CN117892590A - Concrete dam damage identification and safety assessment method based on crack intelligent identification and finite element inversion - Google Patents
Concrete dam damage identification and safety assessment method based on crack intelligent identification and finite element inversion Download PDFInfo
- Publication number
- CN117892590A CN117892590A CN202410071098.3A CN202410071098A CN117892590A CN 117892590 A CN117892590 A CN 117892590A CN 202410071098 A CN202410071098 A CN 202410071098A CN 117892590 A CN117892590 A CN 117892590A
- Authority
- CN
- China
- Prior art keywords
- dam
- finite element
- damage
- crack
- inversion
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 55
- 238000013135 deep learning Methods 0.000 claims abstract description 18
- 238000011156 evaluation Methods 0.000 claims abstract description 11
- 238000012544 monitoring process Methods 0.000 claims abstract description 11
- 238000004364 calculation method Methods 0.000 claims abstract description 5
- 230000006870 function Effects 0.000 claims description 27
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 19
- 238000012360 testing method Methods 0.000 claims description 10
- 230000004044 response Effects 0.000 claims description 7
- 230000011218 segmentation Effects 0.000 claims description 7
- 230000004913 activation Effects 0.000 claims description 6
- 238000004458 analytical method Methods 0.000 claims description 6
- 230000009286 beneficial effect Effects 0.000 claims description 4
- 238000013507 mapping Methods 0.000 claims description 4
- 239000000463 material Substances 0.000 claims description 4
- 238000005259 measurement Methods 0.000 claims description 4
- 238000012545 processing Methods 0.000 claims description 4
- 238000000605 extraction Methods 0.000 claims description 3
- 230000036541 health Effects 0.000 claims description 3
- 238000003709 image segmentation Methods 0.000 claims description 3
- 230000003993 interaction Effects 0.000 claims description 3
- 238000005457 optimization Methods 0.000 claims description 3
- 238000012916 structural analysis Methods 0.000 claims description 3
- 238000006073 displacement reaction Methods 0.000 claims description 2
- 238000011176 pooling Methods 0.000 claims description 2
- 238000004088 simulation Methods 0.000 claims description 2
- 230000008602 contraction Effects 0.000 claims 1
- 238000005070 sampling Methods 0.000 claims 1
- 238000005516 engineering process Methods 0.000 abstract description 12
- 230000005284 excitation Effects 0.000 description 5
- 238000013016 damping Methods 0.000 description 3
- 238000001514 detection method Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000006399 behavior Effects 0.000 description 2
- 230000010339 dilation Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000000638 solvent extraction Methods 0.000 description 2
- 238000011179 visual inspection Methods 0.000 description 2
- 206010047571 Visual impairment Diseases 0.000 description 1
- 230000002411 adverse Effects 0.000 description 1
- 230000032683 aging Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000006735 deficit Effects 0.000 description 1
- 238000003708 edge detection Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000003628 erosive effect Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 230000001939 inductive effect Effects 0.000 description 1
- 230000003902 lesion Effects 0.000 description 1
- 230000000704 physical effect Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000009991 scouring Methods 0.000 description 1
- 208000029257 vision disease Diseases 0.000 description 1
- 230000004393 visual impairment Effects 0.000 description 1
Landscapes
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
The invention relates to the field of dam damage identification and safety evaluation, in particular to a concrete dam damage identification and safety evaluation method based on crack intelligent identification and finite element inversion. By combining the deep learning technology with the finite element inversion method, a technology for identifying dam crack damage based on a U-Net network is provided. After the U-Net network is used for identifying and dividing the surface cracks of the dam, the dam is endowed to a finite element model of the dam for modeling, iteration is carried out based on a finite element inversion method, the size and the position of the cracks are continuously changed, and finally, the value with the minimum phase difference with the field actual value is obtained, so that the damage of the dam is determined, and further, the safety evaluation is carried out on the dam. Compared with the traditional finite element inversion method, the method utilizes the physical information of the dam, namely the identified cracks are initialized, so that the calculated amount of iterative solution is greatly reduced, the calculation efficiency and the accuracy of inversion are improved, and the method has positive significance for dam damage identification and safety monitoring.
Description
Technical Field
The invention relates to the field of dam damage identification and safety evaluation, and aims to improve the monitoring level of hydraulic engineering safety. Relates to a concrete dam damage identification and safety assessment method based on crack intelligent identification and finite element inversion.
