Dam crack sound wave detection method and device based on intelligent discrimination and storage medium
Through the intelligent discrimination method based on drilling acoustic wave measurement method combined with U-HRNet, the accuracy problem of deep crack detection in the dam is solved, and automated and high-precision crack depth discrimination is achieved, especially the accurate detection of water-filled cracks.
Patent Information
- Application Number
- CN202510490537.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-18
- Publication Date
- 2025-08-08
AI Technical Summary
In the prior art, the accuracy of deep crack detection in dams is low, especially the detection of water-filled cracks is difficult, and the error of manual judgment is large.
The drilling sound wave measurement method is used to detect dam cracks, and the trained U-shaped high-resolution network U-HRNet is used to intelligently distinguish the detection waveform data. Through deep learning and transfer learning, automated crack depth discrimination is achieved.
The interpretation accuracy and efficiency of drilling acoustic wave measurement method are improved, and human error is reduced. In particular, the detection effect of water-filled cracks is significant, and automated and high-precision crack depth judgment is achieved.
Smart Images
Figure CN120446281A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of crack detection, and in particular to a dam crack acoustic wave detection method, device and storage medium based on intelligent discrimination. Background Art
[0002] Large concrete dams can develop cracks of varying sizes and depths due to factors such as temperature fluctuations, evaporation, chemical reactions, earthquakes, and changes in water pressure. Cracks can also be filled differently depending on their location, with some being water-filled and others not. Cracks can affect the dam's overall bearing capacity and stability. Furthermore, water-filled cracks on the water-facing surface can be split by the water pressure, accelerating their development into highly dangerous through-hole cracks. Regular testing can detect cracks early, allowing for timely repairs and preventing their expansion and deterioration. Testing data is a crucial basis for dam safety assessments and risk management.
[0003] Currently, most crack detection methods focus on superficial and shallow cracks. Deep cracks require drilling, often using acoustic wave detection. However, when cracks are filled with water, the acoustic wave attenuation characteristics are not obvious, making it difficult to obtain effective detection results. Furthermore, manual interpretation of the detection results is required, resulting in large errors in crack depth determination. Summary of the Invention
[0004] The present invention provides a dam crack acoustic wave detection method, device and storage medium based on intelligent discrimination, which are used to solve the technical problem of low accuracy in detecting deep cracks in dams in the prior art.
[0005] In a first aspect, the present invention provides a method for detecting dam cracks by acoustic waves based on intelligent discrimination, comprising the following steps.
[0006] Detecting the crack to be detected based on the drilling acoustic wave detection method to obtain detection waveform data corresponding to the crack to be detected; Inputting the detection waveform data corresponding to the crack to be detected into a crack depth discrimination model to discriminate the crack depth, thereby obtaining the discrimination depth of the crack to be detected; The crack depth discrimination model is a trained U-shaped high-resolution network U-HRNet.
[0007] In some embodiments, the crack depth discrimination model is specifically obtained by training through the following steps: Get the training dataset; The U-HRNet is trained based on the training data set to obtain a crack depth discrimination model.
[0008] In some embodiments, obtaining a training data set includes: Establish multiple concrete models containing aggregates, and add different cracks to different concrete models to obtain multiple crack models; Based on the measurement observation system, the acoustic wave finite difference forward modeling is carried out on each crack model to obtain the simulated detection waveform data; The simulated detection waveform data corresponding to each crack model is used as input data, and the crack depth corresponding to the crack model is used as label data. The input data and the label data are combined to form a training data set.
[0009] In some embodiments, the training of U-HRNet based on the training dataset includes: Standardize the training data sets corresponding to all fracture models; U-HRNet is trained based on the standardized training dataset.
[0010] In some embodiments, adding different cracks to different concrete models includes: For different concrete models, cracks with different crack widths, different crack depths, different extension directions and / or different filling degrees are added.
[0011] In some embodiments, the method further comprises: Acquiring supplementary training data based on the detected waveform data and actual depth of the crack to be detected; Based on the supplementary training data, the crack depth discrimination model is fine-tuned through transfer learning.
