Pull-type transient electromagnetic detection system and method for improving electromagnetic signal quality

Abstract

The pull-type transient electromagnetic detection system comprises a detection module, a detection assembly comprises a transmitting coil, a receiving coil and a host, the transmitting coil and the receiving coil are arranged in parallel, the height between the transmitting coil and the receiving coil is adjustable, one end of the host is connected with the transmitting coil, and the other end of the host is connected with the receiving coil. The horizontal relative position of the receiving coil and the transmitting coil can also be adjusted, the position of the receiving coil is determined when the mutual inductance of the transmitting coil and the receiving coil is zero, the signal processing module comprises an amplifier and a noise processing module, and the receiving coil, the amplifier, the noise processing module and the host are sequentially connected. Signal processing is carried out through the following steps: 1, coil laying; 2, setting an amplifier circuit; 3, electromagnetic noise suppression; according to the invention, in the electromagnetic signal acquisition process and the electromagnetic signal processing process, the quality of the electromagnetic signal is improved, so that the electromagnetic signal with higher quality can be comprehensively obtained.

Description

Technical Field

[0001] The present invention relates to the field of electromagnetic detection, and specifically to a towed transient electromagnetic detection system and a method for improving the quality of electromagnetic signals. Background Art

[0002] As the pace of urbanization in my country accelerates, urban space is expanding rapidly, and development space is seriously insufficient. The demand for efficient and safe utilization of underground space has increased sharply. Towed electromagnetic detection technology is of great significance in the detection of urban underground space. The rapid development of cities has made the utilization of underground space more and more important, and accurate detection of the structure and properties of underground space is crucial for urban planning, construction and safety.

[0003] When electromagnetic detection technology is used to detect underground space, it mainly includes an electromagnetic system for acquiring electromagnetic signals and a module for signal reception and processing. However, in the prior art, first of all, time domain electromagnetic transmission is usually used to complete the exploration of geological structures. However, the receiving signal of the current time domain electromagnetic method is relatively small compared to the transmitting voltage, and the receiving coil is more seriously affected by the coupling of the transmitting coil, that is, the coupling interference between the transmitting coil and the receiving coil causes the time-frequency domain electromagnetic signal to be distorted. The primary field coupling interference will not only change the attenuation characteristics of the received signal, causing errors in the later data interpretation, but also expand the dynamic range of the received signal and damage the receiving sensor's rear-stage circuit. In addition, the corresponding exploration structure is relatively complex and the production cost is relatively high. For the electromagnetic signal processing module, due to the serious interference of human noise in the urban environment, the electromagnetic signal obtained is mixed with power frequency harmonic noise, random noise and residual noise, among which the power frequency harmonic noise accounts for a large proportion and has a high amplitude. The random noise component is complex and the frequency band is wide, which pollutes the transient electromagnetic late signal. The method in the prior art is no longer suitable for the urban environment. Therefore, it is very necessary to propose a towed transient electromagnetic detection system and a method for improving the quality of electromagnetic signals. Summary of the invention

[0004] The purpose of the present invention is to provide a towed transient electromagnetic detection system and a method for improving the quality of electromagnetic signals, so that when detecting underground space through battery detection technology, the quality of electromagnetic signals can be comprehensively improved in the electromagnetic signal acquisition process and electromagnetic signal processing process, so as to solve the existing technical defects and unmet technical requirements.

[0005] To achieve the above object, the present invention provides the following technical solution: a towed transient electromagnetic detection system, comprising:

[0006] A detection module, the detection module comprising a transmitting coil, a receiving coil and a host, the transmitting coil is laid on a movable towing platform, the receiving coil is laid on a high platform that can move relative to the towing platform, the high platform is arranged above the towing platform, the plane where the receiving coil is located is parallel to the plane where the transmitting coil is located, the host is arranged on the towing vehicle platform, one end of the host is connected to the transmitting coil, and the other end is connected to the receiving coil, the vertical distance between the high platform and the towing vehicle platform is adjustable, the horizontal relative position of the receiving coil and the transmitting coil is also adjustable, and the position of the receiving coil is determined when the mutual inductance of the transmitting coil and the receiving coil is zero;

[0007] A signal processing module is connected between the host and the receiving coil. The signal processing module includes an amplifier and a noise processing module. One end of the amplifier is connected to the receiving coil, and the other end is connected to the noise processing module. The end of the noise processing module away from the amplifier is connected to the host to obtain a higher quality electromagnetic signal.

[0008] In the present application, the host is used to pass a pulse signal to the transmitting coil and simultaneously receive the primary field voltage information fed back to it by the receiving coil.

[0009] Preferably, the radius R of the transmitting coil is 0.5m, the radius r of the receiving coil is 0.125m, the direction of the current emitted by the host to the transmitting coil is counterclockwise, the relative vertical height H between the receiving coil and the transmitting coil is 0.2m, and the horizontal distance L between the center of the receiving coil and the center of the transmitting coil is 0.6108m.

[0010] Preferably, the receiving coil is configured as an equally spaced cross-coupled honeycomb coil, comprising:

[0011] A skeleton body, wherein the skeleton is in an annular shape as a whole;

[0012] First winding parts, which are spaced apart and circumferentially distributed on the outer ring wall of the skeleton body;

[0013] A second winding portion, the second winding portion is also spaced and circumferentially distributed on the outer ring wall of the skeleton body, and the second winding portion is arranged below the first winding portion;

[0014] The enameled wire has multiple turns which are cross-wound on the first winding part and the second winding part at equal intervals in sequence.

[0015] Preferably, the spacing distance between each adjacent first winding portion is the same, the spacing distance between each adjacent second winding portion is also the same, and the spacing distance between each adjacent first winding portion is the same as the distance between each adjacent second winding portion.

[0016] In the present application, it is necessary to further explain or illustrate that the above-mentioned content can be understood as the first winding part and the second winding part are aligned one by one in the vertical direction.

[0017] Preferably, there is a first spacing space between each adjacent first winding portion, there is a second spacing space between each adjacent second winding portion, and there is a third spacing space between each pair of first winding portions and second winding portions corresponding to each other in the vertical direction;

[0018] The enameled wire rests on the upper end surface of at least one first winding part, and then passes through the first spacing space, and then obliquely downwards passes through the third spacing space and the second spacing space in sequence, and then rests on the lower end surface of at least one second winding part, and then passes through the second spacing space, and then obliquely upwards passes through the third spacing space and the first spacing space in sequence, and then rests on the upper end surface of at least one first winding part again, so as to be cyclically wound on the skeleton body.

[0019] Each turn of the enameled wire is wound around the skeleton body in the same manner in sequence and at equal intervals.

[0020] In the present application, what needs to be further explained or illustrated is that the inclination in the above statement is, in the horizontal direction, inclined in the direction away from the last winding portion that the enameled wire abuts before the inclination, and as for the spacing space through which it is inclined, in the horizontal direction is naturally not the third spacing space corresponding to the last winding portion that the enameled wire abuts before the inclination, but the third spacing space corresponding to one or more winding portions away from this winding portion. That is, it can be understood that the enameled wire extends in the horizontal direction all the time in the direction away from the initial winding end.

[0021] A method for improving the quality of electromagnetic signals, characterized by comprising the following steps:

[0022] 1. Coil laying;

[0023] 1.1) Preliminary laying of coil structure;

[0024] 1.2) According to the theoretical calculation and analysis of relevant parameters in the existing coil structure, select a more appropriate vertical distance H between the receiving coil and the transmitting coil;

[0025] 1.3) According to the vertical distance L determined in step 1.2), further theoretically calculate the horizontal distance L between the center of the receiving coil and the center of the transmitting coil corresponding to this height;

[0026] 1.4), according to the results in step 1.2) and step 1.3), adjust the position of the receiving coil in the coil structure;

[0027] 1.5) Performing transient electromagnetic detection using the coil structure adjusted in step 1.4);

[0028] 2. Amplifier Circuit Setup

[0029] The amplifier adopts a hierarchical amplification circuit, which includes a signal receiving and processing circuit, a primary amplification circuit and a secondary amplification circuit. The input end of the signal receiving and processing circuit is connected to the receiving coil, the output end of the signal receiving and processing circuit is connected to the input end of the primary amplification circuit, the output end of the primary amplification circuit is connected to the input end of the secondary amplification circuit, and the output end of the secondary amplification signal circuit is connected to the host;

[0030] 3. Electromagnetic noise suppression (noise processing module)

[0031] 3.1. Use Wavelet-NGSE algorithm to suppress power frequency noise in the noise;

[0032] 3.1.1) Collect a set of electromagnetic signals with environmental noise and pre-process the electromagnetic field signals using a wavelet threshold algorithm;

[0033] 3.1.2) Use Wavelet-NGSE frequency estimator to estimate the fundamental frequency f of the power frequency component in the noise;

[0034] 3.1.3) Adjust the emission current cycle to an even multiple of the power frequency cycle as the transient electromagnetic emission mode;

[0035] 3.1.4) Collect signals, perform difference calculation on adjacent signals, and remove power frequency harmonics in the original data;

[0036] 3.2. Suppressing random noise in noise through minimum noise separation

[0037] 3.2.1) Adjust the data processed in step 1.6) by noise covariance to generate the minimum noise separation component;

[0038] 3.2.2) Forming a rotation matrix based on the use of noise covariance estimation and data covariance;

[0039] 3.2.3) Use the rotation matrix to associate the transformed data with the signal-to-noise ratio;

[0040] On the basis of filtering out power frequency harmonic noise, in order to further suppress random noise, this application explores a data processing method based on minimum noise separation. Minimum noise separation (Minimum Noise Fraction, MNF) is a statistical analysis method based on noise covariance estimation. The decomposition idea is to decompose the original observation data into multiple components with certain rules and relationships. When they are applied to noise reduction processing, signal-noise separation is achieved through reconstruction of a certain threshold value to achieve the purpose of improving the signal-to-noise ratio. The transformation process is to use a rotation matrix to associate the transformed data with the signal-to-noise ratio, and the noise covariance is used to adjust the original data, and the processing becomes the minimum noise separation (Noise Fraction, NF) with the signal-to-noise ratio arranged from high to low. The rotation matrix is ​​formed based on the use of noise covariance estimation and data covariance. Since the low-order minimum noise separation component has a significantly high signal-to-noise ratio, a strong filtering procedure can be applied to it without seriously reducing the signal quality of the data. When transient electromagnetic instruments perform multiple measurements of the same number within a unit equivalent measurement point, the presence of multiple approximately identical information components will be reflected in the magnitude of the error value of the signal covariance matrix, where the covariance of linear algebra represents the correlation of data vectors. Therefore, signal-to-noise separation can be performed through mathematical transformation.

