A method for correcting system response of a time domain electromagnetic detection system

Abstract

The present application relates to a kind of system response correction methods of time domain electromagnetic detection system.The present application method is used to correct the influence of system response on effective signal, removes the influence of inherent characteristics of receiving sensor on effective signal by training adaptive output compensation of the theoretical and actual response of correction device;And by the real measured current waveform and adaptive current sensor compensation, the real transmitting current waveform is obtained, and the response of earth under unit step excitation is obtained according to its inverse function.The present application can remove the influence of system response on time domain electromagnetic detection system signal, improve signal quality, reduce the blind area time of system detection, improve the detection range of system.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of electromagnetic detection, and particularly relates to a system response correction method of a time domain electromagnetic detection system. BACKGROUND

[0002] Currently, time domain electromagnetic detection systems mainly have two research directions, i.e., improving the detection depth and the shallow detection capability of the system. In terms of improving the shallow detection capability of the system, system response correction is one of the hotspots.

[0003] The system response correction of the time domain electromagnetic detection system mainly consists of two aspects, i.e., the transmission characteristic correction of the receiving sensor and the suppression of the influence of the transmitting current.

[0004] CN107024724A discloses a method for quickly optimizing early data of helicopter airborne electromagnetic detection by deconvolution. The influence of the transmitting current waveform is removed by using the direct deconvolution method, but the influence of the receiving sensor on the effective signal and the influence of the current sensor on the current waveform are not considered. Stolz, Edward M., 1998, Legault, Jean M., et al., 2012, Wang Yaran, et al., 2014, Ren Guiying, 2020, etc. all mention that the deconvolution method based on SVD is used to remove the influence of the current waveform on the effective signal, but the influence of the current sensor on the actual transmitting current is not considered.

[0005] Zhang Shuang, et al. proposed that the slightly under-damped matching mode of the receiving sensor can significantly reduce the distortion of the early response, but cannot completely remove the influence of the receiving sensor. Ji Yanju, et al. proposed that the influence of the receiving sensor is removed by using the numerical calculation method, but the actual receiving sensor circuit parameters have precision limitations, and the test results have errors. Wang, Haowen, et al., 2019 proposed a field calibration scheme, which tests the transmission characteristics of the receiving sensor and corrects the characteristics. The scheme does not consider the influence of the actual measurement error and noise. SUMMARY

[0006] The technical problem to be solved by the application is to provide a system response correction method of a time domain electromagnetic detection system, which removes the influence of the receiving sensor and the transmitting current on the effective signal, reduces the detection blind area of the time domain electromagnetic detection system, and improves the accuracy and detection range of the system.

[0007] The application is implemented in the following way,

[0008] A system response correction method of a time domain electromagnetic detection system, comprising:

[0009] A correction coil is placed in the transmitting coil.

[0010] The runtime domain electromagnetic detection system repeatedly closes and opens the correction coil, collects induced voltage generated from the receiving coil when the correction coil is closed and opened, and obtains the measured response of the correction coil according to the collected induced voltage generated from the receiving coil when the correction coil is closed and opened;

[0011] The inverse function of the current sensor is calculated;

[0012] The real transmitting current is obtained according to the inverse function of the current sensor;

[0013] The theoretical response of the correction coil is calculated according to the real transmitting current;

[0014] The inverse function of the receiving sensor transfer function is obtained through a neural network or adaptive filtering according to the measured response and the theoretical response of the correction coil;

[0015] The unit step response after removing the system response is obtained according to the measured voltage collected in the receiving coil, the measured transmitting current, the inverse function of the current sensor and the inverse function of the receiving sensor transfer function.

[0016] Further, the real transmitting current is obtained according to the measured transmitting current i measured (t) and the time-domain representation of the inverse function of the current sensor h current (t) during system operation.

[0017] Further, the inverse function of the current sensor h current (t) is obtained by using a standard current source to generate a sine wave current I(t) with a constant amplitude and a logarithmically equidistantly spaced frequency, measuring the standard current through the current sensor to obtain a measured standard current I measured (t), and taking the measured standard current and the standard current as the input and output of a neural network or adaptive filter, respectively.