[0012] In a second aspect, the present invention provides a dam crack acoustic wave detection device based on intelligent discrimination, comprising the following modules.
[0013] A detection module, configured to detect cracks to be detected using a drilling acoustic wave detection method, and obtain detection waveform data corresponding to the cracks to be detected; A discrimination module, configured to input the detection waveform data corresponding to the crack to be detected into a crack depth discrimination model to discriminate the crack depth, thereby obtaining a discrimination depth of the crack to be detected; The crack depth discrimination model is a trained U-shaped high-resolution network U-HRNet.
[0014] In some embodiments, the crack depth discrimination model is specifically obtained by training through the following steps: Get the training dataset; The U-HRNet is trained based on the training data set to obtain a crack depth discrimination model.
[0015] In some embodiments, obtaining a training data set includes: Establish multiple concrete models containing aggregates, and add different cracks to different concrete models to obtain multiple crack models; Based on the measurement observation system, the acoustic wave finite difference forward modeling is carried out on each crack model to obtain the simulated detection waveform data; The simulated detection waveform data corresponding to each crack model is used as input data, and the crack depth corresponding to the crack model is used as label data. The input data and the label data are combined to form a training data set.
[0016] In some embodiments, the training of U-HRNet based on the training dataset includes: Standardize the training data sets corresponding to all fracture models; U-HRNet is trained based on the standardized training dataset.
[0017] In some embodiments, adding different cracks to different concrete models includes: For different concrete models, cracks with different crack widths, different crack depths, different extension directions and / or different filling degrees are added.
[0018] In some embodiments, it further includes: An acquisition module, configured to acquire supplementary training data based on the detected waveform data and actual depth of the crack to be detected; A fine-tuning module is used to fine-tune the crack depth discrimination model through transfer learning based on the supplementary training data.
[0019] In a third aspect, the present invention provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the method for detecting dam cracks by acoustic waves based on intelligent discrimination as described above is implemented.
[0020] In a fourth aspect, a non-transitory computer-readable storage medium stores a computer program, which, when executed by a processor, implements any of the above-described methods for detecting dam cracks using acoustic waves based on intelligent discrimination.
[0021] In a fifth aspect, the present invention further provides a computer program product, comprising a computer program, which, when executed by a processor, implements any of the above-described methods for detecting dam cracks using acoustic waves based on intelligent discrimination.
[0022] The intelligently discriminated dam crack acoustic wave detection method, device, and storage medium provided by the present invention detect cracks to be detected using the borehole acoustic wave detection method to obtain detection waveform data corresponding to the cracks to be detected. This detection waveform data is then input into a crack depth discrimination model to determine the crack depth and obtain the determined depth of the crack to be detected. The crack depth discrimination model is a trained U-shaped high-resolution network (U-HRNet). The U-HRNet automatically and intelligently discriminates the detection results of the borehole acoustic wave detection method, rapidly obtaining more accurate crack depth determination results. This avoids errors caused by subjective human judgment and improves the interpretation accuracy and efficiency of the borehole acoustic wave detection method. BRIEF DESCRIPTION OF THE DRAWINGS
[0023] In order to more clearly illustrate the technical solutions in the present invention or the prior art, a brief introduction is given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.
[0024] Figure 1 This is one of the flow charts of the dam crack acoustic wave detection method based on intelligent discrimination provided by the present invention.
[0025] Figure 2 It is a structural schematic diagram of the crack depth discrimination model provided by the present invention.
[0026] Figure 3 This is the second flow chart of the dam crack acoustic wave detection method based on intelligent discrimination provided by the present invention.
[0027] Figure 4 It is a structural schematic diagram of the dam crack acoustic wave detection device based on intelligent discrimination provided by the present invention.
[0028] Figure 5 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION
[0029] Current dam crack detection methods primarily focus on superficial and shallow cracks. Borehole acoustic wave detection can detect deeper cracks, but its drawback is that manual identification can lead to significant errors, particularly in the presence of water-filled cracks.