[0041] 3.3. Use the CNN-GRU model to suppress the residual random noise in the noise in step 2.3)

[0042] 3.3.1) Preprocess the transient electromagnetic signal using the power frequency resonance self-cancellation algorithm and the MNF algorithm;

[0043] 3.3.2) Use deep learning model to suppress residual noise.

[0044] The traditional MNF algorithm can only suppress random noise to a certain extent, and when the random noise intensity is high, the effect of the MNF algorithm is poor, resulting in some residual random noise components in the data. Therefore, on the basis of processing data with the traditional MNF algorithm, this application introduces a deep learning algorithm, uses a neural network model to identify effective signal features, and uses a deep learning model to suppress residual random noise on the basis of suppressing power frequency interference and some random noise in turn with the traditional algorithm, and completes strong noise suppression in urban environments.

[0045] In this application, the processing content in the above step 3 takes into account the fluctuation factor of the power frequency base frequency on the basis of the traditional transient electromagnetic power frequency suppression method, so that the method is more in line with the actual situation during the power frequency processing and improves the power frequency suppression effect. At the same time, the difference average of the positive and negative bipolar signals ensures the efficiency of the entire data processing process.

[0046] The NGSE algorithm is an effective time series frequency estimation algorithm. This method can obtain the fundamental frequency of the main periodic components in the time series. However, when the irrelevant components in the sequence interfere severely, the frequency estimation accuracy will be greatly affected. For transient electromagnetic noise data, random noise, as a component unrelated to the power frequency, will seriously affect the estimation accuracy of the power frequency fundamental frequency.

[0047] Preferably, in the step 1,

[0048] The content of the preliminary laying of the coil structure in step 1.1 is: setting the plane where the towing vehicle platform is located parallel to the plane where the platform is located, laying the transmitting coil on the towing vehicle platform, and laying the receiving coil on the platform;

[0049] The specific contents of the step 1.2 are as follows: 1.2.1), taking different vertical distances H between the receiving coil and the transmitting coil, adjusting the horizontal distance L between the center of the circular receiving line and the center of the transmitting coil, and obtaining the relationship between the field strength of the primary field in the vertical direction and the horizontal distance L at different vertical distances H; 1.2.2), taking the vertical distance H at which the primary field changes less drastically with different horizontal distances L; 1.2.3), and taking the approximate horizontal distance L1 at which the primary field changes most drastically with different vertical distances H, and the relevant parameters in the coil structure include the number of turns of the transmitting coil, the number of turns of the receiving coil, the radius R of the transmitting coil, the radius r of the receiving coil, the current intensity I emitted by the main body to the transmitting coil, the current direction and the horizontal distance L between the center of the receiving coil and the center of the transmitting coil;

[0050] The specific contents of step 1.3 are as follows: 1.3.1), placing the circular receiving coil at the vertical distance H in step 2.2) to the approximate horizontal distance L1 in step 2.3); 1.3.2), calculating the mutual inductance coefficient between the transmitting coil and the circular receiving coil to obtain the size of the horizontal distance L at zero coupling;

[0051] In the step 1.4, the requirement for adjusting the coil structure in step 1) according to the calculation result is: in actual operation, the error between the horizontal distance between the center of the receiving coil and the center of the transmitting coil after adjustment and the horizontal distance L between the center of the receiving coil and the center of the transmitting coil obtained by theoretical calculation in step 3) shall not exceed 2 mm.

[0052] Preferably, the specific contents of step 2 include: the signal receiving and processing circuit includes a power supply positive voltage V DD , negative power supply voltage V SS , resistor R1-resistor R6, electromagnetic signal input interface V sig+ And electromagnetic signal input interface V sig- , where the electromagnetic signal input interface Vsig+ And electromagnetic signal input interface V sig- are connected to the receiving coils respectively (which is a common technology in the prior art and will not be described in detail here), the positive voltage V DD A resistor R1 and a resistor R2 are connected, the resistor R1 and the resistor R2 are connected in parallel, and in the branch where the resistor R1 is located, the resistor R1 is connected in series with the electromagnetic signal input interface V sig+ and resistor R3, in the branch where the resistor R2 is located, the resistor R2 is connected in series with the electromagnetic signal input interface V sig- The branch where the resistor R1 is located is connected in parallel with the branch where the resistor R1 is located, and then connected in series with the resistor R6. One end of the resistor R6 is connected to the resistor R3 and the resistor R4 set in parallel, and the other end is connected to the negative voltage V SS The resistor R6 is connected in parallel with the resistor R5, one end of the resistor R5 is connected to the resistor R6, and the other end is grounded;

[0053] The preliminary discharge circuit includes an amplifier A1, an amplifier A2, capacitors C1 to C4, resistors R7, R8, R9, and R 10 , resistor R 13 and resistor R14, the capacitor C1 is connected in series with the resistor R7 and connected to the inverting input terminal of the amplifier A1, and the end of the capacitor C1 away from the resistor R7 is connected to the resistor R1 and the electromagnetic signal input interface V sig+ The capacitor C3 and the resistor R8 are connected in parallel and then connected to the in-phase input terminal of the amplifier A1. The end of the capacitor C3 away from the amplifier A1 is grounded. The output terminal of the amplifier A1 is connected in series with the resistor R 13 and resistor R 15 The capacitor C3 is connected in series with the resistor R9 and then connected to the inverting input terminal of the amplifier A2. The end of the capacitor C3 away from the resistor R9 is connected to the resistor R2 and the electromagnetic signal input interface V sig- The node between the capacitor C4 and the resistor R 10 After being connected in parallel, the capacitor C4 is connected to the in-phase input terminal of the amplifier A2. The end of the capacitor C4 away from the amplifier A2 is grounded. The output terminal of the amplifier A2 is connected in series with the resistor R 14 and resistor R 16 ;

[0054] The secondary amplification circuit includes an amplifier A3, a resistor R 18 -R 20 , capacitor C5 and electromagnetic signal output interface V out , the inverting input terminal of the amplifier A3 is connected to the resistor R 15 Connect the non-inverting input of the amplifier A3 with the resistor R 16 Connection, i.e. amplifier A 1 and amplifier A2 After parallel connection, they are connected to amplifier A 3 The inverting input terminal and the non-inverting input terminal of the resistor R18 are connected to the amplifier A3 and the resistor R 16 node between them, with the other end grounded, the amplifier A3 The output end is connected in series with resistor R 19 , resistor R 19 And electromagnetic signal output interface V out , the electromagnetic signal output interface V out Connected to the host, one end of the capacitor C5 is connected to the resistor R 19 And electromagnetic signal output interface V out The node between them and the other end is grounded.

[0055] Preferably, the preliminary amplification circuit also includes a resistor R 11 and R 12 , the resistor R 11 One end is connected to the node between the resistor R7 and the inverting input terminal of the amplifier A1, and the other end is connected to the output terminal of the amplifier A1 and the resistor R 13 The node between the resistor R 12 One end is connected to the node between the resistor R9 and the inverting input terminal of the amplifier A2, and the other end is connected to the output terminal of the amplifier A2 and the resistor R 14 Nodes between;

[0056] The secondary amplifier circuit also includes a resistor R 17 , the resistor R 17 One end is connected to the resistor R 15 The other end is connected to the output of amplifier A3 and resistor R 19 The nodes between.

[0057] Since the towed electromagnetic system needs to be placed on a mobile platform for mobile detection, it also has extremely high requirements for the acquisition rate of the host computer acquisition system. It is necessary to ensure the stability of data acquisition and the efficiency of acquisition and storage. Towed electromagnetic data will inevitably need to go through the RAM-ROM process in the host computer, which will be limited by the IO bottleneck. It is necessary to overcome this problem as much as possible through multi-channel host computer program structure optimization to ensure the acquisition and storage efficiency of the host computer. To this end, the writing of superimposed fast acquisition program and production and consumption structure program was carried out.

[0058] The original electromagnetic host computer needs to open the storage file and store the acquired data every time it obtains data from the acquisition card interface. It is difficult to break through the storage efficiency limit of ROM during data storage. Even through fast superposition, the traditional storage structure will still lose some drag-type electromagnetic data due to the mismatch between storage rate and acquisition rate. To solve the problem of acquisition and storage mismatch, a new architecture needs to be used to write the host computer program. For this purpose, the "production-consumption" structure is used to write the LabVIEW multi-channel host computer program. The production-consumption structure is divided into two parts: producers and consumers. The producer is responsible for generating data, which corresponds to the acquisition function in the electromagnetic receiving system. The consumer is responsible for data processing, which can be data storage, data superposition and other preprocessing, or both. During work, the producer obtains data according to hardware instructions and stacks it in the "queue" opened in the memory, waiting for consumers to read and process the data in turn. The producer continuously reads and places data in the queue, overcoming the delay problem caused by the architecture of the traditional host computer. The data acquisition interval is only limited by the hardware bottom-level call program. Consumers are not subject to strict functional restrictions, which improves the scalability of the host computer program.

[0059] Preferably, in step three,

[0060] The specific contents of step 3.1.1) are:

[0061] 3.1.1.1) Collect a set of environmental noises, the collection time of which is greater than or equal to 60 seconds;

[0062] 3.1.1.2) Then use the wavelet threshold algorithm to treat the power frequency component in the noise as a valid signal, and pre-process the pure noise using the following formula:

[0063]

[0064] Among them, the wavelet base is 'db8', the threshold function is the soft threshold, n(t) is the pure environmental noise, and n'(t) is the noise after wavelet preprocessing;

[0065] The specific contents of step 3.1.2) are:

[0066] 3.1.2.1) Establish the NGSE frequency estimation equation:

[0067]

[0068] f=ω / 2π

[0069] Where k = 1, 2, ..., k max , k maxis equal to the total number of power frequency harmonics to be estimated, n is the sampling point in the time domain and N corresponds to the total number of sampling points, N is proportional to the sampling frequency and the total duration of the noise, Δt is the time interval between sampling points, which is inversely proportional to the sampling frequency, n'(n) is the preprocessed noise data obtained in step 1.2), ω is the angular frequency, the initial frequency of f is set to 50Hz, A k and B k Similar to the Fourier coefficients, they represent the phase and amplitude estimates of the power frequency harmonics;

[0070]

[0071] In the formula, T represents the total duration of the collected noise in the time domain, and variables Xk and Yk are intermediate transition variables. They are sensitive to the offset between the current estimated angular frequency ω and the actual power frequency fundamental frequency, and are key intermediate variables in the frequency estimation process.