Background
The dam is used as an artificially constructed hydraulic engineering and bears the responsibility of accumulating and scheduling water resources. However, during use, the dam may be adversely affected by aging, scouring, erosion, etc. of the material, resulting in damage such as cracks, crazes, etc., thereby causing hidden danger to safety and stability of the dam. Therefore, the method has great significance in effectively and rapidly detecting and identifying dam damage.
The conventional dam surface damage detection technology mainly comprises visual inspection and monitoring of dam structural parameters by using a sensor, wherein the visual inspection and the monitoring are limited by subjective judgment and visibility of human eyes, and the dam surface damage detection technology is not suitable for large-scale and difficult-to-contact dams. And is time consuming and inefficient; the latter typically uses displacement sensors, strain gauges, accelerometers, etc. to monitor structural change information of the dam, but is limited to monitoring changes in physical properties, and has limitations in acquisition of visual impairment information in the image, thereby impeding identification of such impairment of the crack. In addition, there are other common damage identification methods, such as remote sensing technology, which uses aerial images or satellite images for monitoring, and remote sensing technology can acquire large-scale image information, but has low resolution, and is generally difficult to capture small-scale damage. There are also image processing methods using image processing techniques for lesion recognition, such as edge detection, texture analysis, etc. These methods are generally based on manually extracted image features, have a problem of poor adaptability to complex structures, and are greatly affected by factors such as illumination.
In addition, the application of the traditional finite element inversion to dam damage identification mainly focuses on monitoring dam structure parameters, and damage is identified through inversion of measured data, so that safety evaluation is conducted. However, this method is huge in calculation amount and not accurate enough in recognition, so it is important to provide a convenient and rapid detection method.
Disclosure of Invention
In order to solve the problems, the invention provides a technology for identifying dam crack damage based on a U-Net network by combining a deep learning technology with a finite element inversion method. And identifying the dam surface crack picture acquired by the unmanned aerial vehicle by using a U-Net network, dividing the dam surface crack picture from the background, splicing the picture, giving the picture to a geometric model of the dam for modeling, iterating by using the crack identified by the U-Net network model as an initial value based on a finite element inversion method, continuously changing the size and the position of the crack, and finally obtaining a value with the minimum phase difference with the field actual value, thereby determining the loss condition of the dam, and further carrying out safety evaluation on the dam.
The invention aims to realize the method for identifying and safely evaluating the damage of the concrete dam based on intelligent crack identification and finite element inversion, which comprises the following steps:
Step S1: and acquiring images of the dam by using the unmanned aerial vehicle, obtaining pictures of cracks on the surface of the dam, and numbering the pictures in sequence.
Step S2: and analyzing the acquired data through a deep learning U-Net model, and identifying and partitioning possible crack damage signs of the dam structure.
Step S3: and establishing an initial finite element model of the dam according to the real geometric dimension of the dam, determining the mechanical properties of the concrete material, and determining inversion parameters of a finite element inversion method as the size and the position of crack damage.
Step S4: and splicing and mapping the split picture after the identification and segmentation to a finite element model of the dam.
Step S5: and carrying out dynamic test on the dam to obtain a field actual measurement power result.
Step S6: and modifying and iteratively solving the damage condition in the dam by inversion analysis, and continuously modifying the size, the position and the like of crack damage in the dam body so as to minimize the difference between the finite element calculation result and the actual field value.
And S7, obtaining a parameter value of the inversion tolerance error, and determining the overall damage condition of the dam.
And S8, carrying out safety evaluation on the dam by using the results of surface cracks, overall damage and numerical simulation.
In one embodiment, the specific method of step S1 is as follows:
s1.1, setting the flying height, focal length, camera resolution and the like of the unmanned aerial vehicle according to a fixed flying route related to the site of the dam;
S1.2, shooting a dam surface crack picture by using shooting equipment configured by an unmanned aerial vehicle.
