Weak signal extraction method for three-dimensional excitation data in wells

Abstract

The present invention provides a weak signal extraction method for three-dimensional excitation data in a well, and the weak signal extraction method for three-dimensional excitation data in a well comprises: step 1: collecting seismic data; step 2: selecting a seismic trace near a well site; step 3: selecting the seismic trace as a reference seismic trace; step 4: selecting a trace of collected seismic data; step 5: performing cross-correlation processing on the data and the reference seismic trace; step 6: processing all data according to step 5; step 7: performing time shift processing; step 8: outputting a result profile. The weak signal extraction method for three-dimensional excitation data in a well selects a suitable reference signal and utilizes the correlation principle to extract weak signals, thereby realizing weak signal extraction and improving the signal-to-noise ratio of data.

Description

Technical Field

[0001] The invention relates to the technical field of oilfield development, and in particular to a method for extracting weak signals from three-dimensional excitation data in a well. Background Art

[0002] 3D excitation in a well means to excite signals in the well and arrange detectors on the ground to receive the signals, which is different from conventional vertical seismic profile exploration. The data of 3D excitation in a well usually has a low signal-to-noise ratio due to weak signals and interference.

[0003] There are generally related processing methods for extracting weak signals. The related method is to use a known signal as a reference signal and perform correlation processing with the data to be detected containing the reference signal. The cross-correlation function describes the degree of correlation between random signals, which can extract effective signals and improve the signal-to-noise ratio of the data. In the document "Ground Penetrating Radar Signal Wavelet Extraction and Cross-Correlation Filtering", based on the principle that the reflection coefficient is generally white noise, the wavelet is extracted using seismic records, and the ground penetrating radar signal is subjected to cross-correlation filtering with the extracted wavelet to obtain the filtered profile. Since the reflection coefficient of the stratum does not meet the assumption of white noise for the seismic waves received by the ground in the three-dimensional excitation of the well, the wavelet extracted by this method in the cross-correlation filtering of the three-dimensional excitation data in the well is no longer a seismic wavelet, and the cross-correlation filtering method is not ideal. In the document "Application of Cross-Correlation Filtering in Railway Karst Exploration", the HILbert transform method is used to obtain the seismic wavelet using the amplitude spectrum of the seismic record, and the abnormal body in the profile is obtained by cross-correlation filtering with the seismic record. The problem is that the reflection coefficient needs to be assumed to be white noise when extracting the seismic wavelet. In the document “Ground Penetrating Radar Signal Wavelet Extraction and Cross-Correlation Filtering”, wavelets extracted from seismic records are used to perform cross-correlation filtering on synthetic records containing random noise to suppress random interference in the profile and improve the data signal-to-noise ratio. The problem is that when extracting seismic wavelets, it is necessary to assume that the reflection coefficient is white noise.

[0004] The weak signal extraction method of three-dimensional excitation data in the well is generally to select a channel with a high signal-to-noise ratio as a reference signal and perform correlation processing with the data, which usually has unsatisfactory results.

[0005] In the Chinese patent application with application number: CN201811337420.3, a method for extracting formation acoustic wave velocity in a cased well is involved, including: performing cased acoustic logging in a depth interval to obtain full-wave array data of cased acoustic waves; designing a Gaussian bandpass filter; filtering to obtain a two-dimensional array waveform about time and frequency; coherent processing to obtain a frequency-slowness coherent spectrum; calculating the formation acoustic wave slowness; repeating the steps until the entire depth interval is processed, and obtaining a formation acoustic wave time difference curve within the processed depth interval.

[0006] In the Chinese patent application with application number: CN201210385721.X, a method for extracting the slowness of mode waves based on acoustic logging data is involved. The method includes performing the following steps for each depth point of at least one depth point in the formation: when the transmitter transmits, the receiver receives the original waveform set corresponding to the transmitter; the original waveform set is signal preprocessed to improve the similarity and signal-to-noise ratio of the original waveform set; the preprocessed original waveform set is respectively estimated to obtain the corresponding time-slowness-correlation three-dimensional matrix; the peak points on the correlation dimension in the corresponding time-slowness-correlation three-dimensional matrix are identified; and the type and slowness of the mode wave corresponding to each depth point are determined according to the size relationship between the identified peak points and the wellbore compensation relationship.

[0007] In the Chinese patent application with application number: CN201510170928.9, a real-time inversion and positioning method for underground microseismic events based on grid search is involved. The real-time inversion and positioning method for underground microseismic events based on grid search includes: establishing a velocity model in the space containing the microseismic source under the wellhead; picking the first arrival time of P waves and S waves for microseismic events; calculating the polarization degree and polarization direction of the first arrival of the P wave of each record of the microseismic event; establishing an inversion objective function, and forward-modeling the wave field to calculate the theoretical first arrival travel time from the grid point to the observation system; calculating the objective function value to determine the source position; and calculating the three-dimensional spatial position coordinates of the source.