[0072]

[0073] in, Represents the angular frequency offset, that is, the correction between the current estimated frequency value and the next estimated frequency value, which is calculated by the above four estimated intermediate variables. All are excessive variables;

[0074] 3.1.2.2) Substitute the noise data pre-processed in step 1.1.2) into the NGSE frequency estimation equation and iterate continuously, so that the offset of the angular frequency ω is constantly close to 0, and the angular frequency ω is constantly close to the real power frequency fundamental frequency of the noise;

[0075] 3.1.2.3) Obtain the power frequency base frequency f based on the estimated angular frequency ω;

[0076] The specific content of step 3.1.3) is: according to the industrial frequency base frequency f obtained in step 3.1.2.3), the period of the emission current waveform is adjusted to 2 / f, that is, the emission frequency is f / 2, as the setting transient electromagnetic emission mode, wherein the emission current waveform adopts a bipolar square wave with a duty cycle of 25%;

[0077] After obtaining the actual estimated value of the industrial frequency base frequency, the next step is to set the transient electromagnetic emission mode. The emission current waveform adopts a bipolar square wave with a duty cycle of 25%. The industrial frequency base frequency f is obtained using the Wavelet-NGSE algorithm, and the period of the bipolar emission square wave is adjusted to 2 / f (the emission frequency is f / 2).

[0078] It needs to be explained here that in step 3.1.3) of the present application, the meaning of adjusting the period of the emission current to an even multiple of the power frequency period as the transient electromagnetic emission mode is not limited to 2 / f (the emission frequency is f / 2), but the emission frequency is adjusted to f / 2, f / 4... etc. 2 / f (the emission frequency is f / 2) is just one of the preferred methods.

[0079] The specific content of step 3.1.4) is:

[0080] 3.1.4.1) Start the survey device according to the transient electromagnetic emission mode set in step 1.3);

[0081] 3.1.4.2) The transient electromagnetic field signals generated at the moment of current shutoff are collected by the receiving coil. In each transmission cycle, a group of positive signals and a group of negative signals can be obtained;

[0082] 3.1.4.3) Subtract the collected adjacent positive and negative signals and take the average to complete power frequency suppression;

[0083] The specific content of the step 3.2.1) is: adjusting the content processed in step 3.1.4.3) by noise covariance to process it into the minimum noise separation component with the signal-to-noise ratio arranged from high to low;

[0084] The specific content of step 3.2.2) is:

[0085] 3.2.2.1) Generate multiple secondary field data matrices within equivalent measurement points:

[0086]

[0087] Among them, each channel of the noisy data X has been zero-meaned, and the matrix contains n secondary field signals, each signal contains

[0088] m time channels, X S represents the theoretical electromagnetic response of the secondary field, X N represents noise, the signal and noise are uncorrelated with each other;

[0089] 3.2.2.2) Generate the cosquare matrix of X as:

[0090] C=C S +C N

[0091] Among them C S Represents X S The covariance matrix, C N Yes X N The covariance matrix of

[0092] 3.2.2.3) The covariance matrix CN The decomposition into eigenvalues ​​and eigenvectors is as follows:

[0093]

[0094] Where D N The diagonal matrix representing the eigenvalues ​​in descending order, U N represents the eigenvector associated with each eigenvalue;

[0095] In the step 3.2.3), the specific content is:

[0096] The minimum noise separation component is related to the signal-to-noise ratio:

[0097]

[0098] in, represents the noise variance of the i-th component, represents the variance of the noisy signal of the i-th component;

[0099] The noise covariance matrix C N It is composed of the elements γkq:

[0100]

[0101] Z kj =X k(j+1) -X kj

[0102] Z qj =X q(j+1) -X qj

[0103] Among them, represents the average value of the kth time channel, represents the average value of the qth time channel;

[0104] Construct matrix: P = U N D N -0.5 Make P T C N P is the identity matrix;

[0105] Then Y = PX

[0106] Then perform the following calculation:

[0107] C x =P T CP=VD X V T

[0108] Where D Xand P represent the eigenvalue matrix and eigenvector matrix respectively;

[0109] The second rotation matrix R = PV is constructed, and the electromagnetic data matrix X is linearly transformed to obtain the minimum noise separation component NF, that is, the matrix ψ:

[0110]

[0111] Finally, the data is reconstructed by retaining the first L (L < m) minimum noise separation components. The reconstructed signal is expressed as:

[0112]

[0113] Same as the transient electromagnetic data X before processing, The matrix size remains unchanged.

[0114] Preferably, in step 1.4.2), a group of positive signals and a group of negative signals obtained in each transmission cycle include:

[0115]

[0116] x - (n) = s - (n)+n T- (n)+n r- (n)

[0117] Signals + (n) is a positive signal, and signal s-(n) is a negative signal. The two sets of signals have the same amplitude and opposite positive and negative directions. T+ (n) and n T -(n) all represent the power frequency harmonics coupled into the signal;

[0118] The calculation process in step 1.4.3) is specifically as follows:

[0119]

[0120] In the present application, after the transmitting coil is working, the receiving coil can collect the transient electromagnetic two-field signal generated at the moment of current shutoff. In each transmitting cycle, we can obtain a set of positive signals x+(n) and a set of negative signals x-(n). Since the duty cycle of the bipolar square wave is adjusted to 25% and the transmitting cycle is an even multiple of the power frequency cycle, the power frequency harmonics nT+(n) and nT-(n) have the same phase, so the amplitude is also the same. Finally, by performing a difference operation on adjacent signals, the power frequency harmonics in the original data can be removed. With this method, we only need to perform a subtraction operation once to effectively remove the power frequency harmonics without causing distortion of the TEM signal.

[0121] Compared with the prior art, the present invention has the following beneficial effects:

[0122] 1. In the process of signal reception, within a certain dimensional error range, the influence of the primary field exposure sense can be ignored in the present application, and the transceiver coil is considered to be in a zero coupling state. The signal induced by the receiving coil can be used as a pure secondary field signal for subsequent data processing, avoiding the interference of the primary field on the signal, so that the present application has a higher tolerance to physical dimensional errors, making the acquired signal more fidelity, and the present application has the characteristics of simplicity in structure and low manufacturing cost. By setting an equally spaced cross-coupled honeycomb receiving coil, the turn spacing of the receiving coil is increased, the parasitic capacitance can be reduced, the frequency band is widened, the resonant frequency is higher, and the bandwidth is larger, which effectively improves the bandwidth of multi-turn small loop reception and can improve the shallow detection capability.

[0123] 2. In the process of electromagnetic signal processing, the present application increases the sensitivity of the receiving system through an amplifier via an amplifying circuit, thereby improving the data quality of the urban towed transient electromagnetic device. By setting up a LabVIEW multi-channel host computer program written in a "production-consumption" structure, the delay problem caused by the architecture of the traditional host computer is overcome. The data acquisition interval is only limited by the hardware bottom-level calling program, so that consumers have no strict functional restrictions, thereby improving the scalability of the host computer program.

[0124] 3. In the process of processing electromagnetic field signals, the present application also sets up a signal processing module to refine and classify the noise in the collected transient electromagnetic signals, and remove each type of noise feature separately. A noise reduction strategy based on bipolar square wave emission, wavelet-NGSE, superposition averaging, minimum noise separation and deep learning is proposed to form a filtering process for suppressing power frequency harmonic noise, random noise and residual noise. In view of the interference problem of power frequency harmonic noise, the wavelet-NGSE algorithm is used to extract its precise fundamental frequency value, and then combined with the periodicity of the power frequency, the emission frequency and waveform are adjusted to achieve the suppression of highly correlated power frequency noise; random noise is further suppressed by superposition averaging; finally, minimum noise separation + CNN-GRU network calculates data components and reconstructs the signal after minimum noise separation, so that high-precision residual noise is separated, and electromagnetic data with a high signal-to-noise ratio is obtained, laying the foundation for the accurate interpretation of deep signals. BRIEF DESCRIPTION OF THE DRAWINGS

[0125] Figure 1 It is a simplified diagram of the overall structure of the present invention;

[0126] Figure 2 is a schematic diagram of a coordinate system in Embodiment 1 of the present invention;

[0127] Figure 3 The distribution diagram of the intensity of the transmitted magnetic field in the Z-axis direction at different heights as a function of the horizontal distance L in Example 1 of the present invention;

[0128] Figure 4 is the relationship between the mutual inductance coefficient of the transceiver coil and the horizontal distance L when the vertical distance H is 0.02 m in Example 1 of the present invention;

[0129] Figure 5 is a relationship diagram between the mutual inductance coefficient and the horizontal distance L of the transceiver coils when the vertical distance H is 0.1 m in Example 1 of the present invention;

[0130] Figure 6 is a relationship diagram between the mutual inductance coefficient and the horizontal distance L of the transceiver coils when the vertical distance H is 0.2 m in Example 1 of the present invention;

[0131] Figure 7 is a relationship diagram between the mutual inductance coefficient and the horizontal distance L of the transceiver coils when the vertical distance H is 0.3 m in Example 1 of the present invention;

[0132] Figure 8 is a curve diagram of the change of the conductive ring in Example 1 of the present invention;

[0133] Fig. 9 This is a pure secondary field voltage curve diagram without considering the influence of the receiving coil parasitic parameters in Example 1 of the present invention;

[0134] Fig.10 Schematic diagram of the simulation results of the signals induced when the receiving coil deviates toward the center direction of the transmitting coil under different deviation conditions in Example 1 of the present invention;

[0135] Fig.11 It is a schematic diagram of the simulation results of the signals induced when the receiving coil is offset away from the center of the transmitting coil under different deviation conditions in Example 1 of the present invention;

[0136] Fig.12 This is a schematic diagram of the overall structure of the skeleton when the transmitting coil is not wound with enameled wire in Example 1 of the present invention;

[0137] Fig.13 It is a top view of the skeleton when the transmitting coil is not wound with enameled wire in Example 1 of the present invention;

[0138] Fig.14 It is a bottom view of the frame when the transmitting coil is not wound with enameled wire in Example 1 of the present invention;

[0139] Fig.15 This is a plan view of the structure of a honeycomb coil with medium spacing winding according to Embodiment 1 of the present invention;

[0140] Fig.16 is the bandwidth ratio of each receiving coil in Example 1 of the present invention and Comparative Example 1 under the same conditions;

[0141] Fig.17 is a circuit diagram of the amplifier in the present invention;

[0142] Fig.18 This is a comparison table of background noise of each amplifier under the same conditions in Example 1 and Comparative Example 2 of the present invention;

[0143] Fig.19 is a waveform diagram of a simulation signal in an embodiment of the present invention;

[0144] Fig. 20 A waveform diagram of signal-noise coupling data in an embodiment of the present invention;

[0145] Fig.21 is a waveform diagram after power frequency suppression in a bipolar power frequency suppression method assumed in an embodiment of the present invention;

[0146] Fig. 22 It is a waveform diagram after power frequency suppression in the bipolar power frequency suppression method of the actual simulation base frequency in the embodiment of the present invention;

[0147] Fig.23 This is a schematic diagram of the minimum noise separation processing result in an embodiment of the present invention;

[0148] Fig.24 This is a schematic diagram of the CNN-GRU model structure in an embodiment of the present invention;

[0149] Fig.25 It is a schematic diagram of the results of suppressing urban transient electromagnetic random noise using three algorithms in an embodiment of the present invention;

[0150] In the figure: transmitting coil 1, towing vehicle platform 2, receiving coil 3, platform 4, main unit 5, skeleton body 6, first winding part 7, second winding part 8, first partition space 9, second partition space 10, third partition space 11, first enameled wire 12, second enameled wire 13, third enameled wire 14, fourth enameled wire 15, fifth enameled wire 16, sixth enameled wire 17, seventh enameled wire 18. DETAILED DESCRIPTION

[0151] The following will be combined with the attached embodiment of the present invention Figure 1-25 , the technical solutions in the embodiments of the present invention are clearly and completely described. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0152] In the description of the present invention, it should be understood that the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features.