In one embodiment, the specific method of step S2 is as follows:
S2.1, establishing a deep learning semantic segmentation network U-Net model;
S2.2, recognizing dam surface crack pictures by using a U-Net network, and dividing cracks in the pictures into a classical deep learning architecture for an image segmentation task. The dilation path on the right is part of the decoder, and the feature map of the middle layer is restored to the original resolution of the input image by upsampling and deconvolution operations, thereby producing a segmented picture. The convolution layer uses 3*3 convolution kernels for feature extraction. The loss function selects a Binary Cross entropy loss function (Binary Cross-Entropy Loss),
N is the total number of pixels of the dam crack picture, y i is the binary indication function of the ith pixel corresponding to the classification task, 0 is not in the category, 1 is in the category, p i is the probability of model prediction as positive category, the value range is 0 to 1, the activation function adopts Relu function for introducing nonlinearity,
ReLU(x)＝max(0,x) (2)
The output layer uses the Softmax activation function to generate the final segmentation map.
In one embodiment, the specific method of step S3 is as follows:
Step S3.1 is to build a finite element model for the dam based on Ansys software, and proper boundary conditions are set by considering interaction between the dam and the foundation, and the boundary conditions can be set as follows due to the fixed support:
Ux＝Uy＝Uz＝0 (3)
And S3.2, taking the influence of the water level into consideration, acquiring the water storage height of the dam through water level monitoring equipment, feeding the water storage height back to a finite element model, adjusting boundary conditions, and simulating water pressure loading conditions including the direction, the size, the distribution and the like of water pressure.
In one embodiment, the specific method of step S4 is as follows:
S4.1, after all pictures are identified by the U-Net model, the places with crack damage are segmented;
And S4.2, splicing and mapping all the pictures to the finite element model according to the corresponding numbering sequence, converting the damage information into structural information on the finite element grid, and constructing the finite element model containing the cracks.
In one embodiment, the specific method of step S5 is as follows:
Step S5 involves performing a dynamic test on the dam to obtain a dynamic result measured on site, and in order to fully understand the vibration characteristics of the dam, a manual loading excitation mode may be adopted, and when external excitation is used, a specially designed mechanical vibration device is needed, such as applying mechanical force on the dam structure or inducing vibration through a vibration table, and during the experimental process, complete response data of the dam structure must be collected to perform a comprehensive vibration characteristic analysis, so a plurality of sensors may be disposed at different positions of the dam to obtain dynamic characteristic parameters, such as vibration mode, frequency, damping ratio, and the like, which are critical for deep understanding of the vibration behavior and structural response of the dam.
In one embodiment, the specific method of step S6 is as follows:
s6.1, adopting a finite element inversion method to carry out iterative solution, converting the problem into a problem for solving an optimal solution, and converting the problem into a problem for solving the optimal solution of an independent variable by utilizing an optimization theory, namely
f0＝f(x,y,L) (4)
F 0 is a model function of the fracture, (x, y) is the center coordinates of the fracture, L represents the length of the fracture, and the function of the fracture for a simplified dam model is
Where (x 0,y0) is the coordinates of the central location of the fracture.
The objective function of step S6.2 is
J(P)＝|U0-U(P)| (6)
J(P)<ε (7)
Wherein P is an inverted model parameter, namely the position and the size of a crack, U 0 is a dynamic parameter value measured on the actual site, U (P) is a model predicted value, epsilon is an inversion tolerance error, and when the condition is met, an iterative solution can be considered to be obtained, and an iterative formula is as follows:
Where a is the learning rate and n is the iteration number.
In one embodiment, the specific method of step S8 is as follows:
Step S8, after confirming the size and the position of the dam crack, carrying out structural analysis according to the damage condition, and evaluating the influence of the crack on the structural strength and the stability by simulating structural responses under different loads and working conditions, wherein a plurality of factors are required to be comprehensively considered when finally evaluating the safety of the dam, including inversion results, information of deep learning identification and results of dynamic tests, so that the method is beneficial to accurately knowing the health condition of the dam structure and ensuring to make comprehensive and reliable judgment.
Compared with the prior art, the invention has the following beneficial effects:
The invention provides a technology for identifying dam crack damage based on a U-Net network by combining a deep learning technology with a finite element inversion method. Analyzing whether the pictures acquired by the unmanned aerial vehicle contain cracks or not, segmenting the pictures from the background, splicing the pictures, giving the pictures to a geometric model of the dam for modeling, iterating by taking the cracks identified by the U-Net network model as initial values based on a finite element inversion method, continuously changing the sizes and positions of the cracks, and finally obtaining a value with the minimum phase difference with the actual values on site, thereby determining the loss condition of the dam, and further carrying out safety evaluation on the dam.