[0008] The above existing technologies are significantly different from the present invention and fail to solve the technical problem we want to solve. Therefore, how to select a suitable reference signal and use relevant methods to detect weak signals is an inevitable requirement for extracting weak signals. For this reason, we have invented a new weak signal extraction method for three-dimensional excitation data in the well. Summary of the invention

[0009] The purpose of the present invention is to provide a method for extracting weak signals from three-dimensional excitation data in a well by selecting a suitable reference signal and utilizing a correlation principle to extract weak signals.

[0010] The purpose of the present invention can be achieved by the following technical measures: a method for extracting weak signals from three-dimensional excitation data in a well, the method comprising:

[0011] Step 1: Collect seismic data;

[0012] Step 2: Select the seismic trace near the well site;

[0013] Step 3: Select the seismic trace as the reference seismic trace;

[0014] Step 4: Select a channel of the collected seismic data;

[0015] Step 5: Perform cross-correlation processing on the data and the reference seismic trace;

[0016] Step 6: Process all data according to step 5;

[0017] Step 7: Perform time shift processing;

[0018] Step 8: Output the result profile.

[0019] The purpose of the present invention can also be achieved by the following technical measures:

[0020] In step 1, the collected seismic data is a seismic signal with certain morphological regular characteristics.

[0021] In step 2, the seismic trace near the well site is selected to be closest to the wellhead. The seismic trace is excited in the well, with the source located at a certain depth underground, receiving direct waves and random interference, and not receiving formation reflection waves.

[0022] In step 3, the seismic trace in step 2 is selected as a reference seismic trace. The reference seismic trace is a seismic trace that can be used as a seismic wavelet. A seismic wavelet refers to a seismic wavelet whose convolution with the reflection coefficient plus noise equals the seismic record according to the seismic convolution theory.

[0023] In step 4, the data channel refers to the seismic channel which is excited in the well, the source is located at a certain depth in the well, receives direct waves and random interference, and does not receive formation reflection waves.

[0024] In step 5, the data in step 4 are cross-correlated with the reference seismic trace. The cross-correlation formula is:

[0025]

[0026] n=1,2,3...,N

[0027] x(n) is the nth amplitude value of the signal, y(n) is the nth amplitude value of the scanned signal, N is the number of sampling points, r xy (m) is the mth amplitude value of the cross-correlation signal;

[0028] The specific steps are to set the signal received by the wellhead detector as:

[0029] x=(0,0,a)

[0030] Assume that the signal received by non-wellhead geophone 1 is:

[0031] x1=(0,0,0,0,a,0,0,a)

[0032] Assume that the signal received by non-wellhead geophone 2 is:

[0033] x2=(0,0,0,0,0,0,a,0,0,0,0,a)

[0034] a The amplitude value of the seismic wavelet;

[0035] x1 and x can be calculated using the previous related formula:

[0036] r xy (-2) = x1(1) x(3)

[0037] =0

[0038] r xy (-1)=x1(1)·x(2)+x1(2)·x(3)

[0039] =0

[0040] r xy (0)=x1(1)·x(1)+x1(2)·x(2)+x1(3)·x(3)

[0041] =0

[0042] r xy (1)=x1(2)·x(1)+x1(3)·x(2)+x1(4)·x(3)

[0043] =0

[0044] r xy (2)=x1(3)·x(1)+x1(4)·x(2)+x1(5)·x(3)

[0045] =a·a

[0046] r xy (3)=x1(4)·x(1)+x1(5)·x(2)+x1(6)·x(3)

[0047] =0

[0048] r xy (4)=x1(5)·x(1)+x1(6)·x(2)+x1(7)·x(3)

[0049] =0

[0050] r xy (5)=x1(6)·x(1)+x1(7)·x(2)+x1(8)·x(3)

[0051] =a·a

[0052] r xy(6)=x1(7)·x(1)+x1(8)·x(2)+x1(9)·x(3)

[0053] =0

[0054] r xy (7)=x1(8)·x(1)+x1(9)·x(2)+x1(10)·x(3)

[0055] =0

[0056] r xy =(0,0,0,0,a 2 ,0,0,a 2 ,0,0)

[0057] x2 and x can be calculated using the previous correlation formula:

[0058] r xy (-2) = x2(1) x(3)

[0059] =0

[0060] r xy (-1)=x2(1)·x(2)+x2(2)·x(3)