[0153] See also Figure 1-25 , embodiments of the present invention:

[0154] Embodiment 1:

[0155] like Figure 1 Shown: A towed transient electromagnetic detection system, comprising:

[0156] A detection module, the detection module comprising a transmitting coil 1, a receiving coil 3 and a host 5, the transmitting coil 1 is laid on a movable towing platform, the receiving coil 3 is laid on a high platform 4 that can move relative to the towing platform, the high platform 4 is arranged above the towing platform, the plane where the receiving coil 3 is located is parallel to the plane where the transmitting coil 1 is located, the host 5 is arranged on the towing vehicle platform 2, one end of the host 5 is connected to the transmitting coil 1, and the other end is connected to the receiving coil 3, the vertical distance between the high platform 4 and the towing vehicle platform 2 is adjustable, the horizontal relative position of the receiving coil 3 and the transmitting coil 1 is also adjustable, and the position of the receiving coil 3 is determined when the mutual inductance of the transmitting coil 1 and the receiving coil is zero;

[0157] A signal processing module is connected between the host 5 and the receiving coil 3. The signal processing module includes an amplifier and a noise processing module. One end of the amplifier is connected to the receiving coil 3, and the other end is connected to the noise processing module. The end of the noise processing module away from the amplifier is connected to the host 5 to obtain a higher quality electromagnetic signal.

[0158] The radius of the transmitting coil 1 is greater than the radius of the receiving coil 3 .

[0159] Specifically, the radius R of the transmitting coil 1 is 0.5 m, and the radius r of the receiving coil 3 is 0.125 m.

[0160] like Figure 2-11 As shown: The method for improving the electromagnetic signal corresponding to the overall structure of the coil is:

[0161] 1) Preliminary laying of coil structure;

[0162] The specific contents of the preliminary laying of the coil structure are:

[0163] The plane where the towing vehicle platform 2 is located is arranged parallel to the plane where the platform 4 is located, the transmitting coil 1 is laid on the towing vehicle platform 2 , and the receiving coil 3 is laid on the platform 4 .

[0164] 2) According to the theoretical calculation and analysis of relevant parameters in the existing coil structure, a more appropriate vertical distance H between the receiving coil 3 and the transmitting coil 1 is selected;

[0165] The relevant parameters in the coil structure include:

[0166] The number of turns of the transmitting coil 1, the number of turns of the receiving coil 3, the radius R of the transmitting coil 1, the radius r of the receiving coil 3, the current intensity I emitted by the subject to the transmitting coil 1, the current direction and the horizontal distance L between the center of the receiving coil 3 and the center of the transmitting coil 1.

[0167] 2.1) Take different vertical distances H between the receiving coil 3 and the transmitting coil 1, adjust the horizontal distance L between the center of the circular receiving line and the center of the transmitting coil 1, and obtain the relationship between the field strength of the primary field in the vertical direction and the horizontal distance L under different vertical distances H;

[0168] like Figure 3 As shown: Specifically, in this embodiment, a theoretical calculation is performed by taking the model parameters of a single-turn transmitting coil 1, a transmitting radius R=0.5m, a single-turn receiving coil 3, a receiving radius r=0.125m, a transmitting current intensity I=10A, and a current direction being counterclockwise as an example, and the primary field distribution at different plane positions of H of 0.02m, 0.1m, 0.2m, and 0.3m is analyzed and calculated.

[0169] 2.2) Take the vertical distance H where the field changes less dramatically with different horizontal distances L;

[0170] Establish a coordinate system: take the plane where the transmitting coil 1 is located as the zero plane, the center of the transmitting coil 1 as the coordinate origin, the offset direction of the receiving coil 3 relative to the transmitting coil 1 as the X-axis direction, and the direction perpendicular to the zero plane and upward as the Z-axis direction. In the figure, Φ1 is the angle between the transmitting coil 1 microelement and the x-axis, Φ2 is the angle between the receiving coil 3 microelement and the x-axis, Q is an arbitrary microelement point of the transmitting coil 1, and P is an arbitrary microelement point of the receiving coil 3.

[0171] The magnitude of the magnetic field intensity is represented by both the numerical value of the z-axis and the color bar on the right side of the image. Therefore, by calculating the primary field distribution in the z-direction of planes at different heights, it can be seen that as the plane height increases, not only does the maximum value of the magnetic field intensity gradually decay, but the degree of change of the magnetic field also becomes more gradual, so the vertical distance H is taken as 0.2m.

[0172] 2.3) and take the approximate horizontal distance L1 where the field changes most dramatically with different vertical distances H.

[0173] It can be found that the field strength is weak at a position far away from the transmitting line, and the field strength changes most dramatically at the position of positive and negative H alternation near the transmitting line position (x=±0.5m). However, as H increases, the intensity of the field change is significantly reduced. Therefore, when the receiving coil 3 is placed near x=±0.5m, the receiving coil 3 senses the transmitting magnetic field in both positive and negative directions, so the position x=±0.5m is taken as the approximate horizontal distance L1.

[0174] 3) According to the vertical distance L determined in step 2), further theoretically calculate the horizontal distance L between the center of the receiving coil 3 and the center of the transmitting coil 1 corresponding to this height;

[0175] Since the coupling degree of the two coils is related to the mutual inductance coefficient, when the mutual inductance coefficient is zero, the two coils have no coupling relationship with each other. Therefore, the mutual inductance coefficient of the transmitting and receiving line coils is calculated to obtain the size of L at zero coupling.

[0176] 3.1) Place the circular receiving coil at the vertical distance H in step 2.2) to the approximate horizontal distance L1 in step 2.3);

[0177] 3.2) By calculating the mutual inductance coefficient between the transmitting coil 1 and the circular receiving coil, the horizontal distance L at zero coupling is obtained:

[0178] Calculation results Figure 6 As shown, the horizontal distance L is 0.6108m.

[0179] When the mutual inductance is zero, the receiving coil 3 is divided into two parts, the inner and outer parts, by the transmitting coil 1. The outer area of ​​the transmitting coil 1 with weaker magnetic field strength is obviously larger. At this position, the positive and negative magnetic fluxes B*S sensed by the receiving coil 3 and the transmitting coil 1 are offset, and are completely unaffected by the primary field.

[0180] In this embodiment, in order to facilitate data verification and comprehensive analysis, the parameters other than L=0.2m are also calculated:

[0181] According to the prior art (Neumann formula), the relationship between the offset distance L and the mutual inductance coefficient is calculated when the relative vertical distances H between the single-turn transmitting coil and the single-turn receiving coil 3 are 0.02m, 0.1m, 0.2m, and 0.3m, respectively. Figure 4-7As shown. It can be found that, in theory, when the transmitting and receiving coils 3 are at different H, an L can be found to make the mutual inductance coefficient zero. When the mutual inductance coefficient of the transmitting and receiving coils is 0, the offset distances L corresponding to H of 0.02m, 0.1m, 0.2m, and 0.3m are 0.5308m, 0.5544m, 0.6108m, and 0.6916m, respectively. At zero mutual inductance, the receiving coil 3 is divided into two parts, inner and outer, by the transmitting coil 1. The outer area of ​​the transmitting coil 1 with weaker magnetic field strength is significantly larger. At this position, the positive and negative magnetic fluxes B*S sensed by the receiving coil 3 to the transmitting coil 1 are offset, and are completely unaffected by the primary field. As H increases, the change curve of the mutual inductance coefficient at the zero position becomes smoother, and the offset distance L corresponding to zero mutual inductance also gradually increases.

[0182] 4) According to the results of step 2) and step 3), the position of the receiving coil 3 in the coil structure is adjusted;

[0183] In actual operation, the error between the adjusted horizontal distance between the center of the receiving coil 3 and the center of the transmitting coil 1 and the horizontal distance L between the center of the receiving coil 3 and the center of the transmitting coil 1 obtained by theoretical calculation in step 3) should not exceed 2 mm.

[0184] V) Perform transient electromagnetic detection using the coil structure adjusted in step 4).

[0185] In summary, the theory shows that when the zero flux position of the RX coil is determined, the present embodiment can receive a pure secondary field signal that is not interfered by the primary field. However, in practical applications, any error (processing error, size error, position error) will reintroduce primary field interference. The receiving area must include the area where the positive and negative alternations of the field strength change most dramatically, and the existence of errors may have a significant impact on the compensation effect.

[0186] Therefore, in order to verify the effect of the above structure in eliminating the coupling of the primary field, the simulation experiment based on the above method is as follows:

[0187] The specific parameters of the simulation system are as follows: the radius of the transmitting coil 1 is 0.5m, 1 turn, the transmitting current is 1A, and the shutdown time is 20us (linear shutdown); the radius of the receiving coil 3 is 0.125m, 160 turns; the conductive ring has 1 turn, a radius of 1m, a self-inductance of 0.168uF, an internal resistance of 0.0056Ω, and a height difference H0=1m with the transmitting coil 1. The relative height of the transmitting and receiving coils is fixed at H=0.2m. From the theoretical calculation in the previous article, it can be seen that when the transmitting and receiving coils are zero-coupled, the theoretical horizontal offset distance L=0.6108m. Adjust the offset distance of the receiving coil 3 to simulate the system receiving signal under different horizontal position errors.