According to the invention, the initial dam finite element model is built through the crack damage identified by the deep learning network model, and then the iterative solution is carried out based on the inversion method, so that the calculated amount of the iterative solution is greatly reduced, the accuracy is improved, and the method has positive significance for identifying and safely monitoring the dam damage.
Drawings
FIG. 1 is a schematic flow chart of an embodiment of the present invention;
FIG. 2 is a flow chart of a finite element inversion method according to an embodiment of the present invention.
Detailed Description
The invention is described in more detail below with reference to examples in order to ensure a clearer understanding of the invention by practitioners skilled in the art. It is noted that the specific examples described below are only for explaining the present invention so as to better understand the technical scheme thereof. It should be emphasized that the technical solutions provided by the present invention are not limited to what is presented in the following examples, nor should they be construed to limit the scope of the invention.
The following detailed description of specific embodiments of the invention will be presented in conjunction with the accompanying drawings 1, it being noted that the following detailed description is illustrative and is intended to provide a further explanation of the invention. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
As shown in FIG. 1, the method for identifying damage and evaluating safety of a concrete dam based on deep learning and finite element inversion is designed according to the following principle: by combining the deep learning technology with the finite element inversion method, a technology for identifying dam crack damage based on a U-Net network is provided. Analyzing whether the pictures acquired by the unmanned aerial vehicle contain cracks or not, segmenting the pictures from the background, splicing the pictures, giving the pictures to a finite element model of the dam for modeling, iterating by taking the cracks identified by the U-Net network model as initial values based on a finite element inversion method, continuously changing the sizes and positions of the cracks, finally obtaining a value with the minimum phase difference with the field actual value, determining the loss condition of the dam, and further carrying out safety evaluation on the dam, wherein fig. 2 is a finite element inversion method flow chart of the embodiment of the invention.
In this example, a method for identifying damage and evaluating safety of a concrete dam based on deep learning and finite element inversion comprises the following steps:
1. Dam image acquisition and processing
And acquiring images of the dam by using the unmanned aerial vehicle, obtaining pictures of cracks on the surface of the dam, and numbering the pictures in sequence. When the dam picture is shot, regular pictures with the same size are needed to be shot, the pictures are not overlapped for ensuring the follow-up convenient processing, a fixed flight route is designed for meeting the requirement, and the flight height, focal length, camera resolution and the like of the unmanned aerial vehicle are set and adjusted.
2. Deep learning semantic segmentation technique
And analyzing the acquired data through a deep learning U-Net model, and identifying and partitioning possible crack damage signs of the dam structure. The U-Net network is used, which is a classical deep learning architecture for image segmentation task, and its structure is characterized by U-shape, left side contracted path is encoder part, and is composed of a series of convolution layer and pooling layer for extracting advanced features of image. The dilation path on the right is part of the decoder, and the feature map of the middle layer is restored to the original resolution of the input image by upsampling and deconvolution operations, thereby producing a segmented picture. The convolution layer uses 3*3 convolution kernels for feature extraction. The loss function selects a Binary Cross entropy loss function (Binary Cross-Entropy Loss),
N is the total number of pixels of the dam crack picture, y i is the binary indication function of the ith pixel corresponding to the classification task, 0 is not in the category, 1 is in the category, p i is the probability of model prediction as positive category, the value range is 0 to 1, the activation function adopts Relu function for introducing nonlinearity,
ReLU(x)＝max(0,x) (2)
The output layer uses the Softmax activation function to generate the final segmentation map.
3. Three-dimensional modeling and dynamic testing of dams
And establishing an initial finite element model of the dam according to the real dam size, determining the mechanical properties of the concrete material, and determining inversion parameters of a finite element inversion method as the size and the position of crack damage. And splicing and mapping the split picture after the identification and segmentation to a finite element model of the dam.
Establishing a finite element model for a dam based on Ansys software, setting proper boundary conditions considering interaction between the dam and a foundation, wherein the boundary conditions can be set as follows due to the fixed support:
Ux＝Uy＝Uz＝0 (3)
The finite element model of the dam needs to take the influence of the water level into consideration, the water level monitoring equipment is used for acquiring the water storage height of the dam, feeding the water storage height back to the finite element model, adjusting boundary conditions, and simulating water pressure loading conditions including the direction, the size, the distribution and the like of water pressure. After the U-Net model recognizes all the pictures, the places with crack damage are segmented, all the pictures are spliced and mapped to the finite element model according to the corresponding numbering sequence, damage information is converted into structural information on the finite element grid, and the finite element model containing the crack is constructed.
