.

4. The bridge load testing and detection system based on wireless transmission according to claim 3, characterized in that, The deflection data is obtained in the following ways: Step a1: Selection and marking of target feature points on the bridge; the mobile wireless sensor nodes on the bridge are used as target feature points; Step a2: Using a high-resolution camera, fix the camera to a stable reference point outside the bridge; Step a3: Record the motion video of the bridge target feature points in real time; Step a4: Denoise and enhance the video images in the acquired running video to obtain preprocessed video image data; Step a5: For the first frame of preprocessed video image data, the Hough transform detection method is used to detect the marked target feature points, and the target tracking algorithm is used to obtain the motion trajectory of the feature points from the first frame to the i-th frame; Step a6: Convert the two-dimensional pixel coordinates of the target feature points in the preprocessed video image into three-dimensional coordinates using a perspective projection model; Step a7: Based on the three-dimensional coordinates of feature points on the motion trajectory, calculate the difference between the ordinate of each preprocessed video image frame and the ordinate of the first preprocessed video image data, which is the deflection data.

5. The bridge load testing and detection system based on wireless transmission according to claim 4, characterized in that, The specific method for constructing a sensitivity matrix using initial excitation parameters and structural response data, establishing an objective function, refining the objective function based on the sensitivity matrix, and iteratively optimizing the refined objective function to finally determine the optimal excitation parameters includes: Step b1: Set initial excitation parameters; where the excitation parameters include excitation frequency and excitation mass, where the excitation frequency and excitation mass are the driving speed and load of the heavy vehicle; Step b2: Retrieve structural response data from the test points; Step b3: Using the structural response data from the test points, establish an initial sensitivity matrix between the initial excitation parameters and the structural response data; Step b4, construct the objective function; where the objective function is represented as the sum of two parts: Part 1: Error Term, which calculates the error between the structural response data at the test points and the standard structural response data; Part Two: Penalty Items, which use adjustment coefficients to regulate the magnitude of material damage values; Step b5: Based on the transpose of the initial sensitivity matrix, modify the objective function using the structural response data, standard structural response data, and material damage values; Step b6: Repeat the iteration for the modified objective function until the minimum modified objective function is obtained, and then output the optimal excitation parameters.

6. The bridge load testing and detection system based on wireless transmission according to claim 5, characterized in that, The methods for obtaining the material damage value include: Step c1: Establishing the reference temperature field: Obtain the bridge height at the test point; obtain the standard parameters of the materials, including density, specific heat capacity, and thermal conductivity; obtain solar radiation; obtain the ambient temperature and wind speed; The temperature field coefficient outside the bridge is calculated based on solar radiation, wind speed, and ambient temperature. The temperature field coefficient inside the bridge is calculated based on the thermal conductivity, specific heat capacity, and density of the material. The reference temperature field is obtained by using a weighted average method to calculate the temperature field coefficients outside and inside the bridge. Step c2: Measured temperature field coefficients: The actual temperature field coefficients of the bridge cross section are collected using infrared thermal imaging technology; Step c3: Calculate the difference temperature field between the standard temperature field coefficient and the measured temperature field coefficient; Step c4: Correct the material standard parameters based on the temperature difference field to obtain the material damage value.

7. The bridge load testing and detection system based on wireless transmission according to claim 6, characterized in that, The specific methods for predicting the maximum dynamic load based on the optimal excitation parameters at the test point, the corresponding structural response data, and the material feedback data of the cross section include: Based on the optimal excitation parameters, retrieve the structural response data and material feedback data of the test point within the time period corresponding to the optimal excitation parameters. Time-domain feature extraction, frequency-domain feature extraction, and Lyapunov exponent extraction were performed on the structural response data and material feedback data at the test point during the time period. The structural response data and material feedback data of the corresponding time period, as well as the extracted time-domain feature data, frequency-domain feature data, and Lyapunov exponent, are input into the SVR model to predict the maximum dynamic load of the bridge in the time period corresponding to the optimal excitation parameters.

8. The bridge load testing and detection system based on wireless transmission according to claim 7, characterized in that, The specific methods for predicting the actual dynamic load force based on test point excitation parameters, structural response data, and environmental data include: The excitation parameters, structural response data, and environmental data from the test points are retrieved and timestamp aligned. The aligned actual dataset is then used as input to the trained SVR model to predict the actual dynamic load.

9. A bridge load testing and detection system based on wireless transmission according to claim 8, characterized in that, The specific methods for determining whether to issue an early warning based on the predicted maximum dynamic load and the predicted actual dynamic load include: The predicted dynamic load capacity is compared with the maximum dynamic load capacity of the bridge. If the dynamic load capacity is greater than the maximum dynamic load capacity of the bridge, an early warning is issued to control the number of vehicles passing through the bridge.

Images