[0188] The experimental simulation results are as follows Figure 8-11 As shown, during the off period (0~20us), the conductive loop current ( Figure 8 ) is gradually strengthened by the transmitting current, i.e., the process of establishing the eddy current. After being completely shut down, it gradually decreases due to its own energy consumption, i.e., the process of attenuation of the eddy current. Due to the coupling between the conductive ring and the receiving coil 3, without considering the influence of the parasitic parameters of the receiving coil 3, the pure secondary field voltage waveform generated by the conductive ring induced by the receiving coil 3 during the process of establishing and attenuating the eddy current of the conductive ring is as follows: Fig. 9 As shown in the figure, its waveform trend is related to the derivative of the conductive loop current. In the process of eddy current establishment, the current growth is gradually slow, that is, the derivative of the conductive loop current gradually decreases, resulting in a decrease in the secondary field voltage. After the transmitting current is completely turned off, the eddy current changes suddenly from the establishment process to the attenuation process, that is, the polarity of the derivative of the conductive loop current suddenly changes, causing the positive and negative jump of the secondary field voltage, and the voltage gradually decreases as the eddy current decays slowly. Overall, when the offset of the receiving coil 3 of the eccentric self-compensation system does not exceed 2mm, the system offset direction is different, but the fault tolerance curve is similar and always above 99%. In the simulation results, the influence of the primary field is extremely limited and can be ignored. The signal induced by the receiving coil 3 can be used as a pure secondary field signal for subsequent data processing.

[0189] In summary, the experimental results are consistent with the theoretical results, which verifies that the eccentric self-compensating TEM structure can ignore the influence of the primary field leakage inductance within a certain dimensional error range (within 2mm horizontal error), and it is considered that the transceiver coil is in a zero coupling state. The signal induced by the receiving coil 3 can be used as a pure secondary field signal for subsequent data processing. Compared with the prior art, the present application has a lower cost and a higher tolerance to physical dimensional errors.

[0190] It should be explained that the above content about some parameters of the overall structure is only for better verifying the effect of the overall structure on improving electromagnetic signals, and does not mean that only the above parameters are used in this embodiment. The overall structure within the above error range is adaptable to a variety of coil sizes and parameters.

[0191] The content of the distance structure for the transmitting coil 1 is as follows:

[0192] A skeleton body 6, the skeleton is generally annular;

[0193] First winding parts 7, which are spaced apart and circumferentially distributed on the outer ring wall of the skeleton body 6;

[0194] The second winding parts 8 are also spaced apart and circumferentially distributed on the outer wall of the skeleton body 6, and the second winding parts 8 are arranged below the first winding part 7;

[0195] The enameled wire is a plurality of turns of enameled wire, which are cross-wound on the first winding part 7 and the second winding part 8 in sequence at equal intervals.

[0196] like Figure 12-15As shown in the figure, the spacing distance between each first winding part 7 is the same, and the spacing distance between each second winding part 8 is also the same.

[0197] like Figure 12-15 As shown in FIG. 1 , the spacing distance between each first winding portion 7 is the same as the spacing distance between each second winding portion 8 .

[0198] like Figure 12-15 As shown: there is a first spacing space 9 between each of the first winding parts 7, there is a second spacing space 19 between each of the second winding parts 8, and there is a third spacing space 11 between each pair of first winding parts 7 and second winding parts 8 corresponding to each other in the vertical direction;

[0199] The enameled wire rests on the upper end surfaces of the two first winding parts 7, and then passes through the first spacing space 9, and then obliquely downwards passes through the third spacing space 11 and the second spacing space 19 in sequence, and then rests on the lower end surfaces of the two second winding parts 8, and then passes through the second spacing space 19, and then obliquely upwards passes through the third spacing space 11 and the first spacing space 9 in sequence, and then rests on the upper end surfaces of the two first winding parts 7 again, and is thus cyclically wound on the skeleton body 6.

[0200] Each turn of the enameled wire is wound around the skeleton body 6 in the same manner in sequence and at equal intervals.

[0201] like Fig.15As shown: In the specific embodiment, 7 turns of enameled wire are specifically arranged, namely the first enameled wire 12, the second enameled wire 13, the third enameled wire 14, the fourth enameled wire 15, the fifth enameled wire 16, the sixth enameled wire 17 and the seventh enameled wire 18. In the present embodiment, the wires are wound from left to right. The first enameled wire 12 first abuts against the upper end surfaces of the two first winding parts 7, and then passes through the first spacing space 9 between the last first winding part 7 abutted by the first enameled wire 12 and the first winding part 7 adjacent to the right side of the first winding part 7, and tilts to the lower right, passes through three adjacent third spacing spaces 11 in sequence, and then passes through the second spacing space 19 adjacent to the right side of the last third spacing space 11, and abuts against the formation of this On the lower end face of the second winding portion 8 on the right side of the second spacing space 19, after continuously abutting against two second winding portions 8, it passes through the second spacing space 19 on the right side of the last second winding portion 8 abutted, tilts to the upper right, and passes through three adjacent third spacing spaces 11 in sequence, and then passes through the first spacing space 9 adjacent to the right side of the last third spacing space 11, and abuts against the upper end face of the first winding portion 7 on the right side forming this first spacing space 9, thereby circulating winding, and the second enameled wire 13, the third enameled wire 14, the fourth enameled wire 15, the fifth enameled wire 16, the sixth enameled wire 17 and the seventh enameled wire 18 are respectively and sequentially wound in the same manner, and the skeleton body 6 is annular.

[0202] Specifically: To further verify the physical properties, the coil in this embodiment is used in the towed battery detection technology, and the results are as follows: Fig.16 shown.

[0203] The amplifier adopts a hierarchical amplification circuit, which includes a signal receiving and processing circuit, a primary amplification circuit and a secondary amplification circuit. The input end of the signal receiving and processing circuit is connected to the receiving coil 3, the output end of the signal receiving and processing circuit is connected to the input end of the primary amplification circuit, the output end of the primary amplification circuit is connected to the input end of the secondary amplification circuit, and the output end of the secondary amplification signal circuit is connected to the host 5.

[0204] like Fig.17 As shown: the signal receiving and processing circuit includes a positive power supply voltage V DD , negative power supply voltage V SS , resistor R1-resistor R6, electromagnetic signal input interface V sig+ And electromagnetic signal input interface V sig- , the power supply positive voltage V DD A resistor R1 and a resistor R2 are connected, the resistor R1 and the resistor R2 are connected in parallel, and in the branch where the resistor R1 is located, the resistor R1 is connected in series with the electromagnetic signal input interface V sig+and resistor R3, in the branch where the resistor R2 is located, the resistor R2 is connected in series with the electromagnetic signal input interface V sig- The branch where the resistor R1 is located is connected in parallel with the branch where the resistor R1 is located, and then connected in series with the resistor R6. One end of the resistor R6 is connected to the resistor R3 and the resistor R4 set in parallel, and the other end is connected to the negative voltage V SS The resistor R6 is connected in parallel with the resistor R5, one end of the resistor R5 is connected to the resistor R6, and the other end is grounded;

[0205] The preliminary discharge circuit includes an amplifier A1, an amplifier A2, capacitors C1 to C4, resistors R7, R8, R9, and R 10 , resistor R 13 and resistor R14, the capacitor C1 is connected in series with the resistor R7 and connected to the inverting input terminal of the amplifier A1, and the end of the capacitor C1 away from the resistor R7 is connected to the resistor R1 and the electromagnetic signal input interface V sig+ The capacitor C3 and the resistor R8 are connected in parallel and then connected to the in-phase input terminal of the amplifier A1. The end of the capacitor C3 away from the amplifier A1 is grounded. The output terminal of the amplifier A1 is connected in series with the resistor R 13 and resistor R 15 The capacitor C3 is connected in series with the resistor R9 and then connected to the inverting input terminal of the amplifier A2. The end of the capacitor C3 away from the resistor R9 is connected to the resistor R2 and the electromagnetic signal input interface V sig- The node between the capacitor C4 and the resistor R 10 After being connected in parallel, the capacitor C4 is connected to the in-phase input terminal of the amplifier A2. The end of the capacitor C4 away from the amplifier A2 is grounded. The output terminal of the amplifier A2 is connected in series with the resistor R 14 and resistor R 16 ;

[0206] The secondary amplification circuit includes an amplifier A3, a resistor R 18 -R 20 , capacitor C5 and electromagnetic signal output interface V out , the inverting input terminal of the amplifier A3 is connected to the resistor R 15 Connect the non-inverting input of the amplifier A3 with the resistor R 16 Connection, (Amplifier A 1 and amplifier A 2 After parallel connection, they are connected to amplifier A 3 One end of the resistor R18 is connected to the amplifier A3 and the resistor R 16 node between them, with the other end grounded, the amplifier A3 The output end is connected in series with resistor R 19 , resistor R 19And electromagnetic signal output interface V out , the electromagnetic signal output interface V out Connected to the host 5, one end of the capacitor C5 is connected to the resistor R 19 And electromagnetic signal output interface V out The node between them and the other end is grounded.

[0207] The preliminary amplification circuit also includes a resistor R 11 and R 12 , the resistor R 11 One end is connected to the node between the resistor R7 and the inverting input terminal of the amplifier A1, and the other end is connected to the output terminal of the amplifier A1 and the resistor R 13 The node between the resistor R 12 One end is connected to the node between the resistor R9 and the inverting input terminal of the amplifier A2, and the other end is connected to the output terminal of the amplifier A2 and the resistor R 14 Nodes between;

[0208] The secondary amplifier circuit also includes a resistor R 17 , the resistor R 17 One end is connected to the resistor R 15 The other end is connected to the output of amplifier A3 and resistor R 19 The nodes between.

[0209] Specifically, in order to further verify the physical characteristics, the background noise of the amplifier of this embodiment was tested in an electromagnetic shielding room using a dynamic signal analyzer Agilent 35670. The results are as follows: Fig.18 As shown in the figure, the vertical axis, Thecircuit noise, means that the vertical axis means circuit noise, and the unit is volts per square root hertz, and the horizontal axis, Frequency, means that the horizontal axis means frequency, and the unit is hertz (Hz).

[0210] The amplifier transmits the processed signal to the noise processing module.

[0211] The contents of the noise processing module include:

[0212] like Figure 19-25As shown in the figure: The NGSE algorithm is an effective time series frequency estimation algorithm. This method can obtain the fundamental frequency of the main periodic components in the time series. However, when the irrelevant components in the sequence interfere severely, the frequency estimation accuracy will be greatly affected. For transient electromagnetic noise data, random noise, as a component unrelated to the power frequency, will seriously affect the estimation accuracy of the power frequency fundamental frequency. Therefore, we propose a Wavelet-NGSE algorithm to estimate the frequency of transient electromagnetic noise data. First, a set of environmental noise is collected to ensure that the collection time is greater than or equal to 60ms. Then, using the wavelet threshold algorithm, the power frequency component in the noise is regarded as a valid signal, and the pure noise is preprocessed to remove the noise components other than the power frequency. The wavelet basis selects 'db8', and the threshold function selects the soft threshold.