In order to comprehensively understand the vibration characteristics of the dam body, the dynamic test can be performed by adopting a manual loading excitation mode, when external excitation is used, a specially designed mechanical vibration device is needed, such as applying mechanical force on the dam structure or vibration is caused by a vibration table, in the experimental process, complete response data of the dam structure must be collected to perform comprehensive vibration characteristic analysis, and for this purpose, a plurality of sensors can be arranged at different positions of the dam to obtain dynamic characteristic parameters, such as vibration mode, frequency, damping ratio and the like, which are critical to deep understanding of the vibration behavior and structural response of the dam.
4. Finite element inversion method
And carrying out dynamic test on the dam to obtain a field actual measurement power result, carrying out modification iterative solution on the damage condition in the dam through inversion analysis, and continuously modifying the size, the position and the like of crack damage in the dam body so as to minimize the difference between a finite element calculation result and an actual field value. And obtaining a parameter value of the inversion tolerance error, and determining the overall damage condition of the dam.
Dynamic testing is carried out on the dam to obtain a field actual measurement power result, a corresponding number of sensors are arranged at different positions of the dam, dynamic excitation is loaded, and dynamic characteristic parameters such as vibration mode, frequency, damping ratio and the like are obtained.
The finite element inversion method is adopted for iterative solution, so that the problem can be converted into the problem for solving the optimal solution, and the problem is converted into the problem for solving the optimal solution of the independent variable by utilizing the optimization theory, namely
f0＝f(x,y,L) (4)
F 0 is a model function of the fracture, (x, y) is the center coordinates of the fracture, L represents the length of the fracture, and the function of the fracture for a simplified dam model is
Where (x 0,y0) is the coordinates of the central location of the fracture. The objective function is
J(P)＝|U0-U(P)| (6)
J(P)<ε (7)
Wherein P is an inverted model parameter, namely the position and the size of a crack, U 0 is a dynamic parameter value measured on the actual site, U (P) is a model predicted value, epsilon is an inversion tolerance error, and when the condition is met, an iterative solution can be considered to be obtained, and an iterative formula is as follows:
Where a is the learning rate and n is the iteration number.
5. Safety assessment of a dam
And carrying out safety evaluation of the dam according to the result obtained by the steps. After confirming the size and the position of the dam crack, carrying out structural analysis according to the damage condition, and evaluating the influence of the crack on the structural strength and the stability by simulating structural responses under different loads and working conditions, wherein a plurality of factors are required to be comprehensively considered when finally evaluating the safety of the dam, including inversion results, information of deep learning identification and results of dynamic tests, so that the method is beneficial to accurately knowing the health condition of the dam structure and ensures that comprehensive and reliable judgment is made.
The above embodiments illustrate the principles and effects of the present invention by way of example only, and are not intended to limit the present invention. Modifications and variations may be made to the above-described embodiments by any person skilled in the art without departing from the central spirit and scope of the present invention. Therefore, all equivalent modifications and changes which are generally known to those skilled in the relevant art without departing from the spirit and scope of the present invention are intended to be included within the scope of the following claims.
Claims (9)
1. The invention provides a concrete dam damage identification and safety assessment method based on crack intelligent identification and finite element inversion, which is characterized by comprising the following steps:
Step S1: acquiring images of a dam by using an unmanned aerial vehicle, obtaining pictures of cracks on the surface of the dam, and numbering the pictures in sequence;
step S2: analyzing the acquired data by using a deep learning U-Net model, and identifying possible crack damage signs of the dam structure;
step S3: establishing an initial finite element model of the dam according to the real dam size, determining the mechanical properties of the concrete material, and determining inversion parameters of a finite element inversion method as the size and the position of crack damage;
step S4: splicing and mapping the split picture after identification to a finite element model of the dam;
Step S5: dynamic testing is carried out on the dam to obtain a field actual measurement power result;
step S6: modifying and iteratively solving the damage condition in the dam by inversion analysis, and continuously modifying the size, the position and the like of crack damage in the dam body so as to minimize the difference between a finite element calculation result and an actual field value;
s7, obtaining a parameter value of inversion tolerance error, and determining the overall damage condition of the dam;
And S8, carrying out safety evaluation on the dam by using the results of surface cracks, overall damage and numerical simulation.
