Internal crack detection method for hydraulic structure

Abstract

The invention discloses an internal crack detection method for a hydraulic structure, and belongs to the technical field of crack detection. Electromagnetic echoes are obtained by emitting sine electromagnetic waves; secondly, segmenting the electromagnetic echo according to the same time length, extracting a local maximum value and a local minimum value, and calculating the growth value and the attenuation value of the maximum value and the minimum value; then, according to the extreme values, calculating an abnormal degree, and constructing an abnormal vector; secondly, analyzing the distribution of extreme values, calculating the instability and constructing an instability vector; and finally, processing the extracted abnormal vector and unstable vector by utilizing a crack feature fusion neural network, thereby obtaining an evaluation value of the internal crack of the hydraulic structure. By comprehensively analyzing the characteristics of the electromagnetic echo signals, the accuracy and reliability of crack detection are effectively improved.

Description

Technical Field

[0001] The invention relates to the technical field of crack detection, and in particular to a method for detecting internal cracks of a hydraulic structure. Background Art

[0002] Hydraulic structures are vital infrastructure in water conservancy projects, including dams, sluice gates, canals, etc. Their structural safety is directly related to the stability of project operation and disaster prevention and mitigation capabilities. At present, traditional hydraulic structure inspection methods mainly rely on manual visual inspection, which has problems such as low inspection accuracy, limited inspection depth, and difficulty in identifying internal micro cracks. Especially for hydraulic structures made of complex materials such as concrete, internal potential cracks are often difficult to detect in time, especially some deep and micro structural cracks.

[0003] In order to improve the efficiency and accuracy of detection, researchers have gradually turned to electromagnetic wave detection technology. As a non-invasive detection method, electromagnetic wave detection can penetrate the surface of a building and detect the internal structure. However, the existing electromagnetic wave detection technology still faces the problem of being difficult to effectively distinguish between normal structural echo signals and abnormal signals related to cracks, which limits the reliability and accuracy of the detection results. Summary of the invention

[0004] In view of the above-mentioned deficiencies in the prior art, the present invention provides a method for detecting internal cracks in a hydraulic structure, which solves the problem of low accuracy in detecting internal cracks in a hydraulic structure in the prior art.

[0005] In order to achieve the above-mentioned invention object, the technical solution adopted by the present invention is: a method for detecting internal cracks of a hydraulic structure, comprising the following steps: S1. Transmit sinusoidal electromagnetic waves to hydraulic structures to obtain electromagnetic echoes; S2, dividing the electromagnetic echo into segments of the same time length, extracting the local maximum and local minimum in each segment, and calculating the maximum growth value and the maximum attenuation value; S3, according to the maximum growth value and the maximum attenuation value of each echo segment, calculate the maximum growth abnormality degree value and the maximum attenuation abnormality degree value, and construct the maximum growth abnormality vector and the maximum attenuation abnormality vector; S4. Calculate the maximum growth instability and the maximum attenuation instability according to the distribution of the maximum growth value and the maximum attenuation value of each echo segment, and construct the maximum growth instability vector and the maximum attenuation instability vector; S5. Use crack feature fusion neural network to process the maximum growth anomaly vector, the maximum decay anomaly vector, the maximum growth unstable vector and the maximum decay unstable vector to obtain the internal crack assessment value of the hydraulic structure.

[0006] Further, S2 includes the following sub-steps: S21, dividing the electromagnetic echo into segments of the same time length to obtain multiple echo segments; S22, marking the local maximum and the local minimum in each echo segment, arranging the local maximum and the local minimum in the order of their occurrence, and constructing a local maximum sequence; S23, calculating the maximum value growth value according to the gap between each local minimum value and the next local maximum value in the local maximum value sequence; S24. Calculate the maximum attenuation value according to the gap between each local maximum value and the next local minimum value in the local maximum value sequence.

[0007] Furthermore, the formula for the maximum growth value in S23 is: , where γ up,j is the jth maximum growth value, E min,j is the jth local minimum, E max,j+1 is the j+1th local maximum, T 1 is the time length between the jth local minimum and the j+1th local maximum, where j is a positive integer; The formula for the maximum attenuation value in S24 is: , where γ down,j is the jth maximum attenuation value, E max,j is the jth local maximum, E min,j+1 is the j+1th local minimum, T 2 is the time length between the jth local maximum and the j+1th local minimum.

[0008] Furthermore, S3 includes the following sub-steps: S31, selecting a maximum growth value greater than a growth threshold in each echo segment as an abnormal maximum growth value; S32, screening out the maximum attenuation value greater than the attenuation threshold in each echo segment as the abnormal maximum attenuation value; S33, in each echo segment, according to each abnormal maximum growth value, calculating the maximum growth abnormality degree value; S34, in each echo segment, according to each abnormal maximum attenuation value, calculating the maximum attenuation abnormality degree value; S35, taking each maximum growth anomaly degree value as an element, constructing a maximum growth anomaly vector; S36. Use each maximum attenuation anomaly degree value as an element to construct a maximum attenuation anomaly vector.

[0009] Furthermore, the formula for calculating the maximum growth abnormality value in S33 is: , where s up is the maximum growth abnormality value, γup,un is the average value of the maximum growth of each anomaly in an echo segment, N up It is the number of abnormal maximum growth values ​​in an echo segment; The formula for calculating the maximum attenuation abnormality value in S34 is: , where s down is the maximum attenuation abnormality value, γ down,un is the average value of the maximum attenuation value of each abnormality in an echo segment, M down It is the number of abnormal maximum attenuation values ​​in an echo segment.

[0010] Further, S4 includes the following sub-steps: S41, calculating the average value of the maximum growth values ​​of all echo segments to obtain the maximum growth average value; S42, in each echo segment, calculating the standard deviation of each maximum growth value to obtain the maximum growth fluctuation value of the echo segment; S43, taking the ratio of the maximum growth fluctuation value to the maximum growth average value as the maximum growth instability; S44, calculating the average value of the maximum attenuation values ​​of all echo segments to obtain the maximum attenuation average value; S45, in each echo segment, calculating the standard deviation of each maximum attenuation value to obtain the maximum attenuation fluctuation value of the echo segment; S46, taking the ratio of the maximum attenuation fluctuation value to the maximum attenuation average value as the maximum attenuation instability; S47, taking the maximum growth instability as an element, constructing a maximum growth instability vector; S48. Taking the maximum attenuation instability as an element, constructing a maximum attenuation instability vector.

[0011] Furthermore, the crack feature fusion neural network in S5 includes: a growth data fusion layer, a decay data fusion layer, a first matrix construction layer, a second matrix construction layer, a first CNN network, a second CNN network, an adder A1, a feature self-adjustment layer, and a fully connected layer; The first input end of the growth data fusion layer is used to input the maximum growth anomaly vector, the second input end is used to input the maximum growth unstable vector, and the output end is connected to the input end of the first matrix construction layer; the first input end of the attenuation data fusion layer is used to input the maximum attenuation anomaly vector, the second input end is used to input the maximum attenuation unstable vector, and the output end is connected to the input end of the second matrix construction layer; the input end of the first CNN network is connected to the output end of the first matrix construction layer; the input end of the second CNN network is connected to the output end of the second matrix construction layer; the first input end of the adder A1 is connected to the output end of the first CNN network, the second input end is connected to the output end of the second CNN network, and the output end is connected to the input end of the feature self-adjustment layer; the output end of the feature self-adjustment layer is connected to the input end of the fully connected layer; the output end of the fully connected layer serves as the output end of the crack feature fusion neural network.

[0012] Furthermore, the specific processing processes of the growth data fusion layer and the attenuation data fusion layer are: extracting features from each element input at the first input end through the tanh function, and concatenating the features to form a first concatenated vector; extracting features from each element input at the second input end through the tanh function, and concatenating the features to form a second concatenated vector, and multiplying the first concatenated vector and the second concatenated vector element by element to obtain a fused feature vector.

[0013] Furthermore, the expressions of the matrix construction layers are: , where H is the fused feature matrix, h is the fused feature vector, and T is the transpose operation.

[0014] Furthermore, the expression of the feature self-adjusting layer is: ,in, is the nth feature output by the feature self-adjustment layer, x n is the nth feature output by adder A1, w n is the nth weight of the feature self-adjusting layer, b n is the nth bias of the feature self-adjusting layer, e is a natural constant, and n is a positive integer.

[0015] The beneficial effects of the present invention are: 1. In order to better extract crack-related features, the present invention emits sinusoidal electromagnetic waves to hydraulic structures. Sinusoidal electromagnetic waves have waveform continuity and can provide stable and uniform energy distribution. Therefore, the extracted maximum growth value and maximum attenuation value can reflect the propagation characteristics of sinusoidal electromagnetic waves inside hydraulic structures.

[0016] 2. The present invention processes the electromagnetic echo in segments, extracts the local maximum and local minimum by segment, calculates the maximum growth value and the maximum attenuation value to reflect the sudden change of the electromagnetic echo, and then calculates the maximum growth abnormality degree value and the maximum attenuation abnormality degree value to reflect the abnormal situation of the maximum growth and the abnormal situation of the maximum attenuation in a segment.

[0017] 3. The present invention calculates the maximum growth instability and the maximum attenuation instability according to the distribution of each maximum growth value and the maximum attenuation value in each segment, reflecting the fluctuation of the maximum growth value and the maximum attenuation value.

[0018] 4. When cracks exist, the cracks will cause the propagation of electromagnetic waves to be distorted, resulting in non-uniform energy distribution. Cracks will cause medium discontinuity, resulting in abnormal reflection, scattering and attenuation characteristics of the wave. Therefore, the present invention extracts the maximum growth abnormality value, the maximum attenuation abnormality value, and the maximum growth instability and the maximum attenuation instability to reflect the abnormal morphology of the electromagnetic echo, effectively distinguishes the normal structure echo signal and the abnormal signal of the crack, and then uses the crack feature fusion neural network to process these data, thereby improving the internal crack detection accuracy of hydraulic structures. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 The present invention is a flow chart of a method for detecting internal cracks in hydraulic structures; Figure 2 It is a 50MHz sinusoidal electromagnetic wave; Figure 3 Schematic diagram of the crack feature fusion neural network structure. DETAILED DESCRIPTION

[0020] The specific implementation modes of the present invention are described below so that those skilled in the art can understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific implementation modes. For those of ordinary skill in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the attached claims, these changes are obvious, and all inventions and creations utilizing the concept of the present invention are protected.

[0021] like Figure 1 As shown, a method for detecting internal cracks of a hydraulic structure comprises the following steps: S1. Transmit sinusoidal electromagnetic waves to hydraulic structures to obtain electromagnetic echoes; S2, dividing the electromagnetic echo into segments of the same time length, extracting the local maximum and local minimum in each segment, and calculating the maximum growth value and the maximum attenuation value; S3, according to the maximum growth value and the maximum attenuation value of each echo segment, calculate the maximum growth abnormality degree value and the maximum attenuation abnormality degree value, and construct the maximum growth abnormality vector and the maximum attenuation abnormality vector; S4. Calculate the maximum growth instability and the maximum attenuation instability according to the distribution of the maximum growth value and the maximum attenuation value of each echo segment, and construct the maximum growth instability vector and the maximum attenuation instability vector; S5. Use crack feature fusion neural network to process the maximum growth anomaly vector, the maximum decay anomaly vector, the maximum growth unstable vector and the maximum decay unstable vector to obtain the internal crack assessment value of the hydraulic structure.

[0022] In this embodiment, if Figure 2 As shown, the expression of the sinusoidal electromagnetic wave in S1 is: , where E(t) is a sinusoidal electromagnetic wave, E is the amplitude of the emitted sinusoidal electromagnetic wave, sin is a sinusoidal function, f is the frequency value, and φ is the phase.

[0023] In this embodiment, S2 includes the following sub-steps: S21, dividing the electromagnetic echo into segments of the same time length to obtain multiple echo segments; S22, marking the local maximum and the local minimum in each echo segment, arranging the local maximum and the local minimum in the order of their occurrence, and constructing a local maximum sequence; For example: an echo segment is [0.1, 0.3, 0.5, 0.6, 0.7, 0.4, 0.3, 0.2, 0.1, 0.5, 0.6, 0.8, 0.3, 0.2, 0.4], the local maximum is 0.7, 0.8, the local minimum is 0.1, 0.2, then the local maximum sequence is [0.7, 0.1, 0.8, 0.2]; S23, calculating the maximum value growth value according to the gap between each local minimum value and the next local maximum value in the local maximum value sequence; S24. Calculate the maximum attenuation value according to the gap between each local maximum value and the next local minimum value in the local maximum value sequence.

[0024] In the present invention, the local minimum is defined as: the amplitude of the selected point is lower than the amplitudes of the neighboring points on both sides; the local maximum is defined as: the amplitude of the selected point is higher than the amplitudes of the neighboring points on both sides.

[0025] In this embodiment, the formula for the maximum growth value in S23 is: , where γ up,j is the jth maximum growth value, E min,j is the jth local minimum, E max,j+1 is the j+1th local maximum, T 1 is the time length between the jth local minimum and the j+1th local maximum, j is a positive integer, E min,j and E max,j+1adjacent; The formula for the maximum attenuation value in S24 is: , where γ down,j is the jth maximum attenuation value, E max,j is the jth local maximum, E min,j+1 is the j+1th local minimum, T 2 is the time length between the jth local maximum and the j+1th local minimum, E max,j and E min,j+1 Adjacent.

[0026] From the perspective of structural integrity assessment, a healthy building structure allows electromagnetic waves to propagate in a relatively uniform and predictable manner. When there are internal cracks, the propagation path of electromagnetic waves will change, causing non-uniform energy distribution. By extracting the maximum growth value and the maximum attenuation value, it is possible to sensitively capture small changes in this propagation characteristic.

[0027] In the analysis of wave propagation mechanism, when a sinusoidal electromagnetic wave propagates in a homogeneous medium, its energy attenuation and growth show regularity. Once it encounters a crack, the medium will be discontinuous, resulting in abnormal reflection, scattering and attenuation characteristics of the wave. The maximum growth value reflects the speed of wave energy rise, and the maximum attenuation value reflects the speed of wave energy attenuation.

[0028] In this embodiment, S3 includes the following sub-steps: S31, selecting a maximum growth value greater than a growth threshold in each echo segment as an abnormal maximum growth value; S32, screening out the maximum attenuation value greater than the attenuation threshold in each echo segment as the abnormal maximum attenuation value; S33, in each echo segment, according to each abnormal maximum growth value, calculating the maximum growth abnormality degree value; S34, in each echo segment, according to each abnormal maximum attenuation value, calculating the maximum attenuation abnormality degree value; S35, taking each maximum growth anomaly degree value as an element, constructing a maximum growth anomaly vector; S36. Use each maximum attenuation anomaly degree value as an element to construct a maximum attenuation anomaly vector.

[0029] In the present invention, the elements in the vector are arranged in the order of their occurrence in time.

[0030] In this embodiment, the growth threshold is a threshold set for the maximum growth value, and the attenuation threshold is a threshold set for the maximum attenuation value. The growth threshold and the attenuation threshold can be specifically set based on experience and experiments. For example, the present invention emits a 50MHz sinusoidal electromagnetic wave. For a 50MHz sinusoidal electromagnetic wave, the length of one cycle is 20 nanoseconds, and the distance between the adjacent local maximum and minimum values ​​is half a cycle length. For a 50MHz sinusoidal electromagnetic wave: the maximum growth value and the maximum attenuation value are both 2E / (10 nanoseconds). Considering the energy attenuation, the maximum growth value and the maximum attenuation value of the electromagnetic wave echo are both 2Eμ / (10 nanoseconds), where E is the amplitude of the sinusoidal electromagnetic wave and μ is the attenuation ratio. Therefore, 2Eμ / (10 nanoseconds) can be set as the corresponding threshold.

[0031] The present invention selects the maximum growth value and the maximum attenuation value greater than the threshold value to characterize the abnormal signal characteristics related to the crack, that is, the abnormal maximum growth value and the abnormal maximum attenuation value in S31 and S32 are the abnormal signal characteristics related to the crack, and the maximum growth value less than or equal to the growth threshold and the maximum attenuation value less than or equal to the attenuation threshold are regarded as normal echo signal characteristics.

[0032] In this embodiment, the formula for calculating the maximum growth abnormality value in S33 is: , where s up is the maximum growth abnormality value, γ up,un is the average value of the maximum growth of each anomaly in an echo segment, N up It is the number of abnormal maximum growth values ​​in an echo segment; The formula for calculating the maximum attenuation abnormality value in S34 is: , where s down is the maximum attenuation abnormality value, γ down,un is the average value of the maximum attenuation value of each abnormality in an echo segment, M down It is the number of abnormal maximum attenuation values ​​in an echo segment.

[0033] The present invention calculates the average value of each abnormal maximum growth value and each abnormal maximum attenuation value of each echo segment to reflect the abnormal degree of an echo segment, and then uses the number of abnormal maximum growth values ​​and abnormal maximum attenuation values ​​to further enhance the abnormal degree. When the average value is larger and the number is larger, the abnormal degree value is larger.

[0034] In this embodiment, S4 includes the following sub-steps: S41, calculating the average value of the maximum growth values ​​of all echo segments to obtain the maximum growth average value; S42, in each echo segment, calculating the standard deviation of each maximum growth value to obtain the maximum growth fluctuation value of the echo segment, wherein the maximum growth fluctuation value is equal to the standard deviation of each maximum growth value; S43, taking the ratio of the maximum growth fluctuation value to the maximum growth average value as the maximum growth instability; S44, calculating the average value of the maximum attenuation values ​​of all echo segments to obtain the maximum attenuation average value; S45, in each echo segment, calculating the standard deviation of each maximum attenuation value to obtain the maximum attenuation fluctuation value of the echo segment, wherein the maximum attenuation fluctuation value is equal to the standard deviation of each maximum attenuation value; S46, taking the ratio of the maximum attenuation fluctuation value to the maximum attenuation average value as the maximum attenuation instability; S47, taking the maximum growth instability as an element, constructing a maximum growth instability vector; S48. Taking the maximum attenuation instability as an element, construct a maximum attenuation instability vector.

[0035] The maximum growth instability reflects the fluctuation characteristics of the electromagnetic wave energy accumulation process. In a uniform and complete building structure, the energy accumulation of electromagnetic waves should show a relatively stable and predictable trend. When there are internal cracks, this energy accumulation process will show significant irregularities. By quantifying this irregularity, tiny structural anomalies can be sensitively captured, providing more accurate quantitative indicators for crack detection.

[0036] The maximum attenuation instability reveals the complex dynamic process of electromagnetic wave energy loss. In an ideal homogeneous medium, the attenuation of electromagnetic waves should follow relatively stable physical laws. However, internal cracks will cause discontinuity in the propagation path of electromagnetic waves, resulting in significant changes in the energy attenuation pattern. Calculating the maximum attenuation instability is actually quantifying the fluctuation characteristics of this non-ideal attenuation process, providing a new analytical dimension for identifying potential cracks.

[0037] like Figure 3 As shown, the crack feature fusion neural network in S5 includes: a growth data fusion layer, a decay data fusion layer, a first matrix construction layer, a second matrix construction layer, a first CNN network, a second CNN network, an adder A1, a feature self-adjustment layer and a fully connected layer; The first input end of the growth data fusion layer is used to input the maximum growth anomaly vector, the second input end is used to input the maximum growth unstable vector, and the output end is connected to the input end of the first matrix construction layer; the first input end of the attenuation data fusion layer is used to input the maximum attenuation anomaly vector, the second input end is used to input the maximum attenuation unstable vector, and the output end is connected to the input end of the second matrix construction layer; the input end of the first CNN network is connected to the output end of the first matrix construction layer; the input end of the second CNN network is connected to the output end of the second matrix construction layer; the first input end of the adder A1 is connected to the output end of the first CNN network, the second input end is connected to the output end of the second CNN network, and the output end is connected to the input end of the feature self-adjustment layer; the output end of the feature self-adjustment layer is connected to the input end of the fully connected layer; the output end of the fully connected layer serves as the output end of the crack feature fusion neural network.

[0038] In this embodiment, the specific processing processes of the growth data fusion layer and the attenuation data fusion layer are: extracting features from each element input at the first input end through the tanh function, and splicing the features to form a first splicing vector; extracting features from each element input at the second input end through the tanh function, and splicing the features to form a second splicing vector, and multiplying the first splicing vector and the second splicing vector element by element to obtain a fused feature vector.

[0039] The expression of tanh function feature extraction is: ,in, is the mth feature extracted by the tanh function, r m is the mth element, w m For r m The weight of b m For r m , m is a positive integer, and f is the tanh function.

[0040] For the growth data fusion layer, its input data includes: the maximum growth anomaly vector and the maximum growth instability vector. Therefore, after processing each element through the tanh function, the feature splicing vector corresponding to the maximum growth anomaly vector and the feature splicing vector corresponding to the maximum growth instability vector are obtained. These two splicing vectors are multiplied element by element to achieve the fusion of the maximum growth anomaly features belonging to the same segment of electromagnetic echo.

[0041] For the attenuation data fusion layer, its input data includes: the maximum attenuation anomaly vector and the maximum attenuation instability vector. Therefore, after processing each element through the tanh function, the feature splicing vector corresponding to the maximum attenuation anomaly vector and the feature splicing vector corresponding to the maximum attenuation instability vector are obtained. These two splicing vectors are multiplied element by element to achieve the fusion of the maximum attenuation anomaly features belonging to the same segment of electromagnetic echo.

[0042] In this embodiment, the expressions of the matrix construction layers are all: , where H is the fused feature matrix, h is the fused feature vector, and T is the transpose operation.

[0043] The present invention uses a matrix construction layer to transform the vectors output by the growth data fusion layer and the attenuation data fusion layer into matrices, which is convenient for the CNN network to extract features. Then, adder A1 is used to perform bitwise addition, and the maximum growth and maximum attenuation features are fused to achieve a comprehensive evaluation of the crack situation.

[0044] In this embodiment, the expression of the characteristic self-adjusting layer is: ,in, is the nth feature output by the feature self-adjustment layer, x n is the nth feature output by adder A1, w n is the nth weight of the feature self-adjusting layer, b n is the nth bias of the feature self-adjusting layer, e is a natural constant, and n is a positive integer.

[0045] The present invention adopts characteristic self-adjusting layer allocation adjustment ratio , which enables the crack feature fusion neural network to have the ability to automatically adjust feature attention.

[0046] In this embodiment, the weights and biases in the crack feature fusion neural network can be obtained by training through the existing gradient descent method. For example, multiple maximum growth anomaly vectors, maximum decay anomaly vectors, maximum growth unstable vectors, and maximum decay unstable vectors are collected as samples, and the internal crack evaluation values ​​of the manually labeled hydraulic structures are used as labels. Each sample is input into the crack feature fusion neural network. According to the difference between the actual evaluation value output by the crack feature fusion neural network and the label, the gap between the actual and the label is obtained based on the existing mean square error loss function, and each weight and bias are updated according to the gap. The formula for updating the weight and bias in the implementation of this application is: , , where w k is the weight of the kth training, w k-1 is the weight of the k-1th training, η is a hyperparameter, η is used to control the step size of each parameter update, L is the mean square error loss function, b k is the bias during the kth training, b k-1 is the bias during the k-1th training, and k is the number of training times.

[0047] In this embodiment, the range of the label of the training process is set to 0~1, so the output range of the crack feature fusion neural network is between 0~1. When the internal crack evaluation value output by the fully connected layer is greater than the evaluation threshold of 0.5, such as the evaluation value of 0.7, it indicates that the possibility of cracks is greater, and it can be preliminarily judged that there are cracks inside the hydraulic structure. When the internal crack evaluation value output by the fully connected layer is less than or equal to 0.5, such as the evaluation value of 0.3, it indicates that the possibility of cracks is smaller. The setting of specific thresholds can be verified through multiple experiments. The label setting range and the evaluation threshold are set according to specific needs and are not limited to the values ​​given in the present invention. The label setting range is artificially set during the training process. The specific setting range does not affect the implementation of the present invention. The evaluation threshold can be taken as the middle value of the label range, which is convenient for dividing the output into two categories. When it is greater than the evaluation threshold, the possibility of cracks is greater, and when it is less than the evaluation threshold, the possibility of cracks is smaller.

[0048] In order to better extract crack-related features, the present invention emits sinusoidal electromagnetic waves to hydraulic structures. Sinusoidal electromagnetic waves have waveform continuity and can provide stable and uniform energy distribution. Therefore, the extracted maximum growth value and maximum attenuation value can reflect the propagation characteristics of sinusoidal electromagnetic waves inside hydraulic structures.

[0049] The present invention processes electromagnetic echoes in sections, extracts local maximum and local minimum values ​​by section, calculates maximum growth values ​​and maximum attenuation values ​​to reflect the mutation of electromagnetic echoes, and then calculates maximum growth abnormality degree values ​​and maximum attenuation abnormality degree values ​​to reflect the abnormal situation of maximum growth and maximum attenuation in a section.

[0050] The present invention calculates the maximum growth instability and the maximum attenuation instability according to the distribution of each maximum growth value and the maximum attenuation value in each segment, and reflects the fluctuation of the maximum growth value and the maximum attenuation value.

[0051] When cracks exist, the cracks will cause the propagation of electromagnetic waves to be distorted, resulting in non-uniform energy distribution. Cracks will cause medium discontinuity, resulting in abnormal reflection, scattering and attenuation characteristics of the wave. Therefore, the present invention extracts the maximum growth abnormality degree value, the maximum attenuation abnormality degree value, and the maximum growth instability and the maximum attenuation instability to reflect the abnormal morphology of the electromagnetic echo, effectively distinguishes the normal structure echo signal and the abnormal signal of the crack, and then uses the crack feature fusion neural network to process these data, thereby improving the internal crack detection accuracy of hydraulic structures.

[0052] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for detecting internal cracks in a hydraulic structure, characterized in that: The following steps are involved: S1. Transmit sinusoidal electromagnetic waves to hydraulic structures to obtain electromagnetic echoes; S2, dividing the electromagnetic echo into segments of the same time length, extracting the local maximum and local minimum in each segment, and calculating the maximum growth value and the maximum attenuation value; S3, according to the maximum growth value and the maximum attenuation value of each echo segment, calculate the maximum growth abnormality degree value and the maximum attenuation abnormality degree value, and construct the maximum growth abnormality vector and the maximum attenuation abnormality vector; S4. Calculate the maximum growth instability and the maximum attenuation instability according to the distribution of the maximum growth value and the maximum attenuation value of each echo segment, and construct the maximum growth instability vector and the maximum attenuation instability vector; S5. Use crack feature fusion neural network to process the maximum growth anomaly vector, the maximum decay anomaly vector, the maximum growth unstable vector and the maximum decay unstable vector to obtain the internal crack assessment value of the hydraulic structure.

2. The internal crack detection method of a hydraulic structure according to claim 1, characterized in that: The S2 comprises the following sub-steps: S21, dividing the electromagnetic echo into segments of the same time length to obtain multiple echo segments; S22, marking the local maximum and the local minimum in each echo segment, arranging the local maximum and the local minimum in the order of their occurrence, and constructing a local maximum sequence; S23, calculating the maximum value growth value according to the gap between each local minimum value and the next local maximum value in the local maximum value sequence; S24. Calculate the maximum attenuation value according to the gap between each local maximum value and the next local minimum value in the local maximum value sequence.

3. The internal crack detection method of a hydraulic structure according to claim 2, characterized in that: The formula for the maximum growth value in S23 is: , where γ up,j is the jth maximum growth value, E min,j is the jth local minimum, E max,j+1 is the j+1th local maximum, T1 is the time length between the jth local minimum and the j+1th local maximum, and j is a positive integer; The formula for the maximum attenuation value in S24 is: , where γ down,j is the jth maximum attenuation value, E max,j is the jth local maximum, E min,j+1 is the j+1th local minimum, and T2 is the time length between the jth local maximum and the j+1th local minimum.

