A multi-frequency data anti-interference method for downhole direct current method monitoring

Abstract

This invention discloses a multi-frequency data anti-interference method for downhole DC electrical resistivity monitoring: Step 1: Supply a current signal containing multiple frequency components to the ground and record the time series of the transmitted current and received signal; Step 2: Detect the time series of the transmitted current and received signal to obtain the ground frequency response at the measuring point; Step 3: Calculate the signal-to-noise ratio of the received signal at different frequencies based on the Fourier transform result of the received signal time series at a single measuring point, and discard unqualified ground frequency responses; Step 4: Using the ground responses at different frequencies as the data source, determine whether the data volume at each measuring point is sufficient. If so, proceed to Step 5; otherwise, proceed to Step 6; Step 5: Evaluate the stability of the ground frequency response at each measuring point. If so, proceed to Step 6; otherwise, discard the data with the largest deviation and return to Step 4; Step 6: Calculate the ground frequency response. This invention improves the effective data signal-to-noise ratio while maintaining high working efficiency.

Description

Technical Field

[0001] This invention belongs to the field of downhole DC electrical method monitoring technology, specifically a multi-frequency data anti-interference method for downhole DC electrical method monitoring. Background Technology

[0002] Downhole DC electrical resistivity tomography (DCIS) monitoring is an important means of preventing water hazards to the roof and floor during the mining process. It offers advantages such as automation, intelligence, all-weather operation, full waveform coverage, and uninterrupted data acquisition. However, underground electromagnetic noise is complex, extremely intense, and varies significantly with time and space, resulting in low signal-to-noise ratios (SNR) in the raw data. Currently, there are two main methods to improve the SNR of DCIS data: first, increasing the strength of the effective signal, including increasing the transmission current and decreasing the transmitter-receiver distance; second, reducing the impact of noise on the transmission frequency, including increasing the sampling duration, increasing the sampling frequency, and averaging multiple measurements. However, increasing the transmission current is limited by underground safety production constraints, and decreasing the transmitter-receiver distance is limited by the large working face and limited construction space, making their application in underground DCIS monitoring quite difficult. Furthermore, methods such as increasing the sampling duration, increasing the sampling frequency, and averaging multiple measurements under full waveform acquisition conditions place certain demands on the data storage capacity of the monitoring instrument. Increasing the acquisition duration also means reduced work efficiency and an inability to provide timely feedback on changes in the monitored object over time. Therefore, conventional methods for improving the SNR of DCIS are limited by underground monitoring conditions and rarely achieve ideal results.

[0003] Conventional direct current (DC) electrical resistivity tomography (DCPS) typically employs geometric sounding, meaning the exploration depth is determined by the arrangement of geometric parameters, independent of frequency. Therefore, DCPS often uses single-frequency signals, and statistical methods such as averaging multiple measurements are also applied to signals of a single frequency. In contrast, frequency-domain electromagnetic methods control exploration depth by changing the frequency. Multiple frequencies can be transmitted and received simultaneously, and the geoelectric response at different frequencies can be extracted using methods such as direct spectral analysis or correlation spectral analysis. The differences in geoelectric responses at different frequencies represent geoelectric information at different depths. Because multi-frequency signals have good correlation characteristics and strong anti-interference capabilities, they perform better in environments with strong interference. To improve the data quality of downhole DCPS monitoring signals and ensure the accuracy and real-time nature of monitoring results, the use of multi-frequency signals instead of single-frequency signals in downhole DCPS monitoring is considered.

[0004] However, the one-to-one correspondence between frequency and exploration depth in frequency-domain electromagnetic methods presents two problems in downhole DC electrical resistivity tomography (DMT) monitoring: (1) Due to the low transmission frequency and the generally short transmission-receiver distance (less than 1 km), downhole DMT monitoring typically struggles to achieve frequency-domain depth sounding of multi-frequency signals. This means that the ground responses of different frequencies are nearly identical and essentially represent geoelectric information at the same depth, necessitating the selection of only one frequency's ground response from the multi-frequency signal while discarding others, resulting in a significant waste of data. (2) Compared to single-frequency signals, the transmission current of a single frequency is relatively small, leading to a significant reduction in the strength of the received effective signal and consequently a decrease in the signal-to-noise ratio of a single frequency in the multi-frequency signal. These two problems demonstrate that, in the field of downhole DMT monitoring, simply copying the one-to-one correspondence between frequency and exploration depth in frequency-domain electromagnetic methods cannot fully utilize the anti-interference capabilities of multi-frequency signals. Currently, multi-frequency signals are rarely used in downhole DC electrical resistivity methods, and there is a lack of corresponding multi-frequency data application methods; at the same time, searching with the combination of "DC electrical resistivity method" and "multi-frequency data" yielded no relevant patents.