[0030] Based on the above technical problems, the present invention proposes a dam crack acoustic wave detection method based on intelligent discrimination. The method detects cracks to be detected based on the borehole acoustic wave pairing method to obtain detection waveform data corresponding to the cracks to be detected. The detection waveform data corresponding to the cracks to be detected is then input into a crack depth discrimination model to discriminate the crack depth and obtain the discrimination depth of the cracks to be detected. The crack depth discrimination model is a trained U-shaped high-resolution network U-HRNet. Using U-HRNet to automatically and intelligently discriminate the detection results of the borehole acoustic wave pairing method solves the problem of inaccurate crack depth interpretation by the borehole acoustic wave pairing method and the difficulty in interpreting water-filled cracks. It avoids errors caused by subjective human judgment and improves the interpretation accuracy and efficiency of the borehole acoustic wave pairing method.
[0031] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.
[0032] Figure 1 This is one of the flow charts of the dam crack acoustic wave detection method based on intelligent discrimination provided by the present invention, such as Figure 1 As shown, the present invention provides a method for detecting dam cracks with acoustic waves based on intelligent discrimination. The method includes: Step 101: Use a drilling acoustic wave detection method to detect a crack to be detected, and obtain detection waveform data corresponding to the crack to be detected.
[0033] Step 102: Input the detection waveform data corresponding to the crack to be detected into a crack depth discrimination model to identify the crack depth and obtain the depth of the crack to be detected; wherein, the crack depth discrimination model is obtained by training a U-type high-resolution network U-HRNet based on training sample data.
[0034] Specifically, the borehole acoustic wave detection method is used to detect cracks in the dam, collecting detection waveform data corresponding to the cracks. The obtained detection waveform data is then normalized to obtain data suitable for input into the model. This data is then input into a pre-trained crack depth discrimination model to determine the depth of the corresponding crack and obtain the discrimination depth of the crack to be detected.
[0035] Among them, the crack depth discrimination model is obtained by training the U-type high-resolution network U-HRNet based on the training sample data.
[0036] Figure 2This is a schematic diagram of the structure of the crack depth discrimination model provided by the present invention. Figure 2 As shown in the figure, U-HRNet uses U-Net as the basic network skeleton and embeds HRNet as the basic module into each downsampling and upsampling. The HRNet module consists of stage 1 and stage 2 networks, removing the high-resolution branches of the latter two stages to reduce more computational requirements. Through the embedding of the HRNet module, the high-resolution output incorporates more information compared to the original U-Net. In addition to the fusion between feature maps of adjacent sizes, there are also connection operations similar to those in the original U-Net between stages 2 and 8, stage 3 and stage 7, and stage 4 and stage 6, which fuse feature maps of the same size together, allowing the network to fully utilize previously learned information while avoiding the loss of spatial or semantic knowledge due to resolution changes.
[0037] The intelligently judged acoustic wave detection method for dam cracks provided in the embodiment of the present application utilizes U-HRNet to automatically and intelligently judge the detection results of the borehole acoustic wave pairing method, thereby solving the problem of inaccurate interpretation of crack depth by the borehole acoustic wave pairing method and the difficulty in interpreting water-filled cracks. It avoids the errors caused by subjective human judgment, improves the interpretation accuracy of the borehole acoustic wave pairing method, and reduces the human workload.
[0038] In some embodiments, the crack depth discrimination model is specifically obtained by training through the following steps: Get the training dataset; The U-HRNet is trained based on the training data set to obtain a crack depth discrimination model.
[0039] Specifically, a training dataset is first obtained, which can be collected historical data or simulated or simulated data. Then, the U-HRNet is trained based on the training dataset to obtain a crack depth discrimination model.
[0040] The embodiment of the present application provides a dam crack acoustic wave detection method based on intelligent discrimination, which obtains a crack depth discrimination model through deep learning training to discriminate the detection results, thereby improving the accuracy of the discrimination and reducing the manual workload.
