Concrete bridge pile foundation risk detection method and system

Abstract

The application provides a concrete bridge pile foundation risk detection method and system, and relates to the field of bridge pile foundations.The method comprises the following steps: determining a plurality of risk detection points of a concrete bridge pile foundation to be detected and a risk correlation coefficient between the plurality of risk detection points based on structure information and service information of the concrete bridge pile foundation to be detected through finite element analysis; arranging a risk detection device at each risk detection point; determining a plurality of ultrasonic detection positions of the concrete bridge pile foundation to be detected based on point state data collected by the risk detection devices of the plurality of risk detection points and the risk correlation coefficient between the plurality of risk detection points, and generating an ultrasonic detection scheme; and performing ultrasonic detection on the concrete bridge pile foundation to be detected according to the ultrasonic detection scheme, obtaining ultrasonic detection data of the concrete bridge pile foundation to be detected, and performing concrete bridge pile foundation risk analysis, so that the nondestructive detection of the concrete bridge pile foundation risk is realized.

Description

Technical Field

[0001] The present invention relates to the field of bridge pile foundations, and in particular to a method and system for detecting risks of concrete bridge pile foundations. Background Art

[0002] The pile foundation of a bridge is the main load-bearing component of the bridge. During the use of the bridge, when the bridge is hit or carries too much weight, cracks will appear inside the pile foundation. The damage to the pile foundation is usually reflected inside the pile foundation. At this time, non-destructive testing of the pile foundation of the bridge is required.

[0003] Current pile foundation inspections typically utilize pre-buried acoustic testing tubes with acoustic transducers placed within them. The transducers detect cracks within the piles using parameters generated by sound waves passing through them. However, if the acoustic testing tubes become damaged, the transducers cannot penetrate them to inspect the piles, rendering bridge inspections impossible.

[0004] Therefore, it is necessary to provide a concrete bridge pile foundation risk detection method and system for realizing non-destructive detection of concrete bridge pile foundation risks. Summary of the Invention

[0005] The present invention provides a method for detecting risks of concrete bridge pile foundations, comprising: obtaining structural information and service information of a concrete bridge pile foundation to be detected; determining, through finite element analysis, a plurality of risk detection points of the concrete bridge pile foundation to be detected and a risk correlation coefficient between the plurality of risk detection points based on the structural information and service information of the concrete bridge pile foundation to be detected; providing a risk detection device at each risk detection point; determining a plurality of ultrasonic detection positions of the concrete bridge pile foundation to be detected based on detection point status data collected by the risk detection devices at the plurality of risk detection points and the risk correlation coefficients between the plurality of risk detection points; generating an ultrasonic detection scheme based on the plurality of ultrasonic detection positions of the concrete bridge pile foundation to be detected, wherein the ultrasonic detection scheme includes an ultrasonic detection path and ultrasonic detection parameters corresponding to each ultrasonic detection position; performing ultrasonic detection on the concrete bridge pile foundation to be detected according to the ultrasonic detection scheme to obtain ultrasonic detection data of the concrete bridge pile foundation to be detected; and performing a risk analysis of the concrete bridge pile foundation based on the ultrasonic detection data of the concrete bridge pile foundation to be detected.

[0006] Further, based on the structure information and service information of the concrete bridge pile to be detected, the risk detection points and the risk correlation coefficients between the risk detection points of the concrete bridge pile to be detected are determined through finite element analysis, including: establishing a geometric model of the concrete bridge pile to be detected based on the structure information of the concrete bridge pile to be detected; performing mesh division on the geometric model of the concrete bridge pile to be detected to obtain a plurality of finite element analysis units; assigning material properties to each of the finite element analysis units based on the structure information of the concrete bridge pile to be detected; determining boundary conditions; determining a plurality of test external loads based on the structure information and service information of the concrete bridge pile to be detected; for each test external load, solving to obtain test data of the concrete bridge pile to be detected under the test external load, wherein the test data includes simulated vibration data of the plurality of finite element analysis units; determining a plurality of initial detection points based on the test data of the concrete bridge pile to be detected under each test external load; determining point risk correlation coefficients of the plurality of initial detection points based on the test data of the concrete bridge pile to be detected under each test external load; and screening a plurality of risk detection points from the plurality of initial detection points based on the point risk correlation coefficients of the plurality of initial detection points, and determining risk correlation coefficients between the plurality of risk detection points.

[0007] Further, the service information includes traffic flow information and environmental information; based on the structure information and service information of the concrete bridge pile to be detected, a plurality of test external loads are determined, including: obtaining the structure information of the concrete bridge, determining the static load of the concrete bridge pile to be detected; predicting the traffic load range of the concrete bridge pile to be detected based on the traffic flow information through a first load prediction model; predicting the environmental load range of the concrete bridge pile to be detected based on the environmental information through a second load prediction model; and determining the plurality of test external loads based on the static load, the traffic load range and the environmental load range of the concrete bridge pile to be detected.

[0008] Further, based on the test data of the concrete bridge pile to be detected under each test external load, a plurality of initial detection points are determined, including: for each of the finite element analysis units, determining a unit risk parameter of the finite element analysis unit based on the test data of the finite element analysis unit under each test external load, and regarding the finite element analysis unit with a unit risk parameter greater than a unit risk parameter threshold as a risk finite element analysis unit; clustering a plurality of risk finite element analysis units to determine a plurality of unit clusters, and regarding the risk finite element analysis units corresponding to the cluster centers of the unit clusters as one initial detection point.