[0005] In summary, downhole DC electrical resistivity tomography (DCIP) monitoring faces complex and variable high-noise environments. While multi-frequency signals can be used to improve the signal-to-noise ratio of monitoring data, simply adopting the one-to-one correspondence between frequency and exploration depth in frequency-domain electromagnetic methods makes it difficult to leverage the advantages of multi-frequency signals. Therefore, based on the theoretical characteristic of multiple frequencies corresponding to the same exploration depth in downhole DCIP, it is necessary to develop a multi-frequency data processing technology adapted to downhole DCIP monitoring to achieve the goal of leveraging the characteristics of multi-frequency signals and improving the effective data signal-to-noise ratio. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides a method for frequency selection and multi-frequency averaging of multi-frequency signals from downhole DC electrical resistivity tomography (DCIP) monitoring. This method offers the advantage of improving the effective data signal-to-noise ratio while maintaining high operating efficiency, and solves the problem that multi-frequency signals from downhole DCIP monitoring are difficult to effectively resist interference under existing technological conditions.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] A method for resisting interference from multi-frequency data in downhole DC electrical monitoring includes the following steps:

[0009] Step 1: The transmitter supplies a current signal containing multiple frequency components to the ground through the transmitting electrode and records the time series of the transmitted current. At the same time, it records the time series of the electromagnetic field response signal through the receiving electrode as the time series of the received signal. Both the time series of the transmitted current and the time series of the received signal proceed to step 2.

[0010] Step 2: Perform correlation detection on the time series of the transmitted current and the time series of the received signal to obtain the ground frequency response of each measuring point at each frequency. The time series of the received signal (time domain data) and the ground frequency response (frequency domain data) of the measuring point are both entered into Step 3.

[0011] Step 3: Using a single measurement point as a unit, calculate the signal-to-noise ratio (SNR) of the received signal at different frequencies based on the Fourier transform results of the time series of the received signal, evaluate the degree of noise influence on the received signal at different frequencies, remove unqualified ground frequency responses, and proceed to Step 4 with qualified ground frequency response data.

[0012] Step 4: Using a single measuring point as the unit and the earth response at different frequencies as the data source, determine whether the amount of data at each measuring point is sufficient. If the measuring point has sufficient data to participate in the averaging calculation, proceed to Step 5; otherwise, proceed to Step 6.

[0013] Step 5: Evaluate the stability of the geodetic frequency response of each measurement point, taking a single measurement point as the unit. When the geodetic frequency response of the measurement point is stable, proceed to Step 6; otherwise, remove the data with the largest deviation in the geodetic frequency response of the measurement point, and return the remaining geodetic response data of different frequencies to Step 4.

[0014] Step 6: Calculate the arithmetic mean of the geodetic frequency response for each individual measuring point to obtain the final geodetic frequency response.

[0015] The current signal containing multiple frequency components described in step 1 should contain at least three or more transmission frequencies.

[0016] In step 1, the current signal containing multiple frequency components must simultaneously meet two prerequisites: firstly, the maximum transmission frequency must meet the near-field condition at all measuring points; secondly, the minimum transmission frequency must not exhibit a significant excitation polarization effect at any measuring point. If both conditions cannot be met simultaneously, under the same geological structure and monitoring device conditions, the received signal will vary with the transmission frequency, failing to meet the application prerequisites of this invention. Based on experience, the transmission frequency should generally be between 1 and 150 Hz, and should not be too high or too low.

[0017] Wherein: When processing the actual signal, the time series of the transmitted current in step 1 corresponds to the time series of the received signal in time, that is, the time series of the transmitted current in other time periods cannot be used, and the theoretical transmitted current waveform cannot be used to replace the actual transmitted current waveform.

[0018] Wherein: the received signal in step 1 can be either potential data or potential difference data.

[0019] Among them, the correlation detection described in step 2 includes methods such as direct spectral analysis and correlation spectral analysis.