[0041] In some embodiments, obtaining a training data set includes: Establish multiple concrete models containing aggregates, and add different cracks to different concrete models to obtain multiple crack models; Based on the measurement observation system, the acoustic wave finite difference forward modeling is carried out on each crack model to obtain the simulated detection waveform data; The simulated detection waveform data corresponding to each crack model is used as input data, and the crack depth corresponding to the crack model is used as label data. The input data and the label data are combined to form a training data set.
[0042] Specifically, the training dataset required for deep learning training is made from synthetic data.
[0043] First, multiple concrete models containing aggregates are randomly established in batches, and then different cracks are added to different concrete models to obtain multiple crack models.
[0044] Each fracture model is then simulated using numerical simulation methods to obtain the corresponding detection waveform data. Specifically, based on a countermeasure observation system, the simulated waveform records (i.e., detection waveform data) are obtained through acoustic finite-difference forward modeling. This detection waveform data and the accurate crack depth corresponding to the fracture model are used as the training data set. A countermeasure observation system involves placing drill holes at equal intervals on both sides of the fracture, stimulating acoustic waves in one hole and receiving them in the other. This data acquisition method is generally referred to as a countermeasure observation system.
[0045] In some embodiments, adding different cracks to different concrete models includes: For different concrete models, cracks with different crack widths, different crack depths, different extension directions and / or different filling degrees are added.
[0046] Specifically, different cracks are obtained by setting different crack widths, crack depths, extension directions and / or filling degrees. After being added to the concrete model, a variety of crack models can be obtained.
[0047] The dam crack acoustic wave detection method based on intelligent discrimination provided in the embodiment of the present application obtains a training data set by establishing multiple crack models, thereby improving the richness and effectiveness of the training data set and being able to obtain a large amount of data, thereby improving the accuracy and universality of model training.
[0048] In some embodiments, the training of U-HRNet based on the training dataset includes: Standardize the training data sets corresponding to all fracture models; U-HRNet is trained based on the standardized training dataset.
[0049] Specifically, during the model training phase, the training datasets corresponding to all fracture models must first be standardized to obtain a training dataset suitable for model training. A portion of the training dataset is then extracted as a test dataset. This test dataset does not participate in training but is used to calculate the model error on the test dataset during training, thereby preventing overfitting.
[0050] In some embodiments, the method further comprises: Acquiring supplementary training data based on the detected waveform data and actual depth of the crack to be detected; Based on the supplementary training data, the crack depth discrimination model is fine-tuned through transfer learning.
[0051] Specifically, in the actual work area, if there is a verification method on site to obtain the accurate actual crack depth, the detection waveform data and actual depth can be used as supplementary training data, and through transfer learning, the model can be adaptively fine-tuned to obtain a more accurate crack depth discrimination model, thereby improving the application effect of the crack depth discrimination model in the work area.
[0052] Figure 3 This is the second flow chart of the dam crack acoustic wave detection method based on intelligent discrimination provided by the present invention, such as Figure 3 As shown in the figure, after a large number of forward modeling datasets with different fracture characteristics are obtained through fracture modeling, they are standardized and used as input datasets. Label data is generated based on the accurate fracture depth to complete the training data generation. The U-HRNet network is trained based on this training dataset. In actual work areas, internal fractures are detected using the borehole acoustic wave detection method. The detection waveform data is collected, standardized, and input into the network to obtain intelligently determined fracture depths. If field verification data is available, that is, the actual fracture depths of some cracks to be tested are known, this data can be combined with the corresponding detection waveform data to generate supplementary training data. Through transfer learning, the crack depth discrimination model is adaptively fine-tuned. The fine-tuned fracture depth discrimination model is then used to determine the fracture depth of the next detection structure, improving the network's application performance in the work area. If the accurate actual fracture depth cannot be obtained, that is, the supplementary training data required for transfer learning cannot be generated, no fine-tuning is performed, and the original fracture depth discrimination model is used to determine the fracture depth of the next detection structure.