[0009] Further, based on the test data of the concrete bridge pile to be detected under each set of test external load, a point risk correlation coefficient of a plurality of initial detection points is determined, including: for each set of test external load, based on the test data of each initial detection point under the test external load, a load risk correlation coefficient of any two initial detection points under the test external load is calculated; and based on the load risk correlation coefficients of any two initial detection points under each set of test external load, a point risk correlation coefficient of any two initial detection points is calculated.

[0010] Further, based on the point risk correlation coefficients of the plurality of initial detection points, a plurality of risk detection points are screened from the plurality of initial detection points, and a risk correlation coefficient between the plurality of risk detection points is determined, including: based on the point risk correlation coefficients of any two initial detection points, a first point risk correlation coefficient sequence corresponding to each initial detection point is determined; for each initial detection point, based on the first point risk correlation coefficient sequence corresponding to the initial detection point, a point risk correlation coefficient mean and a point risk correlation coefficient standard deviation are calculated, and an initial detection point with a point risk correlation coefficient mean greater than a point risk correlation coefficient mean threshold and a point risk correlation coefficient standard deviation less than a point risk correlation coefficient standard deviation threshold is taken as a risk detection point; based on the point risk correlation coefficients of any two risk detection points, a second point risk correlation coefficient sequence corresponding to each risk detection point is determined; according to the second point risk correlation coefficient sequence corresponding to each risk detection point, a plurality of point clusters are clustered and determined; and for each point cluster, according to the second point risk correlation coefficient sequence corresponding to each risk detection point included in the point cluster, a risk correlation coefficient between any two risk detection points included in the point cluster is determined.

[0011] Further, based on the risk detection point state data collected by the risk detection device of the plurality of risk detection points and the risk correlation coefficients between the plurality of risk detection points, a plurality of ultrasonic detection positions of the concrete bridge pile to be detected are determined, including: for each point cluster, based on the test data of the concrete bridge pile to be detected under each set of test external load, a training sample corresponding to the point cluster is generated, an initial risk prediction model corresponding to the point cluster is established, the initial risk prediction model corresponding to the point cluster is trained using the training sample corresponding to the point cluster, a risk prediction model corresponding to the point cluster is obtained, and through the risk prediction model corresponding to the point cluster, based on the detection point state data collected by the risk detection device of the risk detection points included in the point cluster and the risk correlation coefficients between any two risk detection points included in the point cluster, an ultrasonic detection position corresponding to the point cluster and risk information corresponding to the ultrasonic detection position are predicted.

[0012] Furthermore, based on the multiple risk detection positions of the concrete bridge pile foundation to be inspected, a risk detection scheme is generated, including: generating a risk detection path through a path generation model based on the structural information of the concrete bridge pile foundation to be inspected and the multiple risk detection positions; and generating ultrasonic detection parameters corresponding to the risk detection position based on the risk information corresponding to the ultrasonic detection position through a parameter generation model.

[0013] Furthermore, based on the ultrasonic detection data of the concrete bridge pile foundation to be detected, a risk analysis of the concrete bridge pile foundation is performed, including: performing the risk analysis of the concrete bridge pile foundation based on the ultrasonic detection data of the concrete bridge pile foundation to be detected through a risk analysis model.

[0014] The present invention provides a concrete bridge pile foundation risk detection system for executing the above-mentioned concrete bridge pile foundation risk detection method, comprising: an information acquisition module for acquiring structural information and service information of the concrete bridge pile foundation to be detected; a risk monitoring module for determining, through finite element analysis, a plurality of risk detection points of the concrete bridge pile foundation to be detected and risk correlation coefficients between the plurality of risk detection points based on the structural information and service information of the concrete bridge pile foundation to be detected, and also for setting a risk detection device at each risk detection point; a risk prediction module for predicting the risk of the risk detection point status data and the risk correlation coefficients between the plurality of risk detection points based on the risk detection devices of the plurality of risk detection points. correlation coefficient, for determining multiple ultrasonic detection positions of the concrete bridge pile foundation to be detected; a scheme generating module, for generating an ultrasonic detection scheme based on the multiple ultrasonic detection positions of the concrete bridge pile foundation to be detected, wherein the ultrasonic detection scheme includes an ultrasonic detection path and ultrasonic detection parameters corresponding to each ultrasonic detection position; a risk monitoring module, for performing ultrasonic detection on the concrete bridge pile foundation to be detected according to the ultrasonic detection scheme, and obtaining ultrasonic detection data of the concrete bridge pile foundation to be detected; a risk analysis module, for performing concrete bridge pile foundation risk analysis based on the ultrasonic detection data of the concrete bridge pile foundation to be detected.