[0020] When using direct spectrum analysis, the formula for calculating the ground frequency response is as follows:

[0021]

[0022] Among them, Z j (iw) represents the geodetic frequency response at the j-th frequency of a single measuring point, U j (t) is the time series of the received signal, I j (t) is the time series of the emission current, and F[*] represents the Fourier transform of “*”.

[0023] When using correlation spectral analysis, the formula for calculating the ground frequency response is as follows:

[0024]

[0025] Among them, Z j (iw) represents the earth frequency response at the j-th frequency of a single measuring point. The periods of the transmitted current and the received signal are cross-correlated. The periodic autocorrelation of the emission current is represented by F[*], which indicates the Fourier transform of “*”.

[0026] The period of the transmitted current and the received signal are cross-correlated. and the periodic autocorrelation of the emission current The calculation formula is as follows:

[0027]

[0028]

[0029] Among them U j (t) is the time series of the received signal, I j (t) is the time series of the emission current, N1 is the sampling length, and Δt is the sampling interval.

[0030] In step 3, the signal-to-noise ratio of the received signals at different frequencies at each measuring point is calculated using the following formula:

[0031]

[0032] Among them, V signal The spectrum of the transmission frequency is obtained by performing a Fourier transform on the full waveform data; V noise This is the spectrum of noise near the transmission frequency obtained by performing a Fourier transform on the full waveform data. The frequency band near the transmission frequency is a frequency band with a certain bandwidth centered on the transmission frequency.

[0033] Specifically, the bandwidth of the frequency band near the transmission frequency is determined based on the sampling frequency and sampling duration, and should include at least 10 frequency points.

[0034] Specifically, the signal-to-noise ratio (SNR) threshold for the data is set to at least 10 dB. If the calculated SNR is higher than this value, the data quality of the received signal at this frequency at the current measurement point is considered to be qualified.

[0035] Specifically: the minimum threshold for the amount of data at the measuring point in step 4 is half the number of frequencies contained in the transmitted current, rounded up; when half the number of frequencies contained in the transmitted current is less than 3, the minimum threshold for the amount of data at the measuring point should be set to 3. When the amount of data at the measuring point is greater than the minimum threshold, it is considered that the measuring point has sufficient data to participate in the averaging calculation.

[0036] Wherein: the stability of the earth frequency response described in step 5 is evaluated by the relative mean square error of the earth frequency response at the measuring point, and the calculation formula is as follows:

[0037]

[0038] Where, m s Z represents the relative mean square error of the geodetic system response at different frequencies at a single measuring point, where n is the amount of data from the measuring point involved in the calculation. j For the earth system response at a single measuring point at the j-th frequency, It represents the average value of the earth system response at n frequencies at a single measuring point.

[0039] The threshold for the relative mean square error of the earth's frequency response is generally set below 10%. The threshold for the relative mean square error should be consistent across all measuring points at the same substation.

[0040] Wherein: the final earth frequency response calculation formula described in step 6 is:

[0041]

[0042] Where Z is the earth frequency response of a single measuring point, Z i Let n be the earth frequency response at the i-th frequency of a single measuring point, and n be the amount of measuring point data involved in the calculation.

[0043] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0044] This invention employs the simultaneous transmission and reception of signals containing multiple frequencies, and utilizes frequency selection and multi-frequency averaging data processing methods to achieve better anti-interference capabilities, improve the quality of downhole DC electrical method monitoring data, and ensure the accuracy and real-time performance of monitoring results.

[0045] (1) The multi-frequency signals transmitted and received by this invention contain multiple frequency components, which are dispersed within a relatively wide frequency band. Under the same acquisition mode, more data can be extracted. Faced with the situation of a wide variety of downhole frequency converters, unknown interference frequency bands, and high interference intensity, frequency selection is used to eliminate data in strong interference frequency bands from measurement points at different locations. The specific methods include evaluating the signal-to-noise ratio of the received signals at each frequency and evaluating the stability of the received signals at different frequencies.

[0046] (2) The downhole DC electrical method monitoring targeted by this invention belongs to geometric depth sounding. Without considering the induced polarization effect, the ground response at different frequencies is basically the same. Therefore, after removing data affected by strong electromagnetic interference, the ground response at each frequency can be replaced by calculating the arithmetic mean of the ground response at each frequency. This calculation method replaces multiple acquisitions of a single frequency signal with the ground response at different frequencies of multi-frequency signals, making full use of all effective signals contained in a single full waveform acquisition. It can further improve the signal-to-noise ratio of effective data without consuming additional instrument memory or reducing work efficiency.