2. The method according to claim 1, characterized in that: when the unmanned aerial vehicle is used for shooting the dam picture in the step S1, regular photos with the same size need to be shot, the photos need not to be overlapped in order to ensure the follow-up convenient processing, a fixed flight route needs to be designed in order to meet the requirement, and the flight height, focal length, camera resolution and the like of the unmanned aerial vehicle are set and adjusted.
3. The method according to claim 1, characterized in that: the step S2 uses a U-Net network, which is a classical deep learning architecture for image segmentation task, and is characterized in that the structure is U-shaped, the left side contraction path is an encoder part, and consists of a series of convolution layers and pooling layers, the right side expansion path is a decoder part, the feature image of the middle layer is restored to the original resolution of the input image through up-sampling and deconvolution operation, so as to generate a segmented picture, the convolution layer adopts a 3*3 convolution kernel to perform feature extraction, the loss function selects a binary cross entropy loss function (Binary cross-Entropy Loss),
N is the total number of pixels of the dam crack picture, y i is the binary indication function of the ith pixel corresponding to the classification task, 0 is not in the category, 1 is in the category, p i is the probability of model prediction as positive category, the value range is 0 to 1, the activation function adopts Relu function for introducing nonlinearity,
ReLU(x)＝max(0,x) (2)
The output layer uses the Softmax activation function to generate the final segmentation map.
4. The method according to claim 1, characterized in that: step S3 is to build a finite element model for the dam based on the Ansys software, and appropriate boundary conditions need to be set in consideration of interaction between the dam and the foundation, and the boundary conditions can be set as follows due to the fixed support:
Ux＝Uy＝Uz＝0 (3)
U x,Uy,Uz is the displacement of the dam in three directions.
5. The method according to claim 1, characterized in that: the finite element model of the dam in the step S3 needs to take the influence of the water level into consideration, the water level monitoring equipment is used for acquiring the water level of the dam, feeding the water level into the finite element model, adjusting boundary conditions, and simulating the water pressure loading condition, including the direction, the size, the distribution and the like of the water pressure.
6. The method according to claim 1, characterized in that: and step S4, after the U-Net model recognizes all the pictures, the places with cracks and damage are segmented, all the pictures are spliced and mapped onto the finite element model according to the corresponding number sequence, damage information is converted into structural information on the finite element grid, and the finite element model containing the cracks is constructed.
7. The method according to claim 1, characterized in that: the step S6 adopts a finite element inversion method to carry out iterative solution, so that the problem can be converted into a problem for solving the optimal solution, and the problem is converted into a problem for solving the optimal solution of the independent variable by utilizing an optimization theory, namely
f0＝f(x,y,L) (4)
F 0 is a model function of the fracture, (x, y) is a center coordinate of the fracture, L represents a length of the fracture, and the fracture function for a simplified dam model is
Where (x 0,y0) is the coordinates of the central location of the fracture.
8. The method according to claim 1, characterized in that: the objective function of the step S6 is
J(P)＝|U0-U(P)| (6)
J(P)<ε (7)
Wherein P is an inverted model parameter, namely the position and the size of a crack, U 0 is a dynamic parameter value measured on the actual site, U (P) is a model predicted value, epsilon is an inversion tolerance error, and when the condition is met, an iterative solution can be considered to be obtained, and an iterative formula is as follows:
Where a is the learning rate and n is the iteration number.