[0015] Compared with the existing technology, the concrete bridge pile foundation risk detection method and system provided by the present invention have at least the following beneficial effects:

[0016] 1. By acquiring structural and service information for the concrete bridge pile foundation to be inspected, a more comprehensive understanding of the actual pile foundation condition is achieved, providing a solid foundation for risk assessment. Finite element analysis, based on this collected information, accurately identifies multiple risk detection points and their associated risk correlation coefficients. This approach considers the complex structure of the pile foundation and the diverse influencing factors, enhancing the scientific nature and accuracy of the risk assessment. Installing risk detection devices at each risk detection point allows for focused monitoring of critical areas of the pile foundation, improving the targeted nature and efficiency of the inspection. Multiple ultrasonic inspection locations are determined based on the status data and risk correlation coefficients of the risk detection points, ensuring coverage of the key risk areas of the pile foundation while avoiding unnecessary testing. The generated ultrasonic inspection plan includes a specific inspection path and corresponding ultrasonic testing parameters for each inspection location, ensuring the accuracy and reliability of the test results. The inspection data generated based on the ultrasonic inspection plan comprehensively reflects the actual condition of the pile foundation, including detailed information on defects such as cracks and deformation. Risk analysis based on this inspection data accurately assesses the safety performance of the pile foundation, promptly identifies potential risk points, and provides a scientific basis for subsequent maintenance and reinforcement work.

[0017] 2. By establishing a geometric model, meshing, specifying material properties, and determining boundary conditions and external loads, the stress conditions of bridge pile foundations under different working conditions can be accurately simulated, thereby more accurately assessing their potential risks. Not only static loads are taken into account, but also traffic flow information and environmental information are combined. The range of traffic loads and environmental loads is determined through a predictive model, making the risk assessment more comprehensive and closer to reality. Based on the simulated vibration data of the finite element analysis unit, risky finite element analysis units are screened out, and the initial detection points are determined through clustering, so that the detection points are more concentrated in potential risk areas. By calculating the point risk correlation coefficient, risk detection points are further screened out, and the risk correlation coefficient between them is determined, providing a scientific basis for subsequent ultrasonic testing. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] This specification will be further described in the form of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. These embodiments are not limiting, and in these embodiments, like numbers represent like structures, wherein:

[0019] Figure 1 is a flow chart of a concrete bridge pile foundation risk detection method according to some embodiments of this specification;

[0020] Figure 2 It is a flowchart of determining multiple risk detection points of a concrete bridge pile foundation to be detected and risk correlation coefficients between the multiple risk detection points according to some embodiments of this specification;

[0021] Figure 3 It is a module schematic diagram of a concrete bridge pile foundation risk detection system according to some embodiments of this specification. DETAILED DESCRIPTION

[0022] To more clearly illustrate the technical solutions of the embodiments of this specification, the following briefly describes the drawings required for describing the embodiments. Obviously, the drawings described below are merely examples or embodiments of this specification. Those skilled in the art can apply this specification to other similar scenarios based on these drawings without inventive effort. Unless otherwise apparent from the context or otherwise noted, the same reference numerals in the figures represent the same structure or operation.

[0023] Figure 1 is a flow chart of a concrete bridge pile foundation risk detection method according to some embodiments of this specification, such as Figure 1 As shown, the concrete bridge pile foundation risk detection method may include the following process.

[0024] Step 110: Obtain structural information and service information of the concrete bridge pile foundation to be inspected.

[0025] Specifically, the structural information of the concrete bridge pile foundation to be inspected may include the pile foundation type (for example, bored cast-in-place piles, sunk tube cast-in-place piles, manually excavated piles, etc.) and size (for example, diameter, length, quantity and arrangement (for example, single row piles, multiple rows of piles, group piles, etc.)), material properties (for example, the strength grade and mix ratio of concrete, the specification, model, quantity and arrangement of steel bars, etc.), geometric structure (for example, the connection method between the pile top and the pedestal or pier body, whether the pile body is provided with special structures such as variable cross-section and enlarged head, etc.), design parameters (for example, design bearing capacity requirements (for example, vertical bearing capacity, horizontal bearing capacity and pull-out bearing capacity, etc.), design deformation limits (for example, settlement, inclination, etc.), etc.).

[0026] The service information of the concrete bridge pile foundation to be inspected includes traffic flow information (for example, daily traffic flow, vehicle type, proportion of heavy-loaded vehicles, pedestrian flow, etc.) and environmental information (for example, climatic conditions (for example, wind force, temperature, humidity, seismic activity, etc.), geological conditions (for example, soil type, groundwater level, etc.), hydrology, etc.).

[0027] Step 120 , determining multiple risk detection points of the concrete bridge pile foundation to be detected and risk correlation coefficients between the multiple risk detection points based on the structural information and service information of the concrete bridge pile foundation to be detected by finite element analysis.

[0028] Figure 2This is a flow chart of determining multiple risk detection points of a concrete bridge pile foundation to be detected and risk correlation coefficients between the multiple risk detection points according to some embodiments of this specification, such as Figure 2 As shown, step 120 specifically includes the following steps:

[0029] Based on the structural information of the concrete bridge pile foundation to be inspected, a geometric model of the concrete bridge pile foundation to be inspected is established;

[0030] Mesh the geometric model of the concrete bridge pile foundation to be inspected to obtain multiple finite element analysis units;

[0031] Based on the structural information of the concrete bridge pile foundation to be inspected, material properties (e.g., elastic modulus, Poisson's ratio, density, etc.) are assigned to each finite element analysis unit;

[0032] Determine boundary conditions (e.g., the interaction between the concrete bridge piles to be tested and the soil or foundation);

[0033] Determine multiple sets of external loads for testing based on the structural and service information of the concrete bridge pile foundation to be tested;

[0034] For each set of test external loads, test data of the concrete bridge pile foundation to be tested under the test external loads is obtained, wherein the test data includes simulated vibration data of multiple finite element analysis units;

[0035] Determine multiple initial inspection points based on the test data of the concrete bridge pile foundation to be inspected under each set of test external loads;

[0036] Based on the test data of the concrete bridge pile foundation to be tested under each set of test external loads, the point risk correlation coefficients of multiple initial test points are determined;

[0037] Based on the point risk correlation coefficients of the multiple initial detection points, multiple risk detection points are screened from the multiple initial detection points, and risk correlation coefficients between the multiple risk detection points are determined.