[0047] The beneficial effects of this invention cannot be achieved by transmitting and receiving signals containing only a single frequency because electromagnetic noise during downhole electrical resistivity monitoring is complex, extremely intense, and varies with time and space. It is impossible to avoid it in time and space, and it is difficult to predict its interference characteristics and types, let alone the interference frequency band. A single-frequency signal has a limited frequency component, resulting in a small amount of data that can be extracted from the receiving end's time series. This makes it impossible to guarantee that the transmission frequency remains in a weak interference band throughout the monitoring process, and it is frequently affected by strong electromagnetic interference. Conventional methods for improving the signal-to-noise ratio of DC electrical resistivity are limited by downhole monitoring conditions and cannot achieve ideal results. The frequency selection and multi-frequency averaging methods of this invention both rely on transmitting and receiving multi-frequency signals and cannot be applied to single-frequency signals.

[0048] The beneficial effect of the frequency selection step in this invention differs from the data quality evaluation when using a single-frequency signal in that: when using a single-frequency signal, only one potential (difference) can be extracted from a single measurement point. After data quality evaluation, if it is unqualified, it will be deleted, resulting in sparse measurement point density and reduced system resolution. When using a multi-frequency signal, multiple potentials (differences) of different frequencies can be extracted from a single measurement point. The potentials (differences) of different frequencies are located in electromagnetic interference frequency bands of different intensities. After frequency selection, the influence of strong interference frequency bands can be eliminated, and the effective information of the measurement points can be retained to the maximum extent, ensuring the measurement point density and the system resolution.

[0049] The beneficial effects of the multi-frequency averaging step in this invention cannot be achieved by using the averaging method after multiple measurements for single-frequency signals. This is because conventional averaging after multiple measurements is for single-frequency signals. Applying it to multi-frequency signals requires averaging different frequency components separately, necessitating multiple measurements during the construction phase to meet the application requirements, and a certain number of measurements are needed to achieve good results. Since downhole DC electrical resistivity monitoring uses rolling data acquisition, multiple measurements not only significantly reduce construction efficiency and fail to provide timely feedback on changes in the monitored object over time, but also place higher demands on the instrument's data storage capabilities. Therefore, the averaging method after multiple measurements for single-frequency signals is unsuitable for downhole DC electrical resistivity monitoring. In contrast, the multi-frequency averaging method, while meeting the application requirements, fully utilizes all effective signals contained in a single full-waveform acquisition, requiring only a single measurement. This not only fully leverages the anti-interference advantages of multi-frequency signals but also aligns with the automated and uninterrupted data acquisition characteristics of downhole DC electrical resistivity monitoring.

[0050] In summary, the multi-frequency data processing technology in this invention is compatible with downhole DC electrical monitoring, effectively leveraging its anti-interference capabilities and improving the effective data signal-to-noise ratio. Attached Figure Description

[0051] Figure 1 This is a flowchart of the multi-frequency data anti-interference method of the present invention;

[0052] Figure 2 Examples include (a) the theoretical multi-frequency signal waveform, (b) the actual noise signal waveform, and (c) the superimposed waveform of the two.

[0053] Figure 3 It is the ground frequency response (a) and the Fourier transform of the received signal at a single measuring point when using multi-frequency signals at different transmission frequencies (b).

[0054] Figure 4 It is the geodetic frequency response of each measuring point after multi-frequency averaging.

[0055] Figure 5 It is the earth frequency response at each measuring point when using a 4Hz single-frequency signal. Detailed Implementation

[0056] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments.

[0057] Please see Figure 1 A multi-frequency data processing technology for downhole DC electrical resistivity monitoring includes the following steps:

[0058] Step 1: The transmitter supplies a current signal containing multiple frequency components (hereinafter referred to as multi-frequency signal) to the ground through the transmitting electrode and records the time series of the transmitted current. At the same time, it records the time series of the electromagnetic field response signal through the receiving electrode as the time series of the received signal. Both the time series of the transmitted current and the time series of the received signal proceed to step 2.

[0059] In this embodiment, the time series of the transmission current includes five transmission frequencies: 4Hz, 8Hz, 16Hz, 32Hz, and 64Hz.

[0060] Step 2: Detect the time series of the transmitted current and the time series of the received signal to obtain the ground frequency response of each measuring point at each frequency. The time series of the received signal (time domain data) and the ground frequency response (frequency domain data) of the measuring point are both entered into Step 3.