[0053] The intelligently judged acoustic wave detection method for dam cracks provided in the embodiment of the present application utilizes U-HRNet to intelligently judge the detection results of the borehole acoustic wave detection method, thereby improving the interpretation accuracy of the borehole acoustic wave detection method, realizing the automated interpretation of the results of the borehole acoustic wave detection method, reducing the manual operation process, and improving the effectiveness of the borehole acoustic wave detection method for water-filled cracks.
[0054] Figure 4 This is a schematic diagram of the structure of the dam crack acoustic wave detection device based on intelligent discrimination provided by the present invention. Figure 4 As shown, the present invention provides a dam crack acoustic wave detection device based on intelligent discrimination, including a detection module 201 and a discrimination module 202.
[0055] The detection module 201 is used to detect the cracks to be detected by using a drilling acoustic wave detection method to obtain detection waveform data corresponding to the cracks to be detected.
[0056] The discrimination module 202 is used to input the detection waveform data corresponding to the crack to be detected into the crack depth discrimination model to discriminate the crack depth and obtain the discrimination depth of the crack to be detected.
[0057] In some embodiments, the crack depth discrimination model is specifically obtained by training through the following steps: Get the training dataset; The U-HRNet is trained based on the training data set to obtain a crack depth discrimination model.
[0058] In some embodiments, obtaining a training data set includes: Establish multiple concrete models containing aggregates, and add different cracks to different concrete models to obtain multiple crack models; Based on the measurement observation system, the acoustic wave finite difference forward modeling is carried out on each crack model to obtain the simulated detection waveform data; The simulated detection waveform data corresponding to each crack model is used as input data, and the crack depth corresponding to the crack model is used as label data. The input data and the label data are combined to form a training data set.
[0059] In some embodiments, the training of U-HRNet based on the training dataset includes: Standardize the training data sets corresponding to all fracture models; U-HRNet is trained based on the standardized training dataset.
[0060] In some embodiments, adding different cracks to different concrete models includes: For different concrete models, cracks with different crack widths, different crack depths, different extension directions and / or different filling degrees are added.
[0061] In some embodiments, further comprising: An acquisition module, configured to acquire supplementary training data based on the detected waveform data and actual depth of the crack to be detected; A fine-tuning module is used to fine-tune the crack depth discrimination model through transfer learning based on the supplementary training data.
[0062] Specifically, the above-mentioned dam crack acoustic wave detection device based on intelligent discrimination provided by the present invention can implement all the method steps implemented in the above-mentioned dam crack acoustic wave detection method embodiment based on intelligent discrimination, and can achieve the same technical effect. The parts and beneficial effects of this embodiment that are the same as those of the method embodiment will not be described in detail here.
[0063] It should be noted that the division of units / modules in the above-mentioned embodiments of the present invention is schematic and is merely a logical functional division. In actual implementation, other division methods may be used. In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The above-mentioned integrated units may be implemented in the form of hardware or software functional units.
[0064] Figure 5 Schematic diagram of the structure of the electronic device provided by the present invention, such as Figure 5 As shown, the electronic device may include: a processor 301, a communications interface 302, a memory 303, and a communication bus 304. The processor 301, the communications interface 302, and the memory 303 communicate with each other via the communication bus 304. The processor 301 may call logic instructions in the memory 303 to execute a dam crack acoustic wave detection method based on intelligent discrimination, which includes: Detecting the crack to be detected based on the drilling acoustic wave detection method to obtain detection waveform data corresponding to the crack to be detected; Inputting the detection waveform data corresponding to the crack to be detected into a crack depth discrimination model to discriminate the crack depth, thereby obtaining the discrimination depth of the crack to be detected; The crack depth discrimination model is a trained U-shaped high-resolution network U-HRNet.
[0065] In some embodiments, the crack depth discrimination model is specifically obtained by training through the following steps: Get the training dataset; The U-HRNet is trained based on the training data set to obtain a crack depth discrimination model.