[0038] Specifically, the geometric model of the concrete bridge pile foundation to be tested is divided into a series of interconnected finite element analysis units. Physical laws and boundary conditions are applied to each finite element to simulate the response of the entire concrete bridge pile foundation. In finite element analysis, physical quantities such as displacement, stress, strain, and vibration within each finite element can be obtained by solving a set of linear or nonlinear equations.

[0039] Preferably, based on the structural information and service information of the concrete bridge pile foundation to be tested, multiple sets of external loads for testing are determined, including:

[0040] Obtain structural information of the concrete bridge and determine the static load of the concrete bridge pile foundation to be inspected (for example, the deadweight of the bridge, etc.);

[0041] Predicting the traffic load range of the concrete bridge pile foundation to be inspected based on traffic flow information using a first load prediction model, wherein the first load prediction model may be a convolutional neural network (CNN) model;

[0042] Predicting the environmental load range of the concrete bridge pile foundation to be inspected based on the environmental information using a second load prediction model, wherein the second load prediction model may be a convolutional neural network model;

[0043] Based on the static load, traffic load range and environmental load range of the concrete bridge pile foundation to be tested, multiple groups of external loads for testing are determined.

[0044] The traffic load range of the concrete bridge pile foundation to be inspected can represent the range of loads imposed by vehicles and pedestrians on the concrete bridge pile foundation to be inspected. The environmental load range of the concrete bridge pile foundation to be inspected can represent the range of loads imposed by the environment on the concrete bridge pile foundation to be inspected. For example, the environmental load range can include a river load range, a wind load range, a seismic load range, a temperature load range, etc.

[0045] The value can be taken from the traffic load range of the concrete bridge pile foundation to be tested as the traffic load, and the value can be taken from the environmental load range as the environmental load, and the application position and direction of the environmental load can be limited. The static load, traffic load and environmental load with determined position and direction constitute a group of external loads for testing.

[0046] Preferably, multiple initial detection points are determined based on the test data of the concrete bridge pile foundation to be tested under each set of test external loads, including:

[0047] For each finite element analysis unit, based on the test data of the finite element analysis unit under each set of test external loads, a unit risk parameter of the finite element analysis unit is determined, and a finite element analysis unit having a unit risk parameter greater than a unit risk parameter threshold is regarded as a risky finite element analysis unit;

[0048] Clustering is performed on multiple risk finite element analysis units to determine multiple unit clusters, and the risk finite element analysis unit corresponding to the cluster center of the unit cluster is used as an initial detection point.

[0049] Specifically, for each set of test external loads, the test data of the finite element analysis unit under the set of test external loads can be subjected to variational modal decomposition to obtain a plurality of modal components corresponding thereto, and component features (such as time domain features (such as amplitude, root mean square value, peak factor, waveform factor, etc.), frequency domain features (such as center frequency, bandwidth, spectral energy, spectral entropy, etc.), etc.) of each modal component are extracted.

[0050] For each finite element analysis unit, a unit risk parameter of the finite element analysis unit can be obtained based on the component features of the plurality of modal components corresponding to the test data of the finite element analysis unit under each set of test external loads. For example only, for each finite element analysis unit, a unit risk parameter of the finite element analysis unit can be obtained by a risk parameter prediction model based on the component features of the plurality of modal components corresponding to the test data of the finite element analysis unit under each set of test external loads, wherein the risk parameter prediction model can be a convolutional neural network model.

[0051] The plurality of risk finite element analysis units can be clustered according to the unit risk parameters and spatial distances of the risk finite element analysis units by a clustering algorithm (such as a K-means clustering algorithm, a hierarchical clustering algorithm, etc.), to determine a plurality of unit clusters, wherein each unit cluster includes a plurality of risk finite element analysis units, and the spatial distance is used to represent the distance between the spatial coordinates of two risk finite element analysis units.

[0052] For example only, the plurality of risk finite element analysis units can be clustered according to the following process:

[0053] S11, based on the plurality of risk finite element analysis units, initializing a plurality of unit clusters, wherein one risk finite element analysis unit corresponds to one initial unit cluster;

[0054] S12, according to the unit risk parameter of each risk finite element analysis unit, calculating the unit risk parameter absolute difference of any two risk finite element analysis units, wherein the unit risk parameter absolute difference is the absolute value of the difference between the unit risk parameters of the two risk finite element analysis units;

[0055] S13, for each risk finite element analysis unit, when the unit risk parameter absolute difference of the risk finite element analysis unit and a certain risk finite element analysis unit is less than the unit risk parameter absolute difference threshold and the spatial distance is less than the spatial distance threshold, the risk finite element analysis unit is taken as a candidate merged risk finite element analysis unit, the merging parameter of the candidate merged risk finite element analysis unit is calculated, and the risk finite element analysis unit is merged with the candidate merged risk finite element analysis unit with the largest merging parameter;