[0061] In this embodiment, correlation spectrum analysis is used to calculate the geosystem response at each measurement point at each frequency. The calculation formula is as follows:

[0062]

[0063] Among them, Z j (iw) represents the earth system response at the j-th frequency of a single measuring point. The periods of the transmitted current and the received signal are cross-correlated. For the periodic cross-correlation of the emission current, F[*] denotes the Fourier transform of “*”.

[0064] The periods of the transmitted current and the received signal are cross-correlated. and the periodic autocorrelation of the emission current The calculation formula is as follows:

[0065]

[0066]

[0067] U j (t) is the time series of the received signal, I j (t) is the time series of the emission current, N1 is the sampling length, and Δt is the sampling interval.

[0068] Step 3: Using a single measurement point as a unit, calculate the signal-to-noise ratio (SNR) of the received signal at different frequencies based on the Fourier transform results of the received signal time series, evaluate the degree of noise influence on the received signal at different frequencies, remove unqualified data, and proceed to Step 4 for qualified data.

[0069] Specifically, the signal-to-noise ratio (SNR) of the received signal at each measurement point at different frequencies is calculated using the following formula:

[0070]

[0071] Among them, V signal The spectrum of the transmission frequency is obtained by performing a Fourier transform on the full waveform data; V noise This is the spectrum of noise near the transmission frequency obtained by performing a Fourier transform on the full waveform data. The frequency band near the transmission frequency is a frequency band with a certain bandwidth centered on the transmission frequency.

[0072] In this embodiment, the bandwidth of the frequency band near the transmission frequency is set to 5Hz; the signal-to-noise ratio threshold of the data is set to 10dB, and data quality is considered to be qualified if it is higher than this value.

[0073] Step 4: Using a single measuring point as the unit and the earth response at different frequencies as the data source, determine whether the amount of data at each measuring point is sufficient. If the measuring point has sufficient data to participate in the averaging calculation, proceed to Step 5; otherwise, proceed to Step 6.

[0074] In this embodiment, the emission current contains 5 frequencies, and a minimum of 3 data points are required for each measurement point to participate in the calculation in step 5.

[0075] Step 5: Evaluate the stability of the geodetic frequency response of each measurement point, taking a single measurement point as the unit. When the geodetic frequency response of the measurement point is stable, proceed to Step 6; otherwise, remove the data with the largest deviation in the geodetic frequency response of the measurement point, and return the remaining geodetic response data of different frequencies to Step 4.

[0076] In this embodiment, the threshold for relative mean square error is set to 5%.

[0077] Step 6: Calculate the arithmetic mean of the geodetic frequency response for each individual measuring point to obtain the final geodetic frequency response.

[0078] The principle of this embodiment will be introduced below.

[0079] When the transmitter emits a current signal, the transmission frequency should not be too high or too low. If the transmission frequency is too high, the received signal strength will be frequency-dependent; if the transmission frequency is too low, excitation polarization may occur, neither of which meets the application prerequisites of this invention and makes averaging of multiple ground frequency responses impossible. Correlation detection is performed on the time series of the emitted current and the time series of the received signal to obtain the ground frequency response at the measuring point. Essentially, this is the ground frequency response generated by a unit current, eliminating the influence of different proportions of the frequency components in the emitted current. During downhole DC electrical method monitoring, electromagnetic noise is complex, extremely intense, and varies with time and space. Therefore, the signal-to-noise ratio (SNR) at different frequencies at a single measuring point differs, necessitating the screening of data used in the calculation. This embodiment evaluates and eliminates unqualified data by using the SNR of the received signals at different frequencies and the relative mean square error of the ground response, while also considering the amount of data required for the multi-frequency averaging step, ensuring the accuracy of the results.

[0080] To demonstrate the feasibility and effectiveness of this invention, the following experiment was designed to illustrate the method of this invention:

[0081] In this experiment, the time series of the transmitted current used in step 1 is the theoretical waveform of the multi-frequency signal. The time series of the transmitted current includes five transmission frequencies: 4Hz, 8Hz, 16Hz, 32Hz, and 64Hz. The time series of the received signal is the signal to be processed constructed by superimposing the theoretical waveform of the multi-frequency signal and the measured downhole electrical resistivity tomography (EDT) monitoring time series. The sampling frequency and sampling duration of the theoretical waveform and the actual time series are consistent. The remaining steps are implemented using the settings in the embodiments of this invention. This experiment uses the method of superimposing theoretical multi-frequency signals and measured downhole EDT monitoring data to construct the signal to be processed, which is beneficial for demonstrating the effectiveness of the multi-frequency averaging data processing method in this invention, while also taking into account the actual situation of electromagnetic interference during downhole monitoring.