[0018] Further, the coefficients of the adaptive filter are defined as w(n), n=0,1,…,N, the measured standard current is represented in a discrete form as I measured (n), n=0,1,…,N, the simulated real current is represented in a discrete form as I actual (n)=I measured (n)*w(n), the error e(n)=I(n)-I actual (n) between the standard current of the standard source and the simulated real current, the adaptive filter parameters are adjusted to obtain a set of coefficients that minimize the error, and the coefficients are the inverse function of the current sensor, and the time-domain representation of the inverse function of the current sensor is

[0019] Further, the formula used to calculate the theoretical response of the correction coil according to the real transmitting current is: ​

[0020] where i actual (t) is the real transmit current, is the unit step response of the correction coil closed, u(t) is the unit step function, and δ(t) is the unit impulse function M LR is the mutual inductance between the receiving coil and the correction coil, M BL is the mutual inductance between the compensation coil and the correction coil, M TL is the mutual inductance between the transmitting coil and the correction coil, and τ is the time constant.

[0021] Further, the inverse function of the receiving sensor transfer function is obtained by a neural network or adaptive filtering, comprising:

[0022] The adaptive filter coefficient is defined as w(n), n=0, 1, …, N, and the measured response of the correction coil in discrete form is represented as The analog response of the analog real correction coil in discrete form is V TLR_actual (n) = V TLR_measured (n)*w(n), the error e(n) between the ideal response of the correction coil and the analog response of the correction coil is V TLR_ideal (n)-V TLR_actual (n), the adaptive filter parameters are adjusted to obtain a set of coefficients that minimize the error, which is the inverse function of the receiving sensor transfer function, and the time domain representation form of the inverse function is satisfies

[0023] Further, the unit step response after removing the system response is:

[0024]

[0025] wherein, is the time domain representation form of the inverse function of the receiving sensor transfer function, is the time domain representation form of the inverse function of the current sensor transfer function, i measured (t) is the measured transmit current, V measured (t) is the unit step response without removing the system response.

[0026] Compared with the prior art, the present application has the beneficial effects that:

[0027] The present application adopts a two-stage output compensation method to remove the influence of the receiving sensor and the transmit current on the effective signal, respectively.

[0028] The application adopts a method of self-adaptive output compensation of receiving sensor characteristics, adopts a combination of measurement and theoretical calculation, fully considers various influences of the system, and avoids signal measurement errors caused by electronic device errors and measurement errors in direct measurement.

[0029] The application adopts a method of self-adaptive deconvolution compensation of transmitting current influence, considers the characteristics of the current sensor, simplifies the application steps, can obtain a more real current signal, and accurately removes the influence. BRIEF DESCRIPTION OF DRAWINGS

[0030] Figure 1 The time domain electromagnetic detection system basic structure (a) provided by the embodiment of the application, the correction device principle diagram (b) correction coil response model;

[0031] Figure 2 The schematic diagram of the method described in the embodiment of the application;

[0032] Figure 3 The flow chart of the method described in the embodiment of the application;

[0033] Figure 4 The basic principle of the self-adaptive output compensation algorithm of the receiving sensor described in the embodiment of the application;

[0034] Figure 5 The basic principle of the self-adaptive deconvolution compensation algorithm of the current sensor described in the embodiment of the application;

[0035] Figure 6 The comparison chart of the correction coil induced voltage before and after removing the transmission characteristics of the receiving sensor in the embodiment of the application;

[0036] Figure 7 The induced voltage of the correction coil under the unit step current excitation in the embodiment of the application; DETAILED DESCRIPTION

[0037] In order to make the purpose, technical scheme and advantages of the application clearer, the application will be further described in detail below with embodiments. It should be understood that the specific embodiments described herein are only used to explain the application, and are not used to limit the application.

[0038] The system basic structure of the time domain electromagnetic detection system of the application, and the correction device placed like Figure 1 The correction device includes: the correction coil is placed in the area inside the transmitting coil, used to simulate the anomaly ring, connected to the receiver through the relay to control opening and closing, and the current acquisition.

[0039] The basic signal flow direction of the system is shown by the solid line in Figure 2 , and the system response correction method is as followsFigure 2 The system response correction method of the time domain electromagnetic detection system adopts a two-stage output compensation method. Figure 2 The first stage output compensation selects an adaptive output compensation mode to remove the influence of the receiving sensor transmission characteristics on the effective signal and obtain the real response of the actual system.

[0040] Referring to Figure 3 The method comprises the following steps:

[0041] A correction coil is placed in the transmitting coil;

[0042] The time domain electromagnetic detection system is operated, the correction coil is repeatedly closed and opened, the induced voltage generated from the receiving coil when the correction coil is closed and opened is collected, and the measured response of the correction coil is obtained according to the induced voltage generated from the receiving coil when the correction coil is closed and opened.