[0056] S14, determining whether the merged multiple unit clusters meet a preset condition, if so, completing clustering, if not, executing S15, wherein the preset condition may include that the number of the merged multiple unit clusters is less than a number threshold or that the merging parameters of any two merged unit clusters are less than a merging parameter threshold;

[0057] S15. For each unit cluster, when the absolute difference in unit risk parameters between the risk finite element analysis unit corresponding to the cluster center of the unit cluster and the risk finite element analysis unit corresponding to the cluster center of a certain unit cluster is less than the unit risk parameter absolute difference threshold and the spatial distance is less than the spatial distance threshold, the unit cluster is used as a candidate merged unit cluster, the merging parameters of the candidate merged unit cluster are calculated, and the unit cluster is merged with the candidate merged unit cluster with the largest merging parameters, and S14 is executed.

[0058] For example, the merging parameters of the candidate merged risk finite element analysis cells can be calculated according to the following formula:

[0059]

[0060] Among them, P (i,j) is the merging parameter of the risk finite element analysis unit of the i-th risk finite element analysis unit and its j-th candidate risk finite element analysis unit, a1 and a2 are weights, and a1 and a2 are both greater than 0, a1+a2=1, η1 and η2 are preset parameters, η1 and η2 are both greater than 0, △R (i,j) is the absolute difference in unit risk parameters between the i-th risk finite element analysis unit and its j-th candidate merged risk finite element analysis unit, D (i,j) is the spatial distance between the i-th risk finite element analysis unit and its j-th candidate merged risk finite element analysis unit.

[0061] The calculation method of the merging parameters of the candidate merged unit clusters is similar to the calculation method of the merging parameters of the candidate merged risk finite element analysis units. The absolute difference and spatial distance of the unit risk parameters between the risk finite element analysis units corresponding to the cluster centers of the two unit clusters can be substituted into the above formula, which will not be repeated here.

[0062] Preferably, based on the test data of the concrete bridge pile foundation to be tested under each set of test external loads, the point risk correlation coefficients of the multiple initial test points are determined, including:

[0063] For each set of test external loads, based on the test data of each initial detection point under the test external load, calculate the load risk correlation coefficient between any two initial detection points under the test external load;

[0064] Based on the load risk correlation coefficient of any two initial detection points under each set of test external loads, the point risk correlation coefficient of any two initial detection points is calculated.

[0065] Specifically, the load risk correlation coefficient of any two initial detection points under the test external load can be calculated according to the following formula:

[0066]

[0067] Among them, r (i,j,m) A is the load risk correlation coefficient between the i-th initial detection point and the j-th initial detection point under the m-th group of test external loads, (i,t,m) A is the instantaneous amplitude of the i-th initial detection point at the t-th moment under the m-th set of test external loads, (j,t,m) is the instantaneous amplitude of the jth initial detection point at the tth moment under the mth set of test external loads, and T is the total number of sampling moments.

[0068] The point risk correlation coefficient of any two initial detection points can be calculated according to the following formula:

[0069]

[0070] Among them, r (i,j) is the point risk correlation coefficient between the i-th initial detection point and the j-th initial detection point, and M is the total number of groups of external loads used for testing.

[0071] Preferably, based on the point risk correlation coefficients of the multiple initial detection points, multiple risk detection points are screened from the multiple initial detection points, and the risk correlation coefficients between the multiple risk detection points are determined, including:

[0072] Based on the point risk correlation coefficients of any two initial detection points, determine a first point risk correlation coefficient sequence corresponding to each initial detection point, wherein one element in the first point risk correlation coefficient sequence corresponds to one initial detection point. For example, the first element in the first point risk correlation coefficient sequence corresponds to initial detection point A, and is used to characterize the risk correlation coefficient between the current initial detection point and initial detection point A. The second element in the first point risk correlation coefficient sequence corresponds to initial detection point B, and is used to characterize the risk correlation coefficient between the current initial detection point and initial detection point B. In addition, the risk correlation coefficient of the initial detection point with itself is set to 1;

[0073] For each initial detection point, based on the first point risk correlation coefficient sequence corresponding to the initial detection point, calculate the point risk correlation coefficient mean and the point risk correlation coefficient standard deviation, and take the initial detection point whose point risk correlation coefficient mean is greater than the point risk correlation coefficient mean threshold and whose point risk correlation coefficient standard deviation is less than the point risk correlation coefficient standard deviation threshold as the risk detection point;

[0074] Based on the point risk correlation coefficients of any two risk detection points, determine the second point risk correlation coefficient sequence corresponding to each risk detection point, wherein one element in the second point risk correlation coefficient sequence corresponds to one risk detection point, the first element in the second point risk correlation coefficient sequence corresponds to risk detection point A, and is used to characterize the risk correlation coefficient between the current risk detection point and risk detection point A, the second element in the second point risk correlation coefficient sequence corresponds to risk detection point B, and is used to characterize the risk correlation coefficient between the current initial detection point and risk detection point B, and the risk correlation coefficient of the risk detection point with itself is set to 1;

[0075] Clustering the multiple risk detection points according to the second point risk correlation coefficient sequence corresponding to each risk detection point to determine multiple point clusters, wherein each point cluster includes multiple risk detection points;

[0076] For each point cluster, the risk correlation coefficient between any two risk detection points included in the point cluster is determined according to the second point risk correlation coefficient sequence corresponding to each risk detection point included in the point cluster.