[0082] Figure 2 Examples include a theoretical multi-frequency signal waveform (a), an actual noise signal waveform (b), and a superposition of the two (c). The theoretical multi-frequency signal comprises five frequency components: 4Hz, 8Hz, 16Hz, 32Hz, and 64Hz, with a total amplitude of 20μV, serving as the effective signal. The actual noise signal waveform is a measured time series from downhole DC electrical method monitoring, serving as the interference signal. This embodiment uses the superposition of the effective signal and the interference signal to create a time series of the received signal for subsequent processing. The theoretical multi-frequency signal waveform also serves as a time series of the transmitted current, with a total amplitude of 20μA. The theoretical earth system response value of the constructed signal to be processed is 1Ω. This experiment assesses the feasibility of the invention by determining whether a more accurate earth system response can be obtained through processing.

[0083] Figure 3This section presents the earth frequency response (a) and the Fourier transform (b) of the received signal at a single measuring point for multi-frequency signals at different transmission frequencies. From the Fourier transform curve (a) of this measuring point, it can be seen that in addition to the five transmission frequencies of 4Hz, 8Hz, 16Hz, 32Hz, and 64Hz, external electromagnetic interference also significantly interferes with the power frequency (50Hz) and low-frequency band (0-4Hz) of the measuring point. Calculations show that the signal-to-noise ratio (SNR) of 4Hz is approximately 8.8dB, lower than the set minimum effective value; therefore, the 4Hz data for this measuring point is discarded. From the earth electromagnetic response (b) of the transmission frequencies at this measuring point, it can be seen that the relative mean square error (RMSE) of the earth system response at 8Hz, 16Hz, 32Hz, and 64Hz is 12.47%. The earth system response at 16Hz is significantly higher than the other three frequencies and should be discarded. After discarding, the relative MSE of the earth system response at the remaining three frequencies is 3.12%, meeting the stability requirements, and the data proceeds to step 6.

[0084] Figure 4 This is the geodetic frequency response at each measuring point after multi-frequency averaging. As can be seen from the figure, for the data constructed in this experiment, after processing by the method of this invention, the geodetic system response fluctuates slightly around 1Ω, with an average value of 0.992Ω at each measuring point, which is close to the theoretical value of the geodetic system response of this data. The relative mean square error of each measuring point is 1.87%, which shows high accuracy.

[0085] In addition, the present invention includes the following comparative experiments for illustration.

[0086] First, the signal to be processed is constructed. A theoretical 4Hz single-frequency signal waveform with an amplitude of 20μV is used as the effective signal, and the measured time series from the downhole DC electrical method used in this example is used as the interference signal. The effective signal and the interference signal are superimposed to form the time series of the received signal for subsequent processing in the comparative experiment. The theoretical multi-frequency signal waveform is also used as the transmitted current waveform, with a total amplitude of 20μA. At this point, the earth system response of the constructed signal to be processed is 1Ω. Then, the earth frequency response at each measuring point is calculated using correlation spectrum analysis. Figure 5 The earth frequency response at each measuring point is calculated using a 4Hz single-frequency signal. The above results are compared with the multi-frequency averaging data processing results from the example. Figure 4 For comparison, Figure 5 The average earth frequency response at each measuring point was 0.978Ω, indicating lower accuracy. The relative mean square error at each measuring point was 4.80%, showing not only a larger error but also more measuring points with significant errors, making it more susceptible to strong electromagnetic interference. These results demonstrate that this invention can fully leverage the anti-interference advantages of multi-frequency signals, achieving superior data quality compared to single-frequency signals under the same total amplitude of the transmitted current and the same received signal acquisition parameters, making it suitable for downhole DC electrical resistivity monitoring.