[0213]

[0214] Among them, the wavelet base is 'db8', the threshold function is the soft threshold, n(t) is the pure environmental noise, and n'(t) is the noise after wavelet preprocessing. After the noise preprocessing, the NGSE frequency estimation equation group is established.

[0215]

[0216] f=ω / 2π

[0217] Where k = 1, 2, ..., k max , k max is equal to the total number of power frequency harmonics to be estimated, n is the sampling point in the time domain and N corresponds to the total number of sampling points, N is proportional to the sampling frequency and the total duration of the noise, Δt is the time interval between sampling points, which is inversely proportional to the sampling frequency, n'(n) is the preprocessed noise data obtained in step 1.2), ω is the angular frequency, and the initial frequency of f is set to 50Hz, which is consistent with the ideal power frequency fundamental frequency. k and B k Similar to the Fourier coefficients, they represent the phase and amplitude estimates of the power frequency harmonics;

[0218]

[0219] In the formula, T represents the total duration of the collected noise in the time domain, and variables Xk and Yk are intermediate transition variables. They are sensitive to the offset between the current estimated angular frequency ω and the actual power frequency fundamental frequency, and are key intermediate variables in the frequency estimation process.

[0220]

[0221] in, Represents the angular frequency offset, that is, the correction between the current estimated frequency value and the next estimated frequency value, which is calculated by the above four estimated intermediate variables. All are excessive variables;

[0222] The NGSE frequency estimation algorithm is an iterative optimization process. The angular frequency ω approaches the real power frequency fundamental frequency of the noise with each iterative optimization, and the angular frequency offset approaches 0 with the iterative optimization. When the NGSE frequency estimation algorithm is finished, the angular frequency ω and the angular frequency offset tend to be stable. Finally, the power frequency fundamental frequency f is converted using the estimated angular frequency ω.

[0223] After obtaining the actual estimated value of the industrial frequency base frequency, the second step is to set the transient electromagnetic emission mode. The emission current waveform uses a bipolar square wave with a duty cycle of 25%, and uses the industrial frequency base frequency f obtained by the Wavelet-NGSE algorithm to adjust the period of the bipolar emission square wave to 2 / f (the emission frequency is f / 2). When the transmitting coil 1 is working, the receiving coil 3 can collect the transient electromagnetic second field signal generated at the moment the current is turned off. In each emission cycle, we can get a set of positive signals x + (n) and a set of negative signals x-(n).

[0224]

[0225] x - (n) = s - (n)+n T- (n)+n r- (n)

[0226] Signals + (n) is a positive signal, signal s - (n) is a negative signal, and the two groups of signals have the same amplitude and opposite positive and negative directions. T+ (n) and n T- (n) represents the power frequency harmonics coupled into the signal. Since the duty cycle of the bipolar square wave is adjusted to 25% and the transmission period is an even multiple of the power frequency period, the power frequency harmonic n T+ (n) and n T- (n) have the same phase, so the amplitude is also the same.

[0227] Finally, by performing a difference operation on adjacent signals, the power frequency harmonics in the original data can be removed. The specific formula is as follows:

[0228]

[0229] By using this method, we only need to perform one subtraction operation to effectively remove the power frequency harmonics without causing distortion of the TEM signal.

[0230] In order to verify the performance of the power frequency algorithm based on Wavelet-NGSE frequency estimation, we conducted experimental tests.

[0231] According to the above power frequency harmonic formula, we set the parameters as follows: the power frequency base frequency is 51Hz, the total number of harmonics is 3, and the initial power frequency phase is generated by a random number. The experimental signal is a one-dimensional forward modeling signal generated by a uniform earth model. The waveform of the simulation signal and the signal-noise coupling data is shown in the figure below. Figure 1-2 As shown. The coupling mode of signal and noise is to simulate the bipolar transmission signal mode, and the simulated transmission frequency is 50.5Hz (half of the industrial frequency base frequency, even multiple minus). Among them, the signal-to-noise ratio of the first positive signal is -38.31dB. Further, it needs to be explained that Figure 19-22 The horizontal axis time represents time in milliseconds (ms), and the vertical axis Amplitude represents amplitude in volts (V). You can also refer to Figure 5 The contents of the horizontal and vertical axes.

[0232] The experiments were conducted using the traditional bipolar power frequency suppression method based on the 50Hz assumption and the bipolar power frequency suppression method based on the actual simulated base frequency of 51Hz. First, assuming that the base frequency of the power frequency harmonic is 50Hz, the same simulation signal is used to simulate the bipolar positive and negative signals to ensure that the positive and negative signal cycle period is 40ms, which is twice the ideal power frequency period. Then, the adjacent positive and negative signals are subtracted and divided by two to complete the power frequency suppression. The waveform after power frequency suppression is shown in the figure below. Fig.21 As shown in the figure, from the local amplified waveform of the signal from 2ms to 10ms, it can be seen that there are still many power frequency harmonics remaining after this power frequency suppression. Through calculation, this power frequency suppression improves the signal-to-noise ratio of the signal from -38.31dB to -5.65dB.

[0233] Then, we used the real simulation base frequency to conduct a bipolar power frequency suppression experiment, with 51Hz as the power frequency base frequency, and used the same simulation signal as the above experiment to simulate the bipolar positive and negative signals, ensuring that the positive and negative signal cycle period is 39.2ms, which is twice the actual simulation power frequency period. Then, the adjacent positive and negative signals are subtracted and divided by two to complete the power frequency suppression. The waveform after power frequency suppression is shown in the figure below. Figure 4 As shown in the figure, from the local amplified waveform of the signal from 2ms to 10ms, it can be seen that after the power frequency suppression, almost all the power frequency harmonic components are suppressed, and only random noise is retained. And through calculation, the power frequency suppression improves the signal-to-noise ratio of the signal from -38.31dB to 1.55dB. This simulation experiment proves that when the power frequency fundamental frequency fluctuates from the ideal 50Hz to 51Hz, the actual fundamental frequency of the power frequency is accurately obtained, and the effect of power frequency suppression based on this is significantly improved compared with the traditional 50Hz fundamental frequency assumption.

[0234] On the basis of filtering out power frequency harmonic noise, in order to further suppress random noise, this application explores a data processing method based on minimum noise separation. Minimum noise separation (Minimum Noise Fraction, MNF) is a statistical analysis method based on noise covariance estimation. The decomposition idea is to decompose the original observation data into multiple components with certain rules and relationships. When they are applied to noise reduction processing, signal-noise separation is achieved through reconstruction of a certain threshold value to achieve the purpose of improving the signal-to-noise ratio. The transformation process is to use a rotation matrix to associate the transformed data with the signal-to-noise ratio, and the noise covariance is used to adjust the original data, and the processing becomes a minimum noise separation component (Noise Fraction, NF) with a signal-to-noise ratio arranged from high to low. The rotation matrix is ​​formed based on the use of noise covariance estimation and data covariance. Since the low-order minimum noise separation component has a significantly high signal-to-noise ratio, a strong filtering procedure can be applied to it without seriously reducing the signal quality of the data. When transient electromagnetic instruments perform multiple measurements of the same number within a unit equivalent measurement point, the presence of multiple approximately identical information components will be reflected in the magnitude of the error value of the signal covariance matrix, where the covariance of linear algebra represents the correlation of data vectors. Therefore, signal-to-noise separation can be performed through mathematical transformation. Multiple secondary field data matrices within the equivalent measurement point can be expressed as:

[0235]

[0236] Among them, each channel of the noisy data X has been zero-meaned, and the matrix contains n secondary field signals, each signal contains m time channels, X S represents the theoretical electromagnetic response of the secondary field, X N Represents noise, the signal and noise are uncorrelated with each other.

[0237] The covariance matrix of X is expressed as:

[0238] C=C S +C N

[0239] Among them C S Represents X S The covariance matrix, C N Yes X N The covariance matrix, C N is decomposed into eigenvalues ​​and eigenvectors as shown below:

[0240]

[0241] Where D N The diagonal matrix representing the eigenvalues ​​in descending order, U NDenotes the eigenvector associated with each eigenvalue. The minimum noise separation component is defined as the ratio of the variance of the noise to the total variance of each minimum noise separation component, and is related to the signal-to-noise ratio:

[0242]

[0243] in, represents the noise variance of the i-th component, Represents the variance of the noisy signal of the ith component. Noise covariance estimation is the key to minimum noise separation. This application takes the maximum autocorrelation factor method as an example to estimate the noise covariance matrix C N , the covariance matrix C N By element γk q constitute:

[0244]

[0245] Z kj =X k(j+1) -X kj

[0246] Z qj =X q(+1) -X qj

[0247] Among them, represents the average value of the kth time channel, Represents the average value of the qth time channel. Construct the matrix

[0248] P=U N D N -0.5

[0249] Make P T C N P is the identity matrix. Then

[0250] Y=PX

[0251] Then perform the following calculation:

[0252] C X =P T CP=VD X V T

[0253] Where D X and P represent the eigenvalue matrix and eigenvector matrix respectively. The second rotation matrix R=PV is constructed, and the electromagnetic data matrix X is linearly transformed to obtain the minimum noise separation component NF, that is, the matrix ψ.

[0254]

[0255] Finally, the data is reconstructed by retaining the first L (L < m) minimum noise separation components. The reconstructed signal is expressed as:

[0256]

[0257] Same as the transient electromagnetic data X before processing, The matrix size remains unchanged.

[0258] The minimum noise separation simulation test is used to verify the effects of the two in the noise reduction process, such as Fig.23 The figure shows the test results of minimum noise separation and stacking average, among which the minimum noise separation takes the first minimum noise separation component for the best effect. The ESNR of the noisy signal before processing is 3.6dB, and the ESNR after using minimum noise separation and its combined stacking average technology are 17.1dB and 22.6dB respectively. It can be seen that each step of processing will improve the signal-to-noise ratio. Compared with stacking average, minimum noise separation improves the signal-to-noise ratio more. It can be seen before and after processing that the random noise suppression effect is obvious, and the quality of transient electromagnetic data is significantly improved within a limited number of stacking times.

[0259] The traditional MNF algorithm can only suppress random noise to a certain extent, and when the random noise intensity is high, the effect of the MNF algorithm is poor, resulting in some residual random noise components in the data. Therefore, based on the traditional MNF algorithm to process data, this chapter introduces a deep learning algorithm, uses a neural network model to identify effective signal features, and uses a deep learning model to suppress residual random noise on the basis of suppressing power frequency interference and some random noise in turn through the traditional algorithm, and completes the strong noise suppression in the urban environment.