[0066] In some embodiments, obtaining a training data set includes: Establish multiple concrete models containing aggregates, and add different cracks to different concrete models to obtain multiple crack models; Based on the measurement observation system, the acoustic wave finite difference forward modeling is carried out on each crack model to obtain the simulated detection waveform data; The simulated detection waveform data corresponding to each crack model is used as input data, and the crack depth corresponding to the crack model is used as label data. The input data and the label data are combined to form a training data set.
[0067] In some embodiments, the training of U-HRNet based on the training dataset includes: Standardize the training data sets corresponding to all fracture models; U-HRNet is trained based on the standardized training dataset.
[0068] In some embodiments, adding different cracks to different concrete models includes: For different concrete models, cracks with different crack widths, different crack depths, different extension directions and / or different filling degrees are added.
[0069] In some embodiments, the method further comprises: Acquiring supplementary training data based on the detected waveform data and actual depth of the crack to be detected; Based on the supplementary training data, the crack depth discrimination model is fine-tuned through transfer learning.
[0070] Specifically, the processor 301 may be a central processing unit (CPU), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or a complex programmable logic device (CPLD), and the processor may also adopt a multi-core architecture.
[0071] The logical instructions in memory 303 can be implemented in the form of software functional units and when sold or used as an independent product, can be stored in a processor-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) or a processor to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program code.
[0072] In some embodiments, a computer program product is further provided. The computer program product includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can perform the dam crack acoustic wave detection method based on intelligent discrimination provided by each of the above method embodiments. The method includes: Detecting the crack to be detected based on the drilling acoustic wave detection method to obtain detection waveform data corresponding to the crack to be detected; Inputting the detection waveform data corresponding to the crack to be detected into a crack depth discrimination model to discriminate the crack depth, thereby obtaining the discrimination depth of the crack to be detected; The crack depth discrimination model is a trained U-shaped high-resolution network U-HRNet.
[0073] Specifically, the above-mentioned computer program product provided in the embodiment of the present application can implement all the method steps implemented in the above-mentioned method embodiments, and can achieve the same technical effects. The parts and beneficial effects of this embodiment that are the same as those of the method embodiment will not be described in detail here.
[0074] In some embodiments, a computer-readable storage medium is further provided, wherein the computer-readable storage medium stores a computer program, wherein the computer program is configured to cause a computer to execute the dam crack acoustic wave detection method based on intelligent discrimination provided in each of the above method embodiments, the method comprising: Detecting the crack to be detected based on the drilling acoustic wave detection method to obtain detection waveform data corresponding to the crack to be detected; Inputting the detection waveform data corresponding to the crack to be detected into a crack depth discrimination model to discriminate the crack depth, thereby obtaining the discrimination depth of the crack to be detected; The crack depth discrimination model is a trained U-shaped high-resolution network U-HRNet.
[0075] Specifically, the computer-readable storage medium provided by the present invention can implement all the method steps implemented by the above-mentioned method embodiments, and can achieve the same technical effects. The parts and beneficial effects that are the same as the method embodiments in this embodiment will not be described in detail here.
[0076] It should be noted that the computer-readable storage medium can be any available medium or data storage device that can be accessed by the processor, including but not limited to magnetic storage (such as floppy disks, hard disks, magnetic tapes, magneto-optical disks (MO), etc.), optical storage (such as CDs, DVDs, BDs, HVDs, etc.), and semiconductor storage (such as ROMs, EPROMs, EEPROMs, non-volatile memories (NAND FLASH), solid-state drives (SSDs)), etc.
[0077] It should also be noted that the terms "first," "second," and the like are used herein to distinguish similar objects, and are not intended to describe a specific order or precedence. It should be understood that the terms used in this manner are interchangeable where appropriate, such that the embodiments of the present invention can be implemented in an order other than that illustrated or described herein. Furthermore, the terms "first" and "second" generally distinguish objects of the same type, and do not limit the number of objects. For example, the first object may be one or more.
[0078] In this disclosure, the term "and / or" describes the relationship between associated objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A exists alone, A and B exist simultaneously, or B exists alone. The character " / " generally indicates that the associated objects are in an "or" relationship.
[0079] In the present invention, the term "plurality" refers to two or more than two, and other quantifiers are similar to it.