[0043] The inverse function of the current sensor is calculated;

[0044] The real transmitting current is obtained according to the inverse function of the current sensor;

[0045] The theoretical response of the correction coil is calculated according to the real transmitting current;

[0046] The inverse function of the receiving sensor transfer function is obtained through a neural network or adaptive filtering according to the measured response and the theoretical response of the correction coil;

[0047] The unit step response after removing the system response is obtained according to the measured voltage collected in the receiving coil, the measured transmitting current, the inverse function of the current sensor and the inverse function of the receiving sensor transfer function.

[0048] As shown in Figure 1 A correction device with known basic electrical parameters is installed on the time domain electromagnetic detection system, and the relative position of the correction device on the pod is determined. Then the time domain electromagnetic detection system is started, the transmitting base frequency is set to 25Hz bipolar trapezoidal wave, the oscilloscope is used to check whether the transmitting signal is normal, and then the induced voltage of the receiving coil and the transmitting current i measured (t) output by the current sensor are collected.

[0049] The closing and opening of the correction coil are controlled through the control button on the receiver, the induced voltage V closed (t) and V open (t) generated from the receiving coil when the correction coil is closed and opened are collected, and the measured response V TLR_measured (t) of the correction coil is obtained by subtracting V closed(t)-V open (t)。Multiple experiments are repeated to obtain multiple sets of measured responses of the correction coil.

[0050] The process of obtaining the real transmit current is as follows:

[0051] A standard current source is used to generate a sinusoidal current I(t) with constant amplitude and logarithmically spaced frequency, and the measured standard current I measured (t) is obtained by measuring the standard current with a current sensor. The measured standard current and the standard current are used as the input and output of the neural network or adaptive filter, respectively.

[0052] The measured current of the transmit coil is represented as Figure 4 The N-order adaptive filter coefficients are defined as w(n), n=0, 1, …, N, and the measured current is represented as The analog real current is represented as I actual (n) = I measured (n) * w(n), and the error e(n) between the standard current of the standard source and the analog current is I actual (n), the adaptive filter parameters are adjusted to obtain a set of coefficients that minimize the error, which is the inverse function of the current sensor transfer function, and its time-domain representation is The measured current of the transmit coil is represented as Figure 4 It is known that the measured current of the transmit coil and the real transmit current satisfy

[0053] The measured transmit current i measured (t) and the inverse function h current (t) of the current sensor transfer function are convolved to obtain the real transmit current i actual (t).

[0054] The theoretical response of the correction coil is calculated according to the real transmit current:

[0055] The unit step response of the correction coil when closed is calculated. After the positions and turns of the transmit coil, receive coil, compensation coil, and correction coil are measured, the following Nee Man formula is used:

[0056]

[0057] The mutual inductances M TL , M BL , M LR between the transmit coil and the correction coil, the compensation coil and the correction coil, and the receive coil and the correction coil are calculated, and after the resistance R L and inductance L L of the correction coil are measured, the time constant τ is obtained, and the formula for the correction coil under unit step current excitation is obtained:

[0058] Calculate the ideal response of the correction coil under real current excitation:

[0059]

[0060] Based on the measured and theoretical responses of the calibration coil, the inverse function of the receiving sensor transfer function is obtained through a neural network or adaptive filtering, including: obtaining the measured response V of the calibration coil. TLR_measured (t) and ideal response V TLR_ideal (t) are used as the input and output of the neural network or adaptive filter, respectively, to obtain the inverse function of the receiving sensor transfer function, and its time-domain representation is as follows. satisfy

[0061] Depend on Figure 5 We can assume that the coefficients of the Nth-order adaptive filter are defined as w(n), n = 0, 1, ..., N, and the discrete form of the measured response of the correction coil is expressed as V. TLR_measured (n), n=0,1,…,N, the discrete form of the simulated real correction coil response is V TLR_actual (n)=V TLR_measured (n)*w(n), the error between the ideal correction coil response and the simulated correction coil response is e(n)=V TLR_ideal (n)-V TLR_actual (n), by adjusting the parameters of the adaptive filter, a set of coefficients is obtained that minimizes the error. These coefficients are the inverse function of the receiver sensor's transfer function, and their time-domain representation is as follows. satisfy

[0062] like Figure 6 As shown, the ideal correction coil response V was obtained based on ideal current simulation. TLR_ideal (t), Measured correction coil response V TLR_measured (t). The response obtained after adaptive filtering is V. TLR_WF (t), by Figure 6 It can be seen that, taking 5% as the signal error limit, the adaptive filter's correction coil response reaches the error 42μs earlier than the measured correction coil response in the off-time phase, meaning the early usable time channel is advanced by 42μs. According to the diffusion depth formula... Where ρ, t, and μ0 represent the uniform earth resistivity, time, and vacuum permeability, respectively, indicating that the shallow detection capability is greatly improved after adaptive filtering of the shallow signals detected by the system.