[0260] Combining the characteristics of transient electromagnetic signals and the special network structure proposed above, in order to suppress residual noise, we propose a new model CNN-GRU model to suppress the residual random noise of transient electromagnetic. The model structure is as follows Fig.24 As shown, Figure 6In the model, Noisy signal is the noise signal, MNF is Minimum Noise Fraction, GRU is Gated Recurrent Unit, Input layer is the input layer, Conv layer is the convolution layer, reconstruction layer is the reconstruction layer, Processed signal is the processed signal, Theoreticalsignal is the theoretical signal, LossA and LossB are both loss functions. The model has two loss functions, where LossA is the loss between the reconstructed signal (from the reconstruction layer) and the processed theoretical signal, LossB is the loss between the noise signal and the processed theoretical signal, and Loss is the total loss. The final loss is the weighted sum of LossA and LossB, with weights α and (1-α) respectively. The CNN-GRU model suppresses noise based on the traditional denoising algorithm. First, the transient electromagnetic signal is preprocessed using the power frequency resonance self-cancellation algorithm and the MNF algorithm to initially suppress random noise, and then the deep learning model is used to complete the residual noise suppression, greatly improving the quality of transient electromagnetic signals. The main structure of the network is composed of a convolutional neural network (CNN) and a GRU. The convolutional neural network and the GRU respectively extract the spatial and temporal features of transient electromagnetic data, while taking advantage of the symmetric high-dimensional feature extraction of the autoencoder.

[0261] In order to verify the random noise suppression performance of the CNN-GRU model, a data processing experiment was conducted using measured urban noise. After power frequency preprocessing, it was coupled with the same transient electromagnetic data, and the signal-to-noise ratio of the data before and after processing was calculated to quantify the performance of the algorithm. The SFSDSA algorithm, LSTM-Autoencoder algorithm and CNN-GRU model were used for noise suppression comparison experiments. The processing results of the three algorithms are shown in Figure 2. Fig.25 As shown, Fig.25 In the figure, amplitude (V) indicates the amplitude of the signal in volts (V), t (ms) indicates the time in milliseconds (ms), signal with noise indicates a signal with noise, and signal indicates a signal without noise.

[0262] Through calculation, the signal-to-noise ratio of the transient electromagnetic signal before processing is -14.3dB. The signal-to-noise ratios of the SFSDSA algorithm, LSTM-Autoencoder algorithm and CNN-GRU model after random noise suppression are increased to 11.1dB, 15.8dB and 23.5dB respectively. All three algorithms effectively improve the quality of transient electromagnetic data, and the signal-to-noise ratio improvement value of the CNN-GRU model is greater. In addition, it can also be clearly seen from the waveforms before and after signal processing that the transient electromagnetic signal processed by the CNN-GRU model is most consistent with the ideal transient electromagnetic signal, which is consistent with the results of the simulated noise suppression experiment. Therefore, the CNN-GRU model has the best effect in processing urban random noise after power frequency suppression, and the quality of transient electromagnetic data is significantly improved.

[0263] Comparative Example 1:

[0264] Compared with Example 1, this comparative example uses enameled wires of the same wire diameter and is wound according to coil skeletons of the same size. The same size here refers to the same overall size, not the same structure, that is, the same overall circular size. The ordinary circular coil in the prior art is used. The results are as follows: Fig.16 shown.

[0265] In summary, the capacitance and inductance of the honeycomb coil are smaller than those of the circular coil with the same number of turns and the same equivalent area. Compared with the circular coil under the same conditions, the honeycomb coil has a higher resonant frequency and a larger bandwidth. Fig.16 As shown, the highest frequency band of the honeycomb coil is 213kHz (i.e., the curve corresponding to 213kHz in the figure is the result of implementation 1, and the other curve is the result of comparative example 1), which can expand the bandwidth of the traditional circular coil by 60%, effectively improving the bandwidth of multi-turn small loop reception, and can improve the shallow detection capability.

[0266] Comparative Example 2:

[0267] This comparative example uses a low-noise integrated amplifier LT1028. The results are as follows Fig.18 shown.

[0268] In summary, the background noise of the amplifier in Example 1 was tested in an electromagnetic shielding room using a dynamic signal analyzer Agilent 35670, and compared with the original low-noise integrated amplifier LT1028. Under other same conditions, when the amplifiers were different, the background noise comparison between Example 1 and Comparative Example 2 was as follows: Fig.18 As shown, the amplifier in Example 1 has lower noise than LT1028 in the transient electromagnetic frequency band, with a minimum of 0.6nV / Hz0.5.

[0269] Based on the urban transient electromagnetic noise suppression strategy proposed in this application, a noise suppression experiment was carried out on a set of measured transient electromagnetic data. According to the above content, three treatments of strong power frequency suppression, random noise suppression and residual random noise suppression were carried out respectively. By analyzing the transient electromagnetic extraction profile data before and after the treatment, the effectiveness of the noise suppression strategy in this paper was verified.

[0270] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and the present invention can be implemented in other specific forms without departing from the spirit or basic features of the present invention. Therefore, no matter from which point of view, the embodiments should be regarded as exemplary and non-restrictive. The scope of the present invention is defined by the attached claims rather than the above description, and it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims are included in the present invention. Any figure mark in the claims should not be regarded as limiting the claims involved.

[0271] In addition, it should be understood that although the present specification is described according to implementation modes, not every implementation mode contains only one independent technical solution. This description of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment can also be appropriately combined to form other implementation modes that those skilled in the art can understand.

Claims

1. A towed transient electromagnetic detection system, characterized in that: include: A detection module, the detection module comprising a transmitting coil, a receiving coil and a host, the transmitting coil is laid on a movable towing platform, the receiving coil is laid on a high platform that can move relative to the towing platform, the high platform is arranged above the towing platform, the plane where the receiving coil is located is parallel to the plane where the transmitting coil is located, the host is arranged on the towing vehicle platform, one end of the host is connected to the transmitting coil, and the other end is connected to the receiving coil, the vertical distance between the high platform and the towing vehicle platform is adjustable, the horizontal relative position of the receiving coil and the transmitting coil is also adjustable, and the position of the receiving coil is determined when the mutual inductance of the transmitting coil and the receiving coil is zero; A signal processing module is connected between the host and the receiving coil. The signal processing module includes an amplifier and a noise processing module. One end of the amplifier is connected to the receiving coil, and the other end is connected to the noise processing module. The end of the noise processing module away from the amplifier is connected to the host to obtain a higher quality electromagnetic signal.

2. A towed transient electromagnetic detection system according to claim 1, characterized in that: The radius R of the transmitting coil is 0.5m, the radius r of the receiving coil is 0.125m, the direction of the current emitted by the host to the transmitting coil is counterclockwise, the relative vertical height H between the receiving coil and the transmitting coil is 0.2m, and the horizontal distance L between the center of the receiving coil and the center of the transmitting coil is 0.6108m.

3. A towed transient electromagnetic detection system according to claim 1 or 2, characterized in that: The receiving coil is configured as an equally spaced cross-coupled honeycomb coil, comprising: A skeleton body, wherein the skeleton is in an annular shape as a whole; First winding parts, which are spaced apart and circumferentially distributed on the outer ring wall of the skeleton body; A second winding portion, the second winding portion is also spaced and circumferentially distributed on the outer ring wall of the skeleton body, and the second winding portion is arranged below the first winding portion; The enameled wire has multiple turns which are cross-wound on the first winding part and the second winding part at equal intervals in sequence.

4. A towed transient electromagnetic detection system according to claim 3, characterized in that: The spacing distances between each adjacent first winding portion are the same, the spacing distances between each adjacent second winding portion are also the same, and the spacing distance between each adjacent first winding portion is the same as the distance between each adjacent second winding portion.

5. A towed transient electromagnetic detection system according to claim 4, characterized in that: There is a first spacing space between each adjacent first winding portion, there is a second spacing space between each adjacent second winding portion, and there is a third spacing space between each pair of first winding portions and second winding portions corresponding to each other in the vertical direction; The enameled wire rests on the upper end surface of at least one first winding part, and then passes through the first spacing space, and then obliquely downwards passes through the third spacing space and the second spacing space in sequence, and then rests on the lower end surface of at least one second winding part, and then passes through the second spacing space, and then obliquely upwards passes through the third spacing space and the first spacing space in sequence, and then rests on the upper end surface of at least one first winding part again, so as to be cyclically wound on the skeleton body.

6. A method for improving electromagnetic signal quality based on a towed transient electromagnetic detection system according to any one of claims 1 to 5, characterized in that: The following steps are involved:

1. Coil laying 1.1) Preliminary laying of coil structure; 1.2) According to the theoretical calculation and analysis of relevant parameters in the existing coil structure, select a more appropriate vertical distance H between the receiving coil and the transmitting coil; 1.3) According to the vertical distance L determined in step 1.2), further theoretically calculate the horizontal distance L between the center of the receiving coil and the center of the transmitting coil corresponding to this height; 1.4), according to the results in step 1.2) and step 1.3), adjust the position of the receiving coil in the coil structure; 1.5) Performing transient electromagnetic detection using the coil structure adjusted in step 1.4); 2. Amplifier Circuit Setup The preamplifier adopts a hierarchical amplification circuit, which includes a signal receiving and processing circuit, a primary amplification circuit and a secondary amplification circuit. The input end of the signal receiving and processing circuit is connected to the receiving coil, the output end of the signal receiving and processing circuit is connected to the input end of the primary amplification circuit, the output end of the primary amplification circuit is connected to the input end of the secondary amplification circuit, and the output end of the secondary amplification signal circuit is connected to the host; 3. Electromagnetic noise suppression 3.

1. Use Wavelet-NGSE algorithm to suppress power frequency noise in the noise; 3.1.1) Collect a set of electromagnetic signals with environmental noise and pre-process the electromagnetic field signals using a wavelet threshold algorithm; 3.1.2) Use Wavelet-NGSE frequency estimator to estimate the fundamental frequency f of the power frequency component in the noise; 3.1.3) Adjust the emission current cycle to an even multiple of the power frequency cycle as the transient electromagnetic emission mode; 3.1.4) Collect signals, perform difference calculation on adjacent signals, and remove power frequency harmonics in the original data; 3.

2. Suppressing random noise in noise through minimum noise separation 3.2.1) Adjust the data processed in step 1.6) by noise covariance to generate the minimum noise separation component; 3.2.2) Forming a rotation matrix based on the use of noise covariance estimation and data covariance; 3.2.3) Use the rotation matrix to associate the transformed data with the signal-to-noise ratio; 3.

3. Use the CNN-GRU model to suppress the residual random noise in the noise in step 2.3) 3.3.1) Preprocess the transient electromagnetic signal using the power frequency resonance self-cancellation algorithm and the MNF algorithm; 3.3.2) Use deep learning model to suppress residual noise.