[0077] Specifically, for any two risk detection points, the sequence similarity between the second-point risk correlation coefficient sequences corresponding to the two risk detection points can be calculated. A clustering algorithm (e.g., a K-means clustering algorithm, a hierarchical clustering algorithm, etc.) can be used to cluster multiple risk detection points based on the sequence similarity between the second-point risk correlation coefficient sequences corresponding to any two risk detection points to determine multiple point clusters.

[0078] For example, sequence similarity can be calculated according to the following formula:

[0079]

[0080] Among them, S (i,j) is the sequence similarity between the second point risk correlation coefficient sequence corresponding to the i-th risk detection point and the second point risk correlation coefficient sequence corresponding to the j-th risk detection point, V (i,k) is the value of the kth element in the second point risk correlation coefficient sequence corresponding to the i-th risk detection point, V (j,k) is the value of the kth element in the second-point risk correlation coefficient sequence corresponding to the jth risk detection point, and K is the total number of risk detection points.

[0081] Step 130: Setting a risk detection device at each risk detection point.

[0082] Specifically, the risk detection device may include a sensor for detecting vibration, for example, a piezoelectric vibration sensor, an eddy current sensor, an inductive vibration sensor, a capacitive vibration sensor, and the like.

[0083] Step 140 : determining a plurality of ultrasonic detection positions of the concrete bridge pile foundation to be inspected based on the detection point status data collected by the risk detection devices of the plurality of risk detection points and the risk correlation coefficients between the plurality of risk detection points.

[0084] Preferably, step 140 specifically includes:

[0085] For each point cluster, based on the test data of the concrete bridge pile foundation to be tested under each set of external loads for testing, a training sample corresponding to the point cluster is generated, an initial risk prediction model corresponding to the point cluster is established, and the initial risk prediction model corresponding to the training point cluster is trained using the training samples corresponding to the point cluster to obtain the risk prediction model corresponding to the point cluster. The risk prediction model corresponding to the point cluster is used to predict the ultrasonic detection position corresponding to the point cluster and the risk information corresponding to the ultrasonic detection position based on the detection point status data collected by the risk detection device of the risk detection point included in the point cluster and the risk correlation coefficient between any two risk detection points included in the point cluster.

[0086] Specifically, the training samples corresponding to the point cluster may include vibration information of multiple risk detection points included in the point cluster under a set of test external loads, as obtained through finite element analysis, and the risk correlation coefficient between any two risk detection points included in the point cluster. The labels of the point cluster training samples may be the test external load and the risk information (e.g., crack information, misalignment information, deformation information, etc.) caused by the test external load on the concrete bridge pile foundation to be inspected, as obtained through finite element analysis. The risk prediction model may be a long short-term memory (LSTM) model.

[0087] Step 150: Generate an ultrasonic testing plan based on multiple ultrasonic testing positions of the concrete bridge pile foundation to be tested.

[0088] The ultrasonic detection scheme includes an ultrasonic detection path and ultrasonic detection parameters corresponding to each ultrasonic detection position.

[0089] Preferably, step 150 specifically includes:

[0090] Generate a risk detection path based on the structural information of the concrete bridge pile foundation to be detected and multiple risk detection locations through a path generation model, wherein the path generation model can be a Generative Adversarial Networks (GAN) model;

[0091] The parameter generation model generates the ultrasonic detection parameters (for example, frequency, gain, dynamic range, focus point, focus quantity, fan window size and depth, wave speed, and medium sound speed) corresponding to the risk information of the ultrasonic detection position based on the risk information corresponding to the ultrasonic detection position. The parameter generation model can be a generative adversarial network model.

[0092] In step 160, the ultrasonic detection data of the concrete bridge pile foundation to be detected is obtained by performing ultrasonic detection on the concrete bridge pile foundation to be detected according to the ultrasonic detection scheme.

[0093] Specifically, the ultrasonic detection data of the concrete bridge pile foundation to be detected can be obtained by operating the ultrasonic detection device according to the ultrasonic detection path and the ultrasonic detection parameters corresponding to each ultrasonic detection position.

[0094] In step 170, the risk analysis of the concrete bridge pile foundation is performed based on the ultrasonic detection data of the concrete bridge pile foundation to be detected.

[0095] As preferred, step 170 specifically includes:

[0096] The risk analysis of the concrete bridge pile foundation is performed based on the ultrasonic detection data of the concrete bridge pile foundation to be detected by the risk analysis model. The risk analysis model can be a convolutional neural network model. The risk analysis model can output the risk information of the concrete bridge pile foundation to be detected, such as crack position, crack size, misalignment position, misalignment size, deformation position, and deformation size.

[0097] Figure 3 Fig. 1 is a schematic diagram of a concrete bridge pile foundation risk detection system according to some embodiments of the present specification. Figure 3 As shown in Fig. 1, the concrete bridge pile foundation risk detection system can include an information acquisition module, a risk monitoring module, a risk prediction module, a scheme generation module, a risk monitoring module, and a risk analysis module.

[0098] The information acquisition module can be used to acquire the structural information and service information of the concrete bridge pile foundation to be detected.

[0099] The risk monitoring module can be used to determine the multiple risk detection points of the concrete bridge pile foundation to be detected and the risk correlation coefficients between the multiple risk detection points based on the structural information and service information of the concrete bridge pile foundation to be detected by finite element analysis. The risk monitoring module can also be used to set a risk detection device at each risk detection point.