Claims

1. A method for resisting interference from multi-frequency data in downhole DC electrical monitoring, characterized in that, Specifically, the following steps are included: Step 1: The transmitter supplies a current signal containing multiple frequency components to the ground through the transmitting electrode and records the time series of the transmitted current. At the same time, it records the time series of the electromagnetic field response signal through the receiving electrode as the time series of the received signal. Both the time series of the transmitted current and the time series of the received signal proceed to step 2. Step 2: Perform correlation detection on the time series of the transmitted current and the time series of the received signal to obtain the ground frequency response of each measuring point at each frequency. The time series of the received signal and the ground frequency response of the measuring point are both proceeded to Step 3. The relevant detection employs either direct spectral analysis or correlation spectral analysis; wherein: When using direct spectrum analysis, the formula for calculating the ground frequency response is as follows: in, For a single measuring point j The ground frequency response at each frequency. U j ( t () is the time series of the received signal. I j ( t () is a time series of the emission current. F [ [Indicates "] Perform a Fourier transform; When using correlation spectral analysis, the formula for calculating the ground frequency response is as follows: in, For a single measuring point j The ground frequency response at each frequency. The periods of the transmitted current and the received signal are cross-correlated. For the periodic autocorrelation of the emission current, F [ [Indicates "] Perform a Fourier transform; in, U j ( t () is the time series of the received signal. I j ( t () is a time series of the emission current. N 1 represents the sampling length, Δ t The sampling interval; Step 3: Using a single measurement point as a unit, calculate the signal-to-noise ratio (SNR) of the received signal at different frequencies based on the Fourier transform results of the time series of the received signal. SNR The degree to which the received signals at different frequencies are affected by noise is evaluated, and unqualified ground frequency responses are eliminated. Qualified ground frequency response data proceeds to step 4. Step 4: Using a single measuring point as the unit and the earth response at different frequencies as the data source, determine whether the amount of data at each measuring point is sufficient. If the measuring point has sufficient data to participate in the averaging calculation, proceed to Step 5; otherwise, proceed to Step 6. Step 5: Evaluate the stability of the geodetic frequency response of each measurement point, taking a single measurement point as the unit. When the geodetic frequency response of the measurement point is stable, proceed to Step 6; otherwise, remove the data with the largest deviation in the geodetic frequency response of the measurement point, and return the remaining geodetic frequency response data of different frequencies to Step 4. The stability of the earth frequency response is evaluated by the relative mean square error of the earth frequency response at the measuring point, and the calculation formula is as follows: in, m s This represents the relative mean square error of the earth system response at different frequencies at a single measuring point. n The amount of measurement data used in the calculation. Z j For a single measuring point at the first j Earth system response at various frequencies For a single measuring point n The average value of the earth system response at each frequency; Step 6: Calculate the arithmetic mean of the geodetic frequency response for each individual measuring point to obtain the final geodetic frequency response.

2. The multi-frequency data anti-interference method for downhole DC electrical monitoring as described in claim 1, characterized in that, In step 1, the current signal includes at least three transmission frequencies.

3. The multi-frequency data anti-interference method for downhole DC electrical monitoring as described in claim 1, characterized in that, In step 1, the current signal containing multiple frequency components simultaneously satisfies two prerequisites: (1) the maximum transmission frequency satisfies the near-zone condition at all measurement points; and (2) the minimum transmission frequency does not exhibit obvious excitation polarization effect at all measurement points.

4. The multi-frequency data anti-interference method for downhole DC electrical monitoring as described in claim 1, characterized in that, In step 1, the frequency of the transmitting current is between 1 and 150 Hz.

5. The multi-frequency data anti-interference method for downhole DC electrical monitoring as described in claim 1, characterized in that, In step 1, the received signal is potential data or potential difference data.

6. The multi-frequency data anti-interference method for downhole DC electrical monitoring as described in claim 1, characterized in that, In step 3, the formula for calculating the signal-to-noise ratio is as follows: in, V signal The spectrum of the transmission frequency obtained by performing a Fourier transform on the full waveform data; V noise This refers to the spectrum of noise near the transmission frequency obtained by performing a Fourier transform on the full waveform data. The frequency band near the transmission frequency is a frequency band with a certain bandwidth centered on the transmission frequency.

7. The multi-frequency data anti-interference method for downhole DC electrical monitoring as described in claim 1, characterized in that, In step 4, the minimum threshold for the amount of data at the measurement point is half the number of frequencies contained in the transmission current, rounded up, and not less than 3.

8. The multi-frequency data anti-interference method for downhole DC electrical monitoring as described in claim 1, characterized in that, In step 6, the formula for calculating the ground frequency response is: in, Z The earth frequency response at a single measuring point. Z i For a single measuring point i The ground frequency response at each frequency. n The amount of measurement data used in the calculation.

Images