9. The method according to claim 1, characterized in that: after confirming the size and the position of the dam crack, the step S8 carries out structural analysis according to the damage condition, and evaluates the influence of the crack on the structural strength and the stability by simulating structural responses under different loads and working conditions, and is remarkable in that a plurality of factors including inversion results, information of deep learning identification and results of dynamic tests need to be comprehensively considered when finally evaluating the safety of the dam, so that the method is beneficial to accurately knowing the health condition of the dam structure and ensuring to make comprehensive and reliable judgment.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410071098.3A CN117892590A (en) | 2024-01-18 | 2024-01-18 | Concrete dam damage identification and safety assessment method based on crack intelligent identification and finite element inversion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410071098.3A CN117892590A (en) | 2024-01-18 | 2024-01-18 | Concrete dam damage identification and safety assessment method based on crack intelligent identification and finite element inversion |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117892590A true CN117892590A (en) | 2024-04-16 |
Family
ID=90644571
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410071098.3A Pending CN117892590A (en) | 2024-01-18 | 2024-01-18 | Concrete dam damage identification and safety assessment method based on crack intelligent identification and finite element inversion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117892590A (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2017096684A (en) * | 2015-11-19 | 2017-06-01 | 日本工営株式会社 | Rubble concrete strength evaluation method and rubble concrete strength evaluation program |
| CN115880594A (en) * | 2022-12-18 | 2023-03-31 | 新乡市水利水电工程质量监测站 | Intelligent dam crack detection method based on unmanned aerial vehicle visual perception and deep learning |
| CN116152674A (en) * | 2021-11-20 | 2023-05-23 | 华能澜沧江水电股份有限公司 | Dam unmanned aerial vehicle image crack intelligent recognition method based on improved U-Net model |
-
2024
- 2024-01-18 CN CN202410071098.3A patent/CN117892590A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2017096684A (en) * | 2015-11-19 | 2017-06-01 | 日本工営株式会社 | Rubble concrete strength evaluation method and rubble concrete strength evaluation program |
| CN116152674A (en) * | 2021-11-20 | 2023-05-23 | 华能澜沧江水电股份有限公司 | Dam unmanned aerial vehicle image crack intelligent recognition method based on improved U-Net model |
| CN115880594A (en) * | 2022-12-18 | 2023-03-31 | 新乡市水利水电工程质量监测站 | Intelligent dam crack detection method based on unmanned aerial vehicle visual perception and deep learning |
Non-Patent Citations (2)
| Title |
|---|
| AMIR PIROOZNIA等: "Investigation of size effect and smeared crack models in ordinary and dam concrete fracture tests", pages 1 - 25, Retrieved from the Internet <URL:《网页在线公开：https://www.sciencedirect.com/science/article/pii/S0013794419305399》> * |
| 李烁: "基于深度学习方法的大坝裂缝检测研究", 《云南水力发电》, vol. 39, no. 8, 4 September 2023 (2023-09-04), pages 344 - 347 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108647585B (en) | Traffic identifier detection method based on multi-scale circulation attention network | |
| Qiu et al. | Automatic visual defects inspection of wind turbine blades via YOLO-based small object detection approach | |
| CN108960135B (en) | Dense ship target accurate detection method based on high-resolution remote sensing image | |
| CN108959794A (en) | A kind of structural frequency response modification methodology of dynamics model based on deep learning | |
| CN114705689A (en) | Unmanned aerial vehicle-based method and system for detecting cracks of outer vertical face of building | |
| CN110751644B (en) | Road surface crack detection method | |
| CN114022770A (en) | Mountain crack detection method based on improved self-attention mechanism and transfer learning | |
| CN114049356B (en) | Method, device and system for detecting structure apparent crack | |
| CN112184687B (en) | Road crack detection method based on capsule feature pyramid and storage medium | |
| CN113610070A (en) | Landslide disaster identification method based on multi-source data fusion | |
| CN109584206B (en) | Method for synthesizing training sample of neural network in part surface flaw detection | |
| CN111507972A (en) | Tunnel surface defect detection method combining convolutional neural network and support vector machine | |
| CN114511710A (en) | Image target detection method based on convolutional neural network | |
| CN114358091B (en) | Pile damage identification method, equipment and medium based on convolutional neural network | |
| CN116342536A (en) | Aluminum strip surface defect detection method, system and equipment based on lightweight model | |
| CN113516652A (en) | Battery surface defect and adhesive detection method, device, medium and electronic equipment | |
| CN117576073A (en) | Road defect detection method, device and medium based on improved YOLOv8 model | |
| CN117150838A (en) | Crack damage intelligent assessment method based on visual information and physical fusion | |
| CN116682045A (en) | Beam pumping unit fault detection method based on intelligent video analysis | |
| CN117892590A (en) | Concrete dam damage identification and safety assessment method based on crack intelligent identification and finite element inversion | |
| CN115984687A (en) | River work moving bed model test water boundary measuring method, device, equipment and medium | |
| CN115375925A (en) | Underwater sonar image matching algorithm based on phase information and deep learning | |
| CN115330705A (en) | Skin paint surface defect detection method based on adaptive weighting template NCC | |
| CN111739057A (en) | Free liquid level identification and extraction method based on U-net convolution neural network model | |
| Bahreini et al. | Dynamic graph CNN based semantic segmentation of concrete defects and as-inspected modeling |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |