Dam hidden danger electromagnetic data interpretation method and system based on anchoring effect

Abstract

The invention belongs to the technical field of dam internal hidden danger detection, particularly provides a dam hidden danger electromagnetic data interpretation method and system based on an anchoring effect, and aims at solving a deviation curve of a detection target based on an anchoring curve, reflecting an apparent resistivity distribution rule according to the deviation curve, and deducing a danger area. The method specifically comprises the following steps: acquiring an attenuation curve of original transient electromagnetism; carrying out noise reduction processing on the attenuation curve, then evaluating the signal-to-noise ratio of the signal and removing waste channels; performing time sequence synthesis processing on the effective signals, and screening out an anchoring curve; calculating a deviation curve based on the anchoring curve; constructing a resistivity distribution model, and calculating a reference curve; correspondingly multiplying the deviation curve by the reference curve according to a time sequence to obtain an inversion curve; calculating apparent resistivity based on the inversion curve, and obtaining a cloud picture according to the apparent resistivity; and deducing a dam hidden danger area based on the cloud picture. According to the invention, the interpretation speed and precision of dam hidden danger detection data can be effectively improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of detecting hidden dangers inside dams, and particularly relates to a method and system for interpreting electromagnetic data of hidden dangers in dams based on the anchoring effect. Background Art

[0002] As an important water conservancy project facility, the safety status of a dam is directly related to the safety of people's lives and property and the stable development of social economy in the downstream area. The transient electromagnetic detection technology, as an efficient and non-destructive geophysical detection method, has been widely used in the detection of hidden dangers in dams in recent years. However, due to the special boundary conditions of dams and the complexity of the detection environment, it brings great difficulties to accurately locate the hidden dangers. It is mainly manifested in two aspects: First, the dam has a trapezoidal cross-section, with water on one side and air on the other side. Applying traditional inversion methods based on semi-space or full-space has a large error. Second, as an important water conservancy building, there are usually a large number of electrical and other auxiliary facilities near the dam body, and there is strong physical field interference. The data obtained by transient electromagnetic detection often contains a large amount of noise and interference information. Summary of the Invention

[0003] The present invention provides a method and system for interpreting electromagnetic data of hidden dangers in dams based on the anchoring effect, which can greatly improve the accuracy and efficiency of detecting hidden dangers in dams and provide technical support for the safety evaluation and maintenance of dams.

[0004] A method for interpreting electromagnetic data of hidden dangers in dams based on the anchoring effect, the method comprising:

[0005] Exciting the electromagnetic field in the hidden danger area inside the dam body, collecting the original transient electromagnetic field signal, and obtaining the attenuation curve of the original transient electromagnetic field changing with time;

[0006] Performing superposition processing on the attenuation curves at the same position, suppressing the uncorrelated noise between the peaks of the original transient electromagnetic field signals, and performing correlated noise suppression processing on the superposed attenuation curves to obtain the transient electromagnetic attenuation curve after noise suppression;

[0007] Performing signal evaluation and waste channel rejection on the transient electromagnetic attenuation curve after noise suppression to obtain the transient electromagnetic attenuation curve after removing the waste channel;

[0008] Performing time series synthesis processing on the transient electromagnetic attenuation curve after removing the waste channel to screen out the anchoring curve;

[0009] Based on the anchoring curve, calculating the deviation between the attenuation curve of the original transient electromagnetic field changing with time and the anchoring curve to form a deviation curve;

[0010] Based on the existing geological data and monitoring data of the dam, constructing a resistance distribution stratified model, and calculating a reference curve based on the resistance distribution stratified model;

[0011] Multiply the deviation curve and the reference curve corresponding to each other in time series to obtain an inversion curve;

[0012] Based on the inversion curve, calculate the apparent resistivity and obtain a cloud map according to the apparent resistivity;

[0013] Based on the cloud map, infer the hidden danger location of the dam, and complete the interpretation of the transient electromagnetic detection data of the dam hidden danger based on the anchoring effect.

[0014] Preferably, the method for superimposing the attenuation curves at the same position includes:

[0015] Superimpose the attenuation curves at the same position, average the superimposed attenuation curves, and complete the superimposing process of the attenuation curves; wherein, the superimposing formula is as follows:

[0016]

[0017] In the formula, represents the summation of the original transient electromagnetic field signals, and n represents the number of signals.

[0018] Preferably, the method for suppressing the correlated noise of the superimposed attenuation curves includes:

[0019] Adopt the first-order difference method to pre-emphasize the superimposed attenuation curve to obtain a pre-emphasized signal;

[0020] Segment and window the pre-emphasized signal;

[0021] Perform short-time Fourier transform on the windowed pre-emphasized signal to obtain the transient electromagnetic field signal in the frequency domain;

[0022] Use a filter to filter out the noise signal of the transient electromagnetic field signal in the frequency domain to obtain a denoised signal;

[0023] Merge and convert the denoised signal into the time domain to complete the suppression process of the correlated noise.

[0024] Preferably, the method for performing time series synthesis on the transient electromagnetic attenuation curve after removing the waste channels to obtain the anchoring curve includes:

[0025] When distinguishing the preset weak areas between the dam layers, use the time series synthesis method to take the average value of the attenuation curves of the original transient electromagnetic fields at each measuring point of the dam at the same moment as the anchoring curve value at the same moment; arrange the anchoring curve values at each moment in time series to obtain the anchoring curve;

[0026] The method for obtaining the anchoring curve also includes the direct selection method and the first value selection point method;

[0027] When the dam structure form meets the preset requirements, the direct selection method is used to directly select the measuring points without anomalies, and the attenuation curve of the measuring points without anomalies is used as the anchoring curve;

[0028] When conducting a comprehensive comparison of the entire dam area, the first-value selection method is used to select the first original transient electromagnetic field signals collected at each measuring point, and the attenuation curve corresponding to the median of all the selected first original transient electromagnetic field signals is used as the anchoring curve.

[0029] Preferably, the calculation formula of the deviation curve is:

[0030]

[0031] In the formula, B z represents the attenuation curve value, and B m represents the anchoring curve value.

[0032] The present invention also provides an electromagnetic data interpretation system for dam hidden dangers based on the anchoring effect, which is used to implement the method. The system includes:

[0033] A signal acquisition module, which is used to excite the electromagnetic field in the hidden danger area inside the dam, collect the original transient electromagnetic field signals, and obtain the attenuation curve of the original transient electromagnetic field changing with time;

[0034] A denoising module, which is used to perform superposition processing on the attenuation curves at the same position, suppress the uncorrelated noise between the peaks of the original transient electromagnetic field signals, and perform correlated noise suppression processing on the superposed attenuation curves to obtain the transient electromagnetic attenuation curve after noise suppression;

[0035] A waste channel removal module, which is used to perform signal evaluation and waste channel removal on the attenuation curve after noise suppression to obtain the transient electromagnetic attenuation curve with waste channels removed;

[0036] An anchoring curve acquisition module, which is used to perform time series synthesis processing on the transient electromagnetic attenuation curve with waste channels removed and screen out the anchoring curve;

[0037] A deviation curve acquisition module, which is used to calculate the deviation between the attenuation curve of the original transient electromagnetic field changing with time and the anchoring curve based on the anchoring curve to form a deviation curve;

[0038] A reference curve acquisition module, which is used to construct a resistance distribution stratified model based on the existing geological data and monitoring data of the dam, and calculate the reference curve based on the resistance distribution stratified model;

[0039] An inversion curve acquisition module, which is used to multiply the deviation curve and the reference curve in time series correspondence to obtain an inversion curve;

[0040] A cloud map acquisition module, configured to calculate apparent resistivity based on the inversion curve and obtain a cloud map according to the apparent resistivity;

[0041] A hidden danger location acquisition module, configured to infer the hidden danger location of the dam based on the cloud map, and complete the interpretation of the transient electromagnetic detection data of the dam hidden danger based on the anchoring effect.

[0042] Preferably, the denoising module includes a stacking processing unit, configured to perform stacking processing on the attenuation curves at the same position; the stacking processing unit includes:

[0043] A stacking subunit, configured to stack the attenuation curves at the same position;

[0044] An averaging subunit, configured to average the stacked attenuation curves to complete the stacking processing of the attenuation curves; wherein, the stacking formula is as follows:

[0045]

[0046] In the formula, represents the summation of the original transient electromagnetic field signals, and n represents the number of signals.

[0047] Preferably, the denoising module further includes a correlated noise filtering unit, configured to perform correlated noise suppression processing on the attenuation curves after stacking processing; the correlated noise filtering unit includes:

[0048] A pre-emphasis subunit, configured to perform pre-emphasis on the attenuation curves after stacking processing by using the first-order difference method to obtain pre-emphasized signals;

[0049] A segmentation and windowing subunit, configured to segment and window the pre-emphasized signals;

[0050] A time-frequency conversion subunit, configured to perform short-time Fourier transform on the windowed pre-emphasized signals to obtain frequency-domain transient electromagnetic field signals;

[0051] A denoising subunit, configured to filter out the noise signals of the frequency-domain transient electromagnetic field signals by using a filter to obtain denoised signals;

[0052] A frequency-time conversion subunit, configured to merge and convert the denoised signals into the time domain to complete the correlated noise suppression processing.

[0053] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein when the processor executes the program, the method described above is implemented.

[0054] The present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed, the method described above is implemented.

[0055] Compared with the prior art, the beneficial effects of the present invention are as follows: The idea of the anchoring effect is introduced and applied to the interpretation of transient electromagnetic detection data for hidden dangers of dams. This interpretation method using the anchoring effect is not mentioned in the existing background technology. Operations such as combining methods of mathematical statistics, prior information fusion, etc. with the anchoring effect to eliminate interference signals and establish anchoring curves are different from traditional methods for interpreting detection data of hidden dangers of dams, and have a certain degree of novelty. Aiming at the special boundary conditions of dams (trapezoidal cross-section, one side facing water and the other side facing air) and complex detection environments (physical field interference caused by a large number of electrical auxiliary facilities), as well as the problem of large errors in traditional inversion methods, a new solution idea is proposed. By introducing the anchoring effect and establishing an anchoring curve to solve the abnormal response, the deficiencies of traditional methods can be solved, and it has prominent substantive features. The combination of various specific methods and steps, such as various ways of eliminating interference and various methods of selecting anchoring curves, etc., cooperate with each other to improve the detection accuracy and efficiency, and there is a significant progress compared with the prior art. The inventive method can improve the accuracy and efficiency of detecting hidden dangers of dams, provide technical support for the safety evaluation and maintenance of dams, has practical application value, can be manufactured and used in the field of detecting hidden dangers of dams, and produces positive effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the technical solutions of the present invention, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0057] Figure 1 It is a flowchart of the method for interpreting electromagnetic data of hidden dangers of dams based on the anchoring effect according to an embodiment of the present invention;

[0058] Figure 2 It is a schematic structural diagram of the system for interpreting electromagnetic data of hidden dangers of dams based on the anchoring effect according to an embodiment of the present invention;

[0059] Figure 3 It is a multi-channel diagram of interference data before and after signal elimination according to an embodiment of the present invention;

[0060] Figure 4 It is a schematic diagram of the attenuation curve according to an embodiment of the present invention;

[0061] Figure 5 It is a multi-channel diagram of deviation data according to an embodiment of the present invention;

[0062] Figure 6 It is a cloud map according to an embodiment of the present invention;

[0063] Brief Description of the Drawings: 1-1 is the primary magnetic field excitation unit; 1-2 is the waveform regulation unit, 1-3 is the magnetic field reception unit, 1-4 is the acquisition and power supply unit; 2 is the data analysis component. Detailed Implementation Manner

[0064] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0065] To make the above objects, features, and advantages of the present invention more clearly understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific implementation manners.

[0066] Embodiment 1

[0067] As Figure 1 shown, the electromagnetic data interpretation method for hidden dangers of dams based on the anchoring effect includes:

[0068] S1: Excite the electromagnetic field in the hidden danger area inside the dam body, collect the original transient electromagnetic field signals, and obtain the decay curve of the original transient electromagnetic field changing with time; in this embodiment, the transient electromagnetic field source excitation method of a magnetic source or an electric source is adopted to excite the electromagnetic field in the hidden danger area inside the dam body; a loop device is used to receive the original transient electromagnetic field signals.

[0069] Furthermore, when using a magnetic source as the field source excitation method, a non-grounded multi-turn small loop coil is selected; when using an electric source as the field source excitation method, grounding electrodes are arranged on the dam body. Further, the current emission waveform is preferably a step wave.

[0070] In this embodiment, the decay curve of the secondary signal of the two original transient electromagnetic fields with time can be the change curve of the voltage of the receiving coil with time, that is, the change rate of the magnetic field changing with time; it can also be the change curve of the magnetic field B with time. Further, the secondary magnetic field data can be selected as three-component magnetic field signals or single-component magnetic fields, but the z-direction component needs to be included. Further, when receiving data, the position coordinates of the receiving point are synchronously recorded, and the receiving method can be manual numbering or RTK positioning.

[0071] S2: Superimpose the attenuation curves at the same location, suppress the uncorrelated noise between the peaks of the original transient electromagnetic field signals, and perform correlated noise suppression processing on the superimposed attenuation curves to obtain the transient electromagnetic attenuation curves after noise suppression. Comprehensive and meticulous preprocessing is carried out on the original data obtained from the transient electromagnetic detection of dam hidden dangers, and its core goal is to completely eliminate various interference information contained therein. In the actual detection environment, the sources of interference information are extensive and complex, mainly including external electromagnetic interference, such as electromagnetic radiation generated by high-voltage transmission lines, street lights, cables, communication base stations, etc. near the dam.

[0072] A further implementation manner lies in that the method for superimposing the attenuation curves at the same location includes:

[0073] The secondary field has the alternating-sign periodicity of the primary field. Superimpose the attenuation curves at the same location, average the superimposed attenuation curves, and any uncorrelated noise between the peaks of the transient electromagnetic secondary field signals in the spectrum will be suppressed, completing the superimposition processing of the attenuation curves; among them, the superimposition formula is as follows:

[0074]

[0075] In the formula, represents the summation of the original transient electromagnetic field signals, n represents the number of signals, is to average the superimposed attenuation curves to obtain a more stable and accurate result, and the signal-to-noise ratio is increased times. However, at the same time, both the transient electromagnetic field signals and any coherent noise in the superimposed attenuation curves will be enhanced, and further suppression of the coherent noise is required.

[0076] A further implementation manner lies in that the method for suppressing correlated noise in the superimposed attenuation curves includes:

[0077] Adopt the first-order difference method to pre-emphasize the superimposed attenuation curves to obtain pre-emphasized signals; in this embodiment, the expression of the first-order difference method is: y(n) = x(n) - αx(n - 1), where x(n) is the original input signal, y(n) is the pre-emphasized signal, and α is the pre-emphasis coefficient, and the value range is (0, 1), and the preferred value is 0.9.

[0078] Segment and window the pre-emphasized signals; specifically, when segmenting the signals, the length overlap of each segmented signal should be no less than 50%; the window function is preferably the Hamming window or the Blackman window. The characteristics of the two window functions are that the values at both ends of the window are smaller and the values in the middle are larger, so that the amplitude of the signal gradually attenuates at both ends after windowing, avoiding sudden changes when the signal is truncated. By windowing, the true frequency of the signal can be more accurately reflected, and the interference is greatly reduced.

[0079] Perform short-time Fourier transform on the windowed pre-emphasized signal to obtain the transient electromagnetic field signal in the frequency domain;

[0080] Use a filter to filter out the noise signal of the transient electromagnetic field signal in the frequency domain to obtain a denoised signal;

[0081] For noise signal filtering, the noise frequency band can be selected by the empirical method for band-pass filtering, or wavelet analysis and other methods can be further used for filtering.

[0082] Merge and convert the denoised signal into the time domain to complete the suppression processing of relevant noise.

[0083] S3: Perform signal evaluation and waste channel rejection on the transient electromagnetic decay curve after noise suppression to obtain the transient electromagnetic decay curve with waste channels removed; specifically, compare the signal after noise suppression with adjacent survey channels. For the signal whose magnitude changes by an order of magnitude compared with adjacent survey channels after denoising the entire curve, it is regarded as a waste channel and is removed.

[0084] S4: Perform time series synthesis processing on the transient electromagnetic decay curve with waste channels removed to screen out the anchor curve.

[0085] A further implementation method is that the method for performing time series synthesis processing on the transient electromagnetic decay curve with waste channels removed to obtain the anchor curve includes:

[0086] When distinguishing the preset weak areas between the dam layers, use the time series synthesis method to take the average value of the decay curves of the original transient electromagnetic fields at each measurement point of the dam at the same moment as the anchor curve value at the same moment; arrange the anchor curve values at each moment in time series to obtain the anchor curve;

[0087] The method for obtaining the anchor curve also includes the direct selection method and the first value selection method;

[0088] When the dam structure form meets the preset requirements, use the direct selection method to directly select the measurement points without abnormalities, and use the decay curve of the measurement points without abnormalities as the anchor curve;

[0089] When conducting comprehensive comparison of the entire dam area, use the first value selection method to select the first original transient electromagnetic field signals collected at each measurement point, and use the decay curve corresponding to the median of all the selected first original transient electromagnetic field signals as the anchor curve. Specifically,

[0090] (1) Direct selection method: For dams with a relatively simple structure type and relatively complete geological data, such as homogeneous dams, the measurement points without abnormalities can be selected and determined according to experience, and the decay curve of this measurement point is used as the anchor curve. That is, f = B i , where i is the measurement point number.

[0091] (2) Time series synthesis method: Since the detection signal is a curve that decays with time, there is a certain correlation between the abnormal occurrence depth and the time series. The decay curves after noise suppression at each measurement point are processed according to time, and the average value at the same moment of each measurement point is taken as the anchor curve value at that moment. The anchor curve values at each moment are arranged in time series to form a curve that changes with time as the anchor curve.

[0092] Specifically, the value at time t is B t = ave(B it ), where i is the number of each measurement point, t is the time, and ave represents the average value. The anchor curve f = {B t : t ∈ T}, where T is the acquisition sequence of the decay curve.

[0093] (3) First value selection method: Select the first acquisition B of each measurement point i1 , and select the curve corresponding to the median as the anchor curve, f = B med .

[0094] Furthermore, when the dam structure type is relatively simple and the geological data is relatively complete, it is preferred to directly select the method to establish the anchor curve; when differentiating the relatively weak areas between the dam layers, the time series synthesis method is preferred; when conducting a comprehensive comparison of the entire dam area, the first value selection method is preferred. It is also possible to comprehensively identify using the time series synthesis method and the first value selection method.

[0095] S5: Based on the anchor curve, calculate the deviation between the decay curve of the original transient electromagnetic field changing with time and the anchor curve to form a deviation curve;

[0096] A further implementation method is that the calculation formula of the deviation curve is:

[0097]

[0098] where B z represents the decay curve value, and B m represents the anchor curve value.

[0099] Perform preliminary judgment and numerical statistical analysis on the obtained deviation data, and draw a deviation multi-trace map. According to the detection principle, in the positive value area of the deviation curve, the underground resistance value of this part is lower than the resistance value corresponding to the anchor curve, and in the negative value area of the deviation curve, the underground resistance value of this part is higher than the resistance value corresponding to the anchor curve. Based on this, the areas with diseases are preliminarily delineated. Regarding the positive and negative value areas: The original decay curves are all positive values, but the above formula calculates the difference with the anchor curve, so the deviation curve has positive and negative values. The positive value area indicates that the magnetic field is stronger than the anchor curve, and the negative value indicates that it is weaker, which explains the difference in underground resistance according to this.

[0100] S6: Based on the existing geological data and monitoring data of the dam, construct a layered model of resistance distribution, and calculate a reference curve based on the layered model of resistance distribution;

[0101] Specifically, calculate the corresponding attenuation curve B according to the layered model of resistance distribution n . Multiply the deviation curve B ω and B n correspondingly according to the time series to obtain the inversion curve B c . In this embodiment, the construction of the resistivity distribution layered model includes: dividing the dam materials above the phreatic line, the dam materials below the phreatic line, and the dam foundation in the vertical direction of the dam; dividing them into upstream dam materials, downstream dam materials, transition materials and core walls in the horizontal direction; using the electrical resistivity characteristic values of the above-mentioned regional dam materials as the resistivity of the region in the model.

[0102] Furthermore, the layered model of resistance distribution is set according to the properties of features such as the phreatic line and bedrock.

[0103] S7: Multiply the deviation curve and the reference curve correspondingly according to the time series to obtain an inversion curve;

[0104] S8: Based on the inversion curve, calculate the apparent resistivity, and obtain a cloud map according to the apparent resistivity;

[0105] The calculation method of the apparent resistivity uses the least squares method:

[0106] The least squares inversion algorithm is a mathematical optimization method that finds the best function match for the data by minimizing the sum of the squares of the errors. It is expressed using the L2 norm:

[0107]

[0108] In the formula, d = [d 1 , d 2 ,..., d n is the actual observed data; f(m) is the theoretical observed value. The purpose of inversion is to find the best model m to minimize the deviation between f(m) and d. To make the data converge, it is often solved by specifying the number of iterations and the minimum error value. In this model experiment, the minimum value is controlled not to be greater than 1% through six iterations.

[0109] S9: Based on the cloud map, infer the hidden danger location of the dam, and complete the interpretation of the transient electromagnetic detection data of the dam hidden danger based on the anchoring effect.

[0110] In summary, the core of the present invention lies in firmly grasping the engineering characteristics of the dam as an artificial structure, where the physical properties of its filling materials are basically known and the seepage pattern is observable. The idea of the anchoring effect is introduced, and interference signals are eliminated through methods such as mathematical statistics and prior information fusion. An anchoring curve for transient electromagnetic observation is established, and then the abnormal responses of other curves are solved based on the anchoring curve. By integrating the spatio-temporal distribution law of the abnormal responses, the abnormal occurrence area inside the dam is located. The present invention solves the interference of the dam boundary conditions on data imaging, can greatly improve the accuracy and efficiency of detecting dam hidden dangers, and provides technical support for the safety evaluation and maintenance of dams.

[0111] Embodiment 2

[0112] The present invention also provides an electromagnetic data interpretation system for dam hidden dangers based on the anchoring effect, which is used to implement the method of Embodiment 1. The system includes: a signal acquisition module, a denoising module, a waste channel removal module, an anchoring curve acquisition module, a deviation curve acquisition module, a reference curve acquisition module, an inversion curve acquisition module, a cloud map acquisition module, a hidden danger location acquisition module, and a display. Among them, the waste channel removal module, the anchoring curve acquisition module, the deviation curve acquisition module, the reference curve acquisition module, the inversion curve acquisition module, the cloud map acquisition module, the hidden danger location acquisition module, and the display integrate a data analysis component 2. The data analysis component 2 is a high-performance computer installed with a data processing program, and this component can connect the points in time sequence to form a multi-channel map. As Figure 2 shown.

[0113] The signal acquisition module is used to excite the electromagnetic field in the hidden danger area inside the dam body, collect the original transient electromagnetic field signal, and obtain the attenuation curve of the original transient electromagnetic field changing with time; in this embodiment, the signal acquisition module includes a primary magnetic field excitation unit 1-1, a waveform regulation unit 1-2, a magnetic field receiving unit 1-3, and a collection and power supply unit 1-4. The primary magnetic field excitation unit 1-1 is divided into an electrode type and a coil type: the electrode type is a long wire, which is connected to the soil through electrodes and can access current; the coil type is a multi-turn coil wound with wires, and can generate a magnetic field after passing different waveform currents. Specifically, the field source excitation coil wound with a multi-turn loop has a diameter of 0.7 m and 33 turns. The magnetic field receiving unit 1-3 is divided into a magnetic flux sensor. When the magnetic field passes through the sensor, it causes an electrical signal change, and the attenuation curve of the magnetic field change rate is obtained through the voltage change, or the magnetic field attenuation curve is obtained by integrating the voltage. The magnetic field receiving unit 1-3 is a coil wound with a multi-turn loop, and has the same center point as the magnetic field excitation module.

[0114] The waveform control unit 1-2 is connected to the primary magnetic field excitation unit 1-1, the magnetic field receiving unit 1-3, and the acquisition and power supply unit 1-4; it can control and emit waveforms of different currents. Specifically, the waveform control unit 1-2 is a current fast turn-off system, which can modulate and emit triangular waves and trapezoidal waves. The acquisition and power supply module is a laptop computer and a 24V lithium battery.

[0115] The denoising module is used to perform superposition processing on the attenuation curves at the same position, suppress the uncorrelated noise between the peaks of the original transient electromagnetic field signals, and perform correlated noise suppression processing on the superposed attenuation curves to obtain the transient electromagnetic attenuation curve after noise suppression.

[0116] A further implementation manner is that the denoising module includes a superposition processing unit for performing superposition processing on the attenuation curves at the same position; the superposition processing unit includes:

[0117] A superposition sub-unit for superposing the attenuation curves at the same position;

[0118] An averaging sub-unit for averaging the superposed attenuation curves to complete the superposition processing of the attenuation curves; where the superposition formula is as follows:

[0119]

[0120] In the formula, represents the summation of the original transient electromagnetic field signals, and n represents the number of signals.

[0121] A further implementation manner is that the denoising module further includes a correlated noise filtering unit for performing correlated noise suppression processing on the superposed attenuation curves; the correlated noise filtering unit includes:

[0122] A pre-emphasis sub-unit for pre-emphasizing the superposed attenuation curves using the first-order difference method to obtain a pre-emphasized signal;

[0123] A segmentation and windowing sub-unit for segmenting and windowing the pre-emphasized signal;

[0124] A time-frequency conversion sub-unit for performing short-time Fourier transform on the windowed pre-emphasized signal to obtain a frequency-domain transient electromagnetic field signal;

[0125] A denoising sub-unit for filtering out the noise signals of the frequency-domain transient electromagnetic field signal using a filter to obtain a denoised signal;

[0126] A frequency-time conversion sub-unit for merging and converting the denoised signal back to the time domain to complete the correlated noise suppression processing.

[0127] A waste channel removal module is used to perform signal evaluation and waste channel removal on the attenuation curve after noise suppression to obtain a transient electromagnetic attenuation curve after removing the waste channel;

[0128] An anchor curve acquisition module is used to perform time-series synthesis processing on the transient electromagnetic attenuation curves after eliminating the waste channel, and screen out the anchor curve;

[0129] A deviation curve acquisition module, used to calculate the deviation between the attenuation curve of the original transient electromagnetic field changing with time and the anchor curve based on the anchor curve, to form a deviation curve;

[0130] A reference curve acquisition module is used to construct a resistance distribution layer model based on existing geological data and monitoring data of the dam, and calculate a reference curve based on the resistance distribution layer model;

[0131] An inversion curve acquisition module is used to multiply the deviation curve and the reference curve correspondingly in time series to obtain an inversion curve;

[0132] A cloud map acquisition module is used to calculate the apparent resistivity based on the inversion curve and obtain a cloud map according to the apparent resistivity;

[0133] The hidden danger location acquisition module is used to infer the hidden danger location of the dam based on the cloud map and complete the interpretation of the transient electromagnetic detection data of the hidden danger of the dam based on the anchoring effect.

[0134] Embodiment 3

[0135] A flood control embankment was built by artificial filling thirty years ago. It adopts a slope embankment structure with a crest elevation of about 7.5m and a crest width of about 5.3m. During the flood season, some sections of the embankment have obvious water leakage. In order to identify the internal leakage channel, the patented system and method are used to carry out work.

[0136] 1. Collect signals.

[0137] The magnetic source transient electromagnetic field source excitation method is used to excite the electromagnetic field in the hidden danger area inside the dam body, and a loop device is used to receive the secondary magnetic field B z Data (raw transient electromagnetic field signal), the acquisition method is continuous sampling, RTK positioning is used during movement. The transmission frequency is 16Hz, the frequency is 1.25MHz, and the transmission area is 10.77m 2 , receiving area 4.02m 2 .

[0138] 2. Eliminate interference.

[0139] In the actual detection environment, interference information includes external electromagnetic interference, high-voltage transmission lines, street lights, cables near the dam, etc. The superposition processing method is used to superimpose the measurement signal multiple times and then average it. The number of superpositions is 100 times, and the signal-to-noise ratio is improved by 10 times.

[0140] For noise suppression processing, first, pre-emphasize the signal to increase the frequency of the high-frequency part; secondly, segment the signal into 120 channels of data, and through windowing, improve the spectral accuracy, with the number of windows being 16; then use the short-time Fourier transform to convert the time-domain transient electromagnetic data into the frequency domain; again, use a band-pass filter to filter out the noise signal; finally, merge the segmented signals and convert them back to the time domain.

[0141] Compare the signal after noise suppression with adjacent survey channels. It is found that the attenuation curve at pile number 4252 shows an order-of-magnitude change compared with adjacent survey channels after noise reduction. This signal is regarded as a defective channel and is excluded. As Figure 3 shown.

[0142] 3. Select the anchor curve

[0143] Two methods are proposed here for selecting the anchor curve: the direct selection method, the time-series synthesis method, and the first-value selection method. This section of the dam is a homogeneous dam, and there are water level observation holes and geological data. The direct selection method is used to determine the anchor curve. The geological data reveals that: within the range of 0m to 10m on the surface of the dam, it is mainly silty clay, grayish-yellow, with a small amount of clay or sand; from 10m to 40m, it is mainly silty sand, with a small amount of thin-layered cohesive soil, and locally fine sand or sandy silt. After preliminary data analysis, there is no abnormality near Hole 1#, and the attenuation curve measured at pile number 4175 (point number 58) is used as the anchor curve. f = B 58 . As Figure 4 shown.

[0144] 4. Obtain the deviation curve

[0145] Taking the processed anchor curve as the reference, calculate the deviation curve between the original signal curve and the anchor curve.

[0146]

[0147] Conduct preliminary judgment and numerical statistical analysis on the obtained deviation data, and draw a multi-channel deviation diagram. According to the detection principle, in the positive region of the deviation curve, the underground resistance value of this part is lower than the resistance value corresponding to the anchor curve, and in the negative region of the deviation curve, the underground resistance value of this part is higher than the resistance value corresponding to the anchor curve. Based on this, it is preliminarily determined that the apparent resistivity deviation near 4462 and 4904 is relatively large and exceeds 20%, and there may be leakage diseases. As Figure 5 shown.

[0148] 5. Construct the inversion curve

[0149] According to the existing geological data and monitoring data of the dam, construct a layered model of resistance distribution. Calculate the corresponding attenuation curve B according to the layered model of resistance distribution n . The deviation curve Bω With B n Multiply them corresponding to each other in time series to obtain the inversion curve B c . The resistance distribution layered model is set according to the properties of features such as the phreatic line and bedrock.

[0150] 6. Generate a map

[0151] Based on the inversion curve, use the particle swarm inversion algorithm to calculate the apparent resistivity, and draw a cloud map according to the apparent resistivity. Further, it is obtained that there are leakage channels in the areas of pile numbers 4800 - 4925, depth 20m - 30m, pile numbers 4975 - 5000, depth 20m - 32m; while there is no concentrated low-resistance area near 4462, and the possibility of a concentrated leakage channel is small. As Figure 6 shown.

[0152] Example 4

[0153] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, it implements the electromagnetic data interpretation method for dam hidden dangers based on the anchoring effect.

[0154] Example 5

[0155] The present invention also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed, it implements the electromagnetic data interpretation method for dam hidden dangers based on the anchoring effect.

[0156] The above-described embodiments are only descriptions of the preferred embodiments of the present invention, and do not limit the scope of the present invention. Without departing from the design spirit of the present invention, various deformations and improvements made by those of ordinary skill in the art to the technical solutions of the present invention shall fall within the protection scope determined by the claims of the present invention.

Claims

1. A method for interpreting electromagnetic data of dam hidden dangers based on anchoring effect, characterized in that: The method comprises: Stimulate the electromagnetic field in the hidden danger area inside the dam body, collect the original transient electromagnetic field signal, and obtain the attenuation curve of the original transient electromagnetic field changing with time; Performing superposition processing on the attenuation curves at the same position to suppress irrelevant noise between the peaks of the original transient electromagnetic field signal, and performing correlated noise suppression processing on the attenuation curves after superposition processing to obtain a transient electromagnetic attenuation curve after noise suppression; Perform signal evaluation and waste channel elimination on the transient electromagnetic attenuation curve after noise suppression to obtain the transient electromagnetic attenuation curve after waste channel elimination; Perform time-series synthesis processing on the transient electromagnetic attenuation curves after eliminating the waste channel, and select the anchor curve; Based on the anchor curve, calculating the deviation between the attenuation curve of the original transient electromagnetic field changing with time and the anchor curve to form a deviation curve; Based on the existing geological data and monitoring data of the dam, a resistance distribution layer model is constructed, and a reference curve is calculated based on the resistance distribution layer model; Multiplying the deviation curve and the reference curve correspondingly in time series to obtain an inversion curve; Based on the inversion curve, calculating the apparent resistivity, and obtaining a cloud map according to the apparent resistivity; Based on the cloud map, the location of the dam hidden dangers is inferred, and the interpretation of the transient electromagnetic detection data of the dam hidden dangers based on the anchoring effect is completed.

2. The method according to claim 1, characterized in that The method for superimposing the attenuation curves at the same position includes: The attenuation curves at the same position are superimposed, and the superimposed attenuation curves are averaged to complete the superposition processing of the attenuation curves; wherein the superposition formula is as follows: In the formula, It represents the summation of the original transient electromagnetic field signals, and n represents the number of signals.

3. The method according to claim 1, characterized in that: The method of performing correlation noise suppression processing on the attenuation curve after superposition processing includes: The first-order difference method is used to pre-emphasize the attenuation curve after superposition processing to obtain a pre-emphasized signal; Segmenting and windowing the pre-emphasized signal; Performing short-time Fourier transform on the windowed pre-emphasized signal to obtain a frequency-domain transient electromagnetic field signal; Using a filter to filter out the noise signal of the frequency-domain transient electromagnetic field signal to obtain a denoised signal; The denoised signals are combined and converted into the time domain to complete the suppression of correlated noise.

4. The method according to claim 1, characterized in that: The method of performing time-series synthesis processing on the transient electromagnetic attenuation curve after eliminating the waste channel to obtain the anchor curve includes: When distinguishing the preset weak area between the layers of the dam, the time series synthesis method is used to take the average value of the attenuation curve of the original transient electromagnetic field of each measuring point of the dam changing with time at the same time as the anchor curve value at the same time; the anchor curve values ​​at each time are arranged in time series to obtain the anchor curve; The method for obtaining the anchor curve also includes a direct selection method and a first value point selection method; When the dam structure meets the preset requirements, the direct selection method is used to directly select the measuring points without abnormalities, and the attenuation curves of the measuring points without abnormalities are used as anchor curves; When conducting a comprehensive comparison of the entire dam area, the first value point selection method is used to select the first original transient electromagnetic field signal collected at each measuring point, and the attenuation curve corresponding to the median of all the selected first original transient electromagnetic field signals is used as the anchor curve.

5. The method according to claim 1, characterized in that The calculation formula of the deviation curve is: In the formula, B z Represents the attenuation curve value, B m Represents the anchor curve value.

6. A dam hidden danger electromagnetic data interpretation system based on anchoring effect, used to implement the method described in any one of claims 1 to 5, characterized in that: The system comprises: The signal acquisition module is used to excite the electromagnetic field in the hidden danger area inside the dam body, collect the original transient electromagnetic field signal, and obtain the attenuation curve of the original transient electromagnetic field changing with time; A denoising module is used to perform superposition processing on the attenuation curves at the same position to suppress irrelevant noise between the peaks of the original transient electromagnetic field signal, and to perform correlated noise suppression processing on the attenuation curves after superposition processing to obtain a transient electromagnetic attenuation curve after noise suppression; A waste channel removal module is used to perform signal evaluation and waste channel removal on the attenuation curve after noise suppression to obtain a transient electromagnetic attenuation curve after removing the waste channel; An anchor curve acquisition module is used to perform time-series synthesis processing on the transient electromagnetic attenuation curves after eliminating the waste channel, and screen out the anchor curve; A deviation curve acquisition module, used to calculate the deviation between the attenuation curve of the original transient electromagnetic field changing with time and the anchor curve based on the anchor curve, to form a deviation curve; A reference curve acquisition module is used to construct a resistance distribution layer model based on existing geological data and monitoring data of the dam, and calculate a reference curve based on the resistance distribution layer model; An inversion curve acquisition module, used for multiplying the deviation curve and the reference curve correspondingly in time series to obtain an inversion curve; A cloud map acquisition module, used to calculate the apparent resistivity based on the inversion curve, and obtain a cloud map according to the apparent resistivity; The hidden danger location acquisition module is used to infer the hidden danger location of the dam based on the cloud map and complete the interpretation of the transient electromagnetic detection data of the hidden danger of the dam based on the anchoring effect.

7. The system according to claim 6, characterized in that The denoising module includes a superposition processing unit, which is used to perform superposition processing on the attenuation curves at the same position; The superposition processing unit comprises: A superposition subunit is used to superimpose the attenuation curves at the same position; The averaging subunit is used to average the superimposed attenuation curves to complete the superposition processing of the attenuation curves; wherein the superposition formula is as follows: In the formula, It represents the summation of the original transient electromagnetic field signals, and n represents the number of signals.

8. The system according to claim 6, characterized in that The denoising module further comprises a correlation noise filtering unit, which is used to perform correlation noise suppression processing on the attenuation curve after the superposition processing; The correlation noise filtering unit comprises: A pre-emphasis subunit, used for pre-emphasizing the attenuation curve after superposition processing by using a first-order difference method to obtain a pre-emphasis signal; A segmentation and windowing subunit, used for segmenting and windowing the pre-emphasis signal; The time-frequency conversion subunit is used to perform short-time Fourier transform on the windowed pre-emphasized signal to obtain a frequency-domain transient electromagnetic field signal; A denoising subunit, used for filtering out the noise signal of the frequency domain transient electromagnetic field signal by using a filter to obtain a denoised signal; The frequency-time conversion subunit is used to combine and convert the denoised signals into the time domain to complete the suppression of correlated noise.

9. An electronic device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method according to any one of claims 1 to 5 is implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed, the method according to any one of claims 1 to 5 is implemented.

Images