[0100] The risk prediction module can be used to determine the multiple ultrasonic detection positions of the concrete bridge pile foundation to be detected based on the detection point state data collected by the risk detection devices of the multiple risk detection points and the risk correlation coefficients between the multiple risk detection points.

[0101] The solution generation module can be used to generate an ultrasonic detection solution based on multiple ultrasonic detection positions of the concrete bridge pile foundation to be detected, wherein the ultrasonic detection solution includes an ultrasonic detection path and ultrasonic detection parameters corresponding to each ultrasonic detection position;

[0102] The risk monitoring module can be used to perform ultrasonic testing on the concrete bridge pile foundation to be tested according to the ultrasonic testing scheme, and obtain ultrasonic testing data of the concrete bridge pile foundation to be tested;

[0103] The risk analysis module can be used to perform risk analysis of concrete bridge pile foundations based on ultrasonic testing data of the concrete bridge pile foundations to be tested.

[0104] The concrete bridge pile foundation risk detection system can be used to implement the concrete bridge pile foundation risk detection method. For more descriptions of the concrete bridge pile foundation risk detection system, please refer to the relevant descriptions of the concrete bridge pile foundation risk detection method, which will not be repeated here.

[0105] Finally, it should be understood that the embodiments described in this specification are intended only to illustrate the principles of the embodiments of this specification. Other variations may also fall within the scope of this specification. Therefore, by way of example and not limitation, alternative configurations of the embodiments of this specification may be considered consistent with the teachings of this specification. Accordingly, the embodiments of this specification are not limited to the embodiments explicitly described and illustrated in this specification.

Claims

1. A method for detecting risks of concrete bridge pile foundations, characterized in that: include: Obtain structural and service information of the concrete bridge pile foundation to be inspected; Determining, by finite element analysis, a plurality of risk detection points of the concrete bridge pile foundation to be inspected and risk correlation coefficients between the plurality of risk detection points based on structural information and service information of the concrete bridge pile foundation to be inspected; Risk detection devices are installed at each risk detection point; Determining multiple ultrasonic detection positions of the concrete bridge pile foundation to be inspected based on detection point status data collected by risk detection devices at multiple risk detection points and risk correlation coefficients between the multiple risk detection points; generating an ultrasonic testing scheme based on the plurality of ultrasonic testing positions of the concrete bridge pile foundation to be tested, wherein the ultrasonic testing scheme includes an ultrasonic testing path and ultrasonic testing parameters corresponding to each ultrasonic testing position; According to the ultrasonic detection scheme, ultrasonic detection is performed on the concrete bridge pile foundation to be detected to obtain ultrasonic detection data of the concrete bridge pile foundation to be detected; Based on the ultrasonic detection data of the concrete bridge pile foundation to be detected, a risk analysis of the concrete bridge pile foundation is performed.

2. The concrete bridge pile foundation risk detection method according to claim 1, characterized in that: By finite element analysis, based on the structural information and service information of the concrete bridge pile foundation to be inspected, a plurality of risk detection points of the concrete bridge pile foundation to be inspected and risk correlation coefficients between the plurality of risk detection points are determined, including: Establishing a geometric model of the concrete bridge pile foundation to be inspected based on the structural information of the concrete bridge pile foundation to be inspected; Meshing the geometric model of the concrete bridge pile foundation to be inspected to obtain a plurality of finite element analysis units; Based on the structural information of the concrete bridge pile foundation to be inspected, specifying material properties for each of the finite element analysis units; Determine boundary conditions; determining a plurality of sets of external loads for testing based on the structural information and service information of the concrete bridge pile foundation to be tested; For each set of test external loads, solving and obtaining test data of the concrete bridge pile foundation to be inspected under the test external loads, wherein the test data includes simulated vibration data of multiple finite element analysis units; Determining a plurality of initial detection points based on the test data of the concrete bridge pile foundation to be detected under each set of the test external loads; determining point risk correlation coefficients of a plurality of initial inspection points based on the test data of the concrete bridge pile foundation to be inspected under each set of the test external loads; Based on the point risk correlation coefficients of the multiple initial detection points, multiple risk detection points are screened from the multiple initial detection points, and risk correlation coefficients between the multiple risk detection points are determined.

3. The concrete bridge pile foundation risk detection method according to claim 2, characterized in that: The service information includes traffic flow information and environmental information; Based on the structural information and service information of the concrete bridge pile foundation to be tested, multiple groups of external loads for testing are determined, including: Acquiring structural information of the concrete bridge and determining the static load of the pile foundation of the concrete bridge to be inspected; Predicting the traffic load range of the concrete bridge pile foundation to be inspected based on the traffic flow information using a first load prediction model; Predicting the environmental load range of the concrete bridge pile foundation to be inspected based on the environmental information using a second load prediction model; The multiple groups of external loads for testing are determined based on the static load, traffic load range and environmental load range of the concrete bridge pile foundation to be tested.

4. The concrete bridge pile foundation risk detection method according to claim 2 or 3, characterized in that: Based on the test data of the concrete bridge pile foundation to be tested under each set of the test external loads, a plurality of initial test points are determined, including: For each of the finite element analysis units, determining a unit risk parameter of the finite element analysis unit based on test data of the finite element analysis unit under each set of the test external loads, and taking a finite element analysis unit having a unit risk parameter greater than a unit risk parameter threshold as a risky finite element analysis unit; Clustering is performed on multiple risk finite element analysis units to determine multiple unit clusters, and the risk finite element analysis unit corresponding to the cluster center of the unit cluster is used as an initial detection point.