[0063] The induced voltage V of the correction coil under the unit step current excitation after removing the system response TLR_step (t), such as Figure 7As shown, the actual correction coil response and the correction coil response under unit step current excitation have the following relationship:

[0064]

[0065] For the measured voltage signal collected from the receiving coil at a certain time during the system operation, on the basis of the measured transmitting current, the unit step response after removing the system response can be obtained as:

[0066]

[0067] Wherein, is the time domain representation of the inverse function of the receiving sensor transfer function, is the time domain representation of the inverse function of the current sensor transfer function, measured (t) is the measured transmitting current, V measured (t) is the unit step response without removing the system response.

[0068] The above only describes the preferred embodiments of the present application and is not intended to limit the present application. Any modification, equivalent replacement and improvement made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method of correcting the system response of a time domain electromagnetic survey system, characterized by, The method comprises: placing a correction coil inside the transmitting coil; running the time domain electromagnetic detection system, repeatedly closing and opening the correction coil, collecting the induced voltage generated from the receiving coil when the correction coil is closed and opened, and obtaining the measured response of the correction coil according to the collected induced voltage generated from the receiving coil when the correction coil is closed and opened; calculating the inverse function of the current sensor; obtaining the real transmitting current according to the inverse function of the current sensor; calculating the theoretical response of the correction coil according to the real transmitting current; obtaining the inverse function of the receiving sensor transfer function through a neural network or adaptive filtering according to the measured response and the theoretical response of the correction coil; obtaining the unit step response after removing the system response according to the measured voltage collected in the receiving coil, the measured transmitting current, the inverse function of the current sensor, and the inverse function of the receiving sensor transfer function.

2. The method of claim 1, wherein the system response correction method is applied to a time domain electromagnetic survey system, and wherein the system response correction method comprises: The real emission current is obtained from the real-time measured emission current i measured (t) and the inverse function h current (t) of the current sensor The convolution operation of the time-domain representation of (t) and 3. The method of claim 2, wherein the system response correction method of a time domain electromagnetic survey system is characterized by, The current sensor inverse function h current (t) is a standard current source to generate a sine wave current I(t) with constant amplitude and logarithmic interval frequency variation, and the measured standard current I measured (t) is obtained by measuring the standard current through the current sensor, and the measured standard current and the standard current are taken as the input and output of the neural network or adaptive filter, respectively.

4. The method of claim 3, wherein the system response correction method of a time domain electromagnetic survey system is characterized by, The coefficients of the adaptive filter are defined as w(n), n = 0, 1, ..., N, and the measured standard current is expressed in discrete form as I. measured (n), n=0,1,…,N, simulating the discrete form of real current as I actual (n)=I measured (n)*w(n), the error between the standard current of the standard source and the simulated real current is e(n)=I(n)-I actual (n), by adjusting the parameters of the adaptive filter, a set of coefficients is obtained that minimizes the error. These coefficients represent the inverse function of the current sensor, and their time-domain representation is as follows.

5. The method of claim 1, wherein the system response correction method is applied to a time domain electromagnetic survey system, and The formula used to calculate the theoretical response of the correction coil from the true transmit current is: where i actual (t) is the true transmit current, is the unit step response of the correction coil when closed, u(t) is the unit step function, δ(t) is the unit impulse function, M LR is the mutual inductance between the receive coil and the correction coil, M BL is the mutual inductance between the compensation coil and the correction coil, M TL is the mutual inductance between the transmit coil and the correction coil, and τ is the time constant.

6. The method of claim 1, wherein the system response correction of the time domain electromagnetic survey system is characterized by, The method for obtaining the inverse function of the receiving sensor transfer function through a neural network or adaptive filtering comprises: The adaptive filter coefficients are defined as w(n), n = 0, 1,..., N, the measured response of the correction coil is represented in discrete form as V TLR_measured (n), n = 0, 1,..., N, the analog response of the analog true correction coil is V TLR_actual (n) = V TLR_measured (n) = V TLR_ideal (n) = V TLR_actual (n), the adaptive filter parameters are adjusted to obtain a set of coefficients that minimize the error, which is the inverse function of the transfer function of the receiving sensor, and its time-domain representation is satisfies 7. The method of claim 1, wherein the system response correction of the time domain electromagnetic survey system is characterized by, The unit step response after removing the system response is: wherein is a time domain representation of the inverse of the sensor transfer function, is a time domain representation of the inverse of the current sensor transfer function, i measured (t) is the measured emitted current, V measured (t) is the unit step response without removal of the system response.

Images