7. A method for improving the quality of electromagnetic signals according to claim 6, characterized in that: In the step 1, The content of the preliminary laying of the coil structure in step 1.1 is: setting the plane where the towing vehicle platform is located parallel to the plane where the platform is located, laying the transmitting coil on the towing vehicle platform, and laying the receiving coil on the platform; The specific contents of the step 1.2 are as follows: 1.2.1), taking different vertical distances H between the receiving coil and the transmitting coil, adjusting the horizontal distance L between the center of the circular receiving line and the center of the transmitting coil, and obtaining the relationship between the field strength of the primary field in the vertical direction and the horizontal distance L at different vertical distances H; 1.2.2) Take the vertical distance H where the primary field changes less drastically with different horizontal distances L; 1.2.3) Take the approximate horizontal distance L1 where the primary field changes most drastically with different vertical distances H, and the relevant parameters in the coil structure include the number of turns of the transmitting coil, the number of turns of the receiving coil, the radius R of the transmitting coil, the radius r of the receiving coil, the current intensity I emitted by the subject to the transmitting coil, the current direction and the horizontal distance L between the center of the receiving coil and the center of the transmitting coil; The specific contents of step 1.3 are as follows: 1.3.1), placing the circular receiving coil at the vertical distance H in step 2.2) to the approximate horizontal distance L1 in step 2.3); 1.3.2), calculating the mutual inductance coefficient between the transmitting coil and the circular receiving coil to obtain the size of the horizontal distance L at zero coupling; In the step 1.4, the requirement for adjusting the coil structure in step 1) according to the calculation result is: in actual operation, the error between the horizontal distance between the center of the receiving coil and the center of the transmitting coil after adjustment and the horizontal distance L between the center of the receiving coil and the center of the transmitting coil obtained by theoretical calculation in step 3) shall not exceed 2 mm.

8. A method for improving the quality of electromagnetic signals according to claim 6, characterized in that: The specific contents of the step 2 include: the signal receiving and processing circuit includes a positive power supply voltage VDD, a negative power supply voltage VSS, resistors R1-resistor R6, an electromagnetic signal input interface Vsig+ and an electromagnetic signal input interface Vsig-, the positive power supply voltage VDD is connected to resistors R1 and R2, the resistors R1 and R2 are connected in parallel, in the branch where the resistor R1 is located, the resistor R1 is connected in series with the electromagnetic signal input interface Vsig+ and the resistor R3 in sequence, in the branch where the resistor R2 is located, the resistor R2 is connected in series with the electromagnetic signal input interface Vsig- and the resistor R4 in sequence, the branch where the resistor R1 is located and the branch where the resistor R1 is located are connected in parallel with the resistor R4. R6 is connected in series, one end of the resistor R6 is connected to the resistor R3 and the resistor R4 arranged in parallel, and the other end is connected to the negative power supply voltage VSS, and the resistor R6 is connected in parallel with the resistor R5, one end of the resistor R5 is connected to the resistor R6, and the other end is grounded; the preliminary discharge circuit includes an amplifier A1, an amplifier A2, a capacitor C1-capacitor C4, a resistor R7, a resistor R8, a resistor R9, a resistor R10, a resistor R13 and a resistor R14, the capacitor C1 is connected in series with the resistor R7 and connected to the inverting input end of the amplifier A1, the end of the capacitor C1 away from the resistor R7 is connected to the node between the resistor R1 and the electromagnetic signal input interface Vsig+, the capacitor C 3 is connected in parallel with the resistor R8 and then connected to the in-phase input terminal of the amplifier A1; the end of the capacitor C3 away from the amplifier A1 is grounded; the output terminal of the amplifier A1 is connected in series with the resistor R13 and the resistor R15 in sequence; the capacitor C3 is connected in series with the resistor R9 and then connected to the inverting input terminal of the amplifier A2; the end of the capacitor C3 away from the resistor R9 is connected to the node between the resistor R2 and the electromagnetic signal input interface Vsig-; the capacitor C4 is connected in parallel with the resistor R10 and then connected to the in-phase input terminal of the amplifier A2; the end of the capacitor C4 away from the amplifier A2 is grounded; the output terminal of the amplifier A2 is connected in series with the resistor R14 and the resistor R16 in sequence; the secondary The amplifier circuit includes an amplifier A3, resistors R18-R20, a capacitor C5 and an electromagnetic signal output interface Vout, wherein the inverting input terminal of the amplifier A3 is connected to the resistor R15, the non-inverting input terminal of the amplifier A3 is connected to the resistor R16, one end of the resistor R18 is connected to the node between the amplifier A3 and the resistor R16, and the other end is grounded, the output end of the amplifier A3 is connected in series with a resistor R19, a resistor R19 and an electromagnetic signal output interface Vout in sequence, the electromagnetic signal output interface Vout is connected to the host, one end of the capacitor C5 is connected to the node between the resistor R19 and the electromagnetic signal output interface Vout, and the other end is grounded;10. The preliminary amplification circuit also includes resistors R11 and R12. One end of the resistor R11 is connected to the node between the resistor R7 and the inverting input terminal of the amplifier A1, and the other end is connected to the node between the output terminal of the amplifier A1 and the resistor R13. One end of the resistor R12 is connected to the node between the resistor R9 and the inverting input terminal of the amplifier A2, and the other end is connected to the node between the output terminal of the amplifier A2 and the resistor R14. The secondary amplification circuit also includes a resistor R17. One end of the resistor R17 is connected to the node between the resistor R15 and the inverting input terminal of the amplifier A3, and the other end is connected to the node between the output terminal of the amplifier A3 and the resistor R19. ; 9. The method for improving the quality of electromagnetic signals according to claim 6, characterized in that: In the step three, The specific contents of step 3.1.1) are: 3.1.1.1) Collect a set of environmental noises, the collection time of which is greater than or equal to 60 seconds; 3.1.1.2) Then use the wavelet threshold algorithm to treat the power frequency component in the noise as a valid signal, and pre-process the pure noise using the following formula: in, The wavelet base is 'db8', the threshold function is the soft threshold, n(t) is the pure environmental noise, and n'(t) is the noise after wavelet preprocessing; The specific contents of step 3.1.2) are: 3.1.2.1) Establish the NGSE frequency estimation equation: f=ω / 2π in, k=1,2,...,k max , k max is equal to the total number of power frequency harmonics to be estimated, n is the sampling point in the time domain and N corresponds to the total number of sampling points, N is proportional to the sampling frequency and the total duration of the noise, Δt is the time interval between sampling points, which is inversely proportional to the sampling frequency, n'(n) is the preprocessed noise data obtained in step 1.2), ω is the angular frequency, the initial frequency of f is set to 50Hz, A k and B k Similar to the Fourier coefficients, they represent the phase and amplitude estimates of the power frequency harmonics; In the formula, T represents the total duration of the collected noise in the time domain, and variables Xk and Yk are intermediate transition variables. They are sensitive to the offset between the current estimated angular frequency ω and the actual power frequency fundamental frequency, and are key intermediate variables in the frequency estimation process. in, Represents the angular frequency offset, that is, the correction between the current estimated frequency value and the next estimated frequency value, which is calculated by the above four estimated intermediate variables. All are excessive variables; 3.1.2.2) Substitute the noise data pre-processed in step 1.1.2) into the NGSE frequency estimation equation and iterate continuously, so that the offset of the angular frequency ω is constantly close to 0, and the angular frequency ω is constantly close to the real power frequency fundamental frequency of the noise; 3.1.2.3) Obtain the power frequency base frequency f based on the estimated angular frequency ω; The specific content of step 3.1.3) is: according to the industrial frequency base frequency f obtained in step 3.1.2.3), the period of the emission current waveform is adjusted to 2 / f, that is, the emission frequency is f / 2, as the setting of the transient electromagnetic emission mode, wherein, The transmitting current waveform adopts a bipolar square wave with a duty cycle of 25%; The specific content of step 3.1.4) is: 3.1.4.1) Start the survey device according to the transient electromagnetic emission mode set in step 1.3); 3.1.4.2) The transient electromagnetic field signals generated at the moment of current shutoff are collected by the receiving coil. In each transmission cycle, a group of positive signals and a group of negative signals can be obtained; 3.1.4.3) Subtract the collected adjacent positive and negative signals and take the average to complete power frequency suppression; The specific content of the step 3.2.1) is: adjusting the content processed in step 3.1.4.3) by noise covariance to process it into the minimum noise separation component with the signal-to-noise ratio arranged from high to low; The specific content of step 3.2.2) is: 3.2.2.1) Generate multiple secondary field data matrices within equivalent measurement points: in, Each channel of the noisy data X has been zero-meaned, and the matrix contains n secondary field signals, each of which contains m time channels, X S represents the theoretical electromagnetic response of the secondary field, X N represents noise, the signal and noise are uncorrelated with each other; 3.2.2.2) Generate the cosquare matrix of X as: C=C S +C N Among them C S Represents X S The covariance matrix, C N Yes X N The covariance matrix of 3.2.2.3) The covariance matrix C N The decomposition into eigenvalues ​​and eigenvectors is as follows: Where D N The diagonal matrix representing the eigenvalues ​​in descending order, U N represents the eigenvector associated with each eigenvalue; In the step 3.2.3), the specific content is: The minimum noise separation component is related to the signal-to-noise ratio: in, represents the noise variance of the i-th component, represents the variance of the noisy signal of the i-th component; The noise covariance matrix C N By element γ kq constitute: Z kj =X k(j+1) -X kj Z qj =X q(j+1) -X qj Among them, represents the average value of the kth time channel, represents the average value of the qth time channel; Construct matrix: P = U N D N -0.5 Make P T C N P is the identity matrix; Then Y = PX Then perform the following calculation: C x =P T CP=VD x V T Where D X and P represent the eigenvalue matrix and eigenvector matrix respectively; The second rotation matrix R = PV is constructed, and the electromagnetic data matrix X is linearly transformed to obtain the minimum noise separation component NF, that is, the matrix ψ: Finally, the data is reconstructed by retaining the first L (L < m) minimum noise separation components. The reconstructed signal is expressed as: Same as the transient electromagnetic data X before processing, The matrix size remains unchanged.

10. The method for suppressing urban electromagnetic noise based on multi-source noise according to claim 5, characterized in that: In step 1.4.2), a group of positive signals and a group of negative signals obtained in each transmission cycle include: x - (n)=s - (n)+n T- (n)+n r- (n) Signals + (n) is a positive signal, signal s - (n) is a negative signal. The two groups of signals have the same amplitude and opposite positive and negative directions. T+ (n) and n T- (n) all represent the power frequency harmonics coupled into the signal; The calculation process in step 1.4.3) is specifically as follows:

Images