4. The internal crack detection method of a hydraulic structure according to claim 1, characterized in that: The S3 comprises the following sub-steps: S31, selecting a maximum growth value greater than a growth threshold in each echo segment as an abnormal maximum growth value; S32, screening out the maximum attenuation value greater than the attenuation threshold in each echo segment as the abnormal maximum attenuation value; S33, in each echo segment, according to each abnormal maximum growth value, calculating the maximum growth abnormality degree value; S34, in each echo segment, according to each abnormal maximum attenuation value, calculating the maximum attenuation abnormality degree value; S35, taking each maximum growth anomaly degree value as an element, constructing a maximum growth anomaly vector; S36. Use each maximum attenuation anomaly degree value as an element to construct a maximum attenuation anomaly vector.

5. The internal crack detection method of a hydraulic structure according to claim 4, characterized in that: The formula for calculating the maximum growth abnormality value in S33 is: , where s up is the maximum growth abnormality value, γ up,un is the average value of the abnormal maximum growth value in a segment of echo, N up It is the number of abnormal maximum growth values ​​in an echo segment; The formula for calculating the maximum attenuation abnormality value in S34 is: , where s down is the maximum attenuation abnormality value, γ down,un is the average value of the maximum attenuation value of each abnormality in an echo segment, M down It is the number of abnormal maximum attenuation values ​​in an echo segment.

6. The internal crack detection method of a hydraulic structure according to claim 1, characterized in that: The S4 comprises the following sub-steps: S41, calculating the average value of the maximum growth values ​​of all echo segments to obtain the maximum growth average value; S42, in each echo segment, calculating the standard deviation of each maximum growth value to obtain the maximum growth fluctuation value of the echo segment; S43, taking the ratio of the maximum growth fluctuation value to the maximum growth average value as the maximum growth instability; S44, calculating the average value of the maximum attenuation values ​​of all echo segments to obtain the maximum attenuation average value; S45, in each echo segment, calculating the standard deviation of each maximum attenuation value to obtain the maximum attenuation fluctuation value of the echo segment; S46, taking the ratio of the maximum attenuation fluctuation value to the maximum attenuation average value as the maximum attenuation instability; S47, taking the maximum growth instability as an element, constructing a maximum growth instability vector; S48. Taking the maximum attenuation instability as an element, constructing a maximum attenuation instability vector.

7. The internal crack detection method of a hydraulic structure according to claim 1, characterized in that: The crack feature fusion neural network in S5 includes: a growth data fusion layer, a decay data fusion layer, a first matrix construction layer, a second matrix construction layer, a first CNN network, a second CNN network, an adder A1, a feature self-adjustment layer and a fully connected layer; The first input end of the growth data fusion layer is used to input the maximum growth abnormal vector, the second input end is used to input the maximum growth unstable vector, and the output end is connected to the input end of the first matrix construction layer; The first input end of the attenuation data fusion layer is used to input the maximum attenuation abnormal vector, the second input end is used to input the maximum attenuation unstable vector, and the output end is connected to the input end of the second matrix construction layer; The input end of the first CNN network is connected to the output end of the first matrix construction layer; The input end of the second CNN network is connected to the output end of the second matrix construction layer; The first input end of the adder A1 is connected to the output end of the first CNN network, the second input end thereof is connected to the output end of the second CNN network, and the output end thereof is connected to the input end of the feature self-adjusting layer; The output end of the feature self-adjustment layer is connected to the input end of the fully connected layer; The output end of the fully connected layer serves as the output end of the crack feature fusion neural network.

8. The method for detecting internal cracks of hydraulic structures according to claim 7, characterized in that: The specific processing processes of the growth data fusion layer and the attenuation data fusion layer are: extracting features from each element input at the first input end through the tanh function, and splicing the features to form a first splicing vector; extracting features from each element input at the second input end through the tanh function, and splicing the features to form a second splicing vector, and multiplying the first splicing vector and the second splicing vector element by element to obtain a fused feature vector.

9. The method for detecting internal cracks of a hydraulic structure according to claim 8, characterized in that: The expressions of the matrix construction layers are all: , where H is the fused feature matrix, h is the fused feature vector, and T is the transpose operation.

10. The internal crack detection method of a hydraulic structure according to claim 8, characterized in that: The expression of the characteristic self-adjusting layer is: ,in, is the nth feature output by the feature self-adjustment layer, x n is the nth feature output by adder A1, w n is the nth weight of the feature self-adjusting layer, b n is the nth bias of the feature self-adjusting layer, e is a natural constant, and n is a positive integer.

Images