[0080] In the present invention, "determining B based on A" means that A is considered when determining B. This is not limited to "determining B based solely on A" and should also include: "determining B based on A and C," "determining B based on A, C, and E," "determining C based on A, and further determining B based on C," etc. It can also include using A as a condition for determining B, for example, "when A meets the first condition, determine B using the first method," "when A meets the second condition, determine B," etc., and "when A meets the third condition, determine B based on the first parameter," etc. Of course, it can also include using A as a factor in determining B, for example, "when A meets the first condition, determine C using the first method, and further determine B based on C," etc.
[0081] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage and optical storage) containing computer-usable program code.
[0082] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer-executable instructions. These computer-executable instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0083] These processor-executable instructions may also be stored in a processor-readable memory that can direct a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the processor-readable memory produce an article of manufacture comprising an instruction device that implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.
[0084] These processor-executable instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are performed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for executing on the computer or other programmable device to implement the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.
[0085] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.
Claims
1. A dam crack acoustic wave detection method based on intelligent discrimination, characterized in that: include: Detecting the crack to be detected based on the drilling acoustic wave detection method to obtain detection waveform data corresponding to the crack to be detected; Inputting the detection waveform data corresponding to the crack to be detected into a crack depth discrimination model to discriminate the crack depth, thereby obtaining the discrimination depth of the crack to be detected; The crack depth discrimination model is a trained U-shaped high-resolution network U-HRNet.
2. The dam crack acoustic wave detection method based on intelligent discrimination according to claim 1 is characterized in that: The crack depth discrimination model is specifically obtained by training through the following steps: Get the training dataset; The U-HRNet is trained based on the training data set to obtain a crack depth discrimination model.
3. The dam crack acoustic wave detection method based on intelligent discrimination according to claim 2 is characterized in that: The obtaining of the training data set includes: Establish multiple concrete models containing aggregates, and add different cracks to different concrete models to obtain multiple crack models; Based on the measurement observation system, the acoustic wave finite difference forward modeling is carried out on each crack model to obtain the simulated detection waveform data; The simulated detection waveform data corresponding to each crack model is used as input data, and the crack depth corresponding to the crack model is used as label data. The input data and the label data are combined to form a training data set.
4. The dam crack acoustic wave detection method based on intelligent discrimination according to claim 3 is characterized in that: The training of U-HRNet based on the training data set includes: Standardize the training data sets corresponding to all fracture models; U-HRNet is trained based on the standardized training dataset.
5. The dam crack acoustic wave detection method based on intelligent discrimination according to claim 3 is characterized in that: The method of adding different cracks to different concrete models includes: For different concrete models, cracks with different crack widths, different crack depths, different extension directions and / or different filling degrees are added.
6. The dam crack acoustic wave detection method based on intelligent discrimination according to claim 1 is characterized in that: The method further comprises: Acquiring supplementary training data based on the detected waveform data and actual depth of the crack to be detected; Based on the supplementary training data, the crack depth discrimination model is fine-tuned through transfer learning.
7. A dam crack acoustic wave detection device based on intelligent discrimination, characterized in that: include: A detection module, configured to detect cracks to be detected based on a drilling acoustic wave detection method, and obtain detection waveform data corresponding to the cracks to be detected; A discrimination module, configured to input the detection waveform data corresponding to the crack to be detected into a crack depth discrimination model to discriminate the crack depth, thereby obtaining a discrimination depth of the crack to be detected; The crack depth discrimination model is a trained U-shaped high-resolution network U-HRNet.
8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the method for detecting dam cracks by acoustic waves based on intelligent discrimination as claimed in any one of claims 1 to 6 is implemented.
9. A non-transitory computer-readable storage medium, characterized in that The non-transitory computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method for detecting dam cracks by acoustic waves based on intelligent discrimination according to any one of claims 1 to 6 is implemented.
10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the method for detecting dam cracks by acoustic waves based on intelligent discrimination as claimed in any one of claims 1 to 6 is implemented.