5. The concrete bridge pile foundation risk detection method according to claim 4, characterized in that: Based on the test data of the concrete bridge pile foundation to be tested under each set of the test external loads, the point risk correlation coefficients of the multiple initial test points are determined, including: For each set of the test external loads, based on the test data of each of the initial detection points under the test external loads, calculating the load risk correlation coefficient of any two initial detection points under the test external loads; Based on the load risk correlation coefficients of any two initial detection points under each set of the test external loads, the point risk correlation coefficients of any two initial detection points are calculated.

6. The concrete bridge pile foundation risk detection method according to claim 5, characterized in that: Based on the point risk correlation coefficients of the multiple initial detection points, screening multiple risk detection points from the multiple initial detection points, and determining the risk correlation coefficients between the multiple risk detection points, including: Based on the point risk correlation coefficients of any two initial detection points, determining a first point risk correlation coefficient sequence corresponding to each of the initial detection points; For each of the initial detection points, based on the first point risk correlation coefficient sequence corresponding to the initial detection point, calculate the point risk correlation coefficient mean and the point risk correlation coefficient standard deviation, and take the initial detection point whose point risk correlation coefficient mean is greater than the point risk correlation coefficient mean threshold and whose point risk correlation coefficient standard deviation is less than the point risk correlation coefficient standard deviation threshold as the risk detection point; Based on the point risk correlation coefficients of any two risk detection points, determining a second point risk correlation coefficient sequence corresponding to each of the risk detection points; Clustering the plurality of risk detection points according to the second point risk correlation coefficient sequence corresponding to each of the risk detection points to determine a plurality of point clusters; For each point cluster, the risk correlation coefficient between any two risk detection points included in the point cluster is determined according to the second point risk correlation coefficient sequence corresponding to each risk detection point included in the point cluster.

7. The concrete bridge pile foundation risk detection method according to claim 6, characterized in that: Determining multiple ultrasonic detection positions of the concrete bridge pile foundation to be inspected based on detection point status data collected by risk detection devices at multiple risk detection points and risk correlation coefficients between the multiple risk detection points includes: For each of the point clusters, based on the test data of the concrete bridge pile foundation to be tested under each group of the external loads for testing, a training sample corresponding to the point cluster is generated, an initial risk prediction model corresponding to the point cluster is established, and the initial risk prediction model corresponding to the point cluster is trained using the training samples corresponding to the point cluster to obtain the risk prediction model corresponding to the point cluster. The risk prediction model corresponding to the point cluster predicts the ultrasonic detection position corresponding to the point cluster and the risk information corresponding to the ultrasonic detection position based on the detection point status data collected by the risk detection device of the risk detection point included in the point cluster and the risk correlation coefficient between any two risk detection points included in the point cluster.

8. The concrete bridge pile foundation risk detection method according to claim 7, characterized in that: Based on the multiple risk detection positions of the concrete bridge pile foundation to be detected, a risk detection plan is generated, including: Generate a risk detection path based on the structural information of the concrete bridge pile foundation to be detected and multiple risk detection positions through a path generation model; The ultrasonic detection parameters corresponding to the risk detection position are generated based on the risk information corresponding to the ultrasonic detection position through a parameter generation model.

9. The concrete bridge pile foundation risk detection method according to any one of claims 1 to 3, characterized in that: Based on the ultrasonic detection data of the concrete bridge pile foundation to be detected, a risk analysis of the concrete bridge pile foundation is performed, including: The concrete bridge pile foundation risk analysis is performed based on the ultrasonic detection data of the concrete bridge pile foundation to be detected by using a risk analysis model.

10. The concrete bridge pile foundation risk detection system is characterized by: The method for detecting risk of concrete bridge pile foundations according to any one of claims 1 to 9 comprises: An information acquisition module is used to obtain structural information and service information of the concrete bridge pile foundation to be inspected; a risk monitoring module, configured to determine, through finite element analysis and based on structural information and service information of the concrete bridge pile foundation to be inspected, a plurality of risk detection points of the concrete bridge pile foundation to be inspected and risk correlation coefficients between the plurality of risk detection points, and further configured to set a risk detection device at each risk detection point; a risk prediction module, configured to determine a plurality of ultrasonic detection positions of the concrete bridge pile foundation to be inspected based on detection point status data collected by risk detection devices at the plurality of risk detection points and risk correlation coefficients between the plurality of risk detection points; a scheme generating module, configured to generate an ultrasonic testing scheme based on a plurality of ultrasonic testing positions of the concrete bridge pile foundation to be tested, wherein the ultrasonic testing scheme includes an ultrasonic testing path and ultrasonic testing parameters corresponding to each ultrasonic testing position; a risk monitoring module, configured to perform ultrasonic testing on the concrete bridge pile foundation to be tested according to the ultrasonic testing scheme, and obtain ultrasonic testing data of the concrete bridge pile foundation to be tested; The risk analysis module is used to perform risk analysis of the concrete bridge pile foundation based on the ultrasonic detection data of the concrete bridge pile foundation to be detected.

Images