[0061] =0

[0062] r xy (0)=x2(1)·x(1)+x2(2)·x(2)+x2(3)·x(3)

[0063] =0

[0064] r xy (1)=x2(2)·x(1)+x2(3)·x(2)+x2(4)·x(3)

[0065] =0

[0066] r xy (2)=x2(3)·x(1)+x2(4)·x(2)+x2(5)·x(3)

[0067] =0

[0068] r xy (3)=x2(4)·x(1)+x2(5)·x(2)+x2(6)·x(3)

[0069] =0

[0070] r xy (4)=x2(5)·x(1)+x2(6)·x(2)+x2(7)·x(3)

[0071] =a·a

[0072] r xy (5)=x2(6)·x(1)+x2(7)·x(2)+x2(8)·x(3)

[0073] =0

[0074] r xy (6)=x2(7)·x(1)+x2(8)·x(2)+x2(9)·x(3)

[0075] =0

[0076] r xy (7)=x2(8)·x(1)+x2(9)·x(2)+x2(10)·x(3)

[0077] =0

[0078] r xy (8)=x2(9)·x(1)+x2(10)·x(2)+x2(11)·x(3)

[0079] =0

[0080] r xy (9)=x2(10)·x(1)+x2(11)·x(2)+x2(12)·x(3)

[0081] =a·a

[0082] r xy (10)=x2(11)·x(1)+x2(12)·x(2)+x2(13)·x(3)=0

[0083] r xy (11)=x2(12)·x(1)+x2(13)·x(2)+x2(14)·x(3)=0

[0084] r xy =(0,0,0,0,0,0,a 2 ,0,0,0,0,a 2 ,0,0)

[0085] The detection signal is obtained by correlation processing.

[0086] In step 5, when the received signal has random interference, then:

[0087] x=(0,0,a)+y1

[0088] Assume that the signal received by non-wellhead geophone 1 is:

[0089] x1=(0,0,0,0,a,0,0,a)+y2

[0090] Assume that the signal received by non-wellhead geophone 2 is:

[0091] x2=(0,0,0,0,0,0,a,0,0,0,0,a)+y3

[0092] y1,y2,y3 are the amplitude values ​​of the random signals.

[0093] x1 and x are calculated using the previous correlation formula. Due to random interference, the correlation result is 0, which can be obtained:

[0094] r xy (-2) = x1(1) x(3)

[0095] =0

[0096] r xy (-1)=x1(1)·x(2)+x1(2)·x(3)

[0097] =0

[0098] r xy (0)=x1(1)·x(1)+x1(2)·x(2)+x1(3)·x(3)

[0099] =0

[0100] r xy (1)=x1(2)·x(1)+x1(3)·x(2)+x1(4)·x(3)

[0101] =0

[0102] r xy (2)=x1(3)·x(1)+x1(4)·x(2)+x1(5)·x(3)

[0103] =a·a

[0104] r xy (3)=x1(4)·x(1)+x1(5)·x(2)+x1(6)·x(3)

[0105] =0

[0106] r xy (4)=x1(5)·x(1)+x1(6)·x(2)+x1(7)·x(3)

[0107] =0

[0108] r xy (5)=x1(6)·x(1)+x1(7)·x(2)+x1(8)·x(3)

[0109] =a·a

[0110] r xy (6)=x1(7)·x(1)+x1(8)·x(2)+x1(9)·x(3)

[0111] =0

[0112] r xy (7)=x1(8)·x(1)+x1(9)·x(2)+x1(10)·x(3)

[0113] =0

[0114] r xy =(0,0,0,0,a 2 ,0,0,a 2 ,0,0)

[0115] x2 and x are calculated using the previous correlation formula. Due to random interference, the correlation result is 0, which can be obtained:

[0116] r xy (-2) = x2(1) x(3)

[0117] =0

[0118] r xy (-1)=x2(1)·x(2)+x2(2)·x(3)

[0119] =0

[0120] r xy (0)=x2(1)·x(1)+x2(2)·x(2)+x2(3)·x(3)

[0121] =0

[0122] r xy (1)=x2(2)·x(1)+x2(3)·x(2)+x2(4)·x(3)

[0123] =0

[0124] r xy (2)=x2(3)·x(1)+x2(4)·x(2)+x2(5)·x(3)

[0125] =0

[0126] r xy (3)=x2(4)·x(1)+x2(5)·x(2)+x2(6)·x(3)

[0127] =0

[0128] r xy(4)=x2(5)·x(1)+x2(6)·x(2)+x2(7)·x(3)

[0129] =a·a

[0130] r xy (5)=x2(6)·x(1)+x2(7)·x(2)+x2(8)·x(3)

[0131] =0

[0132] r xy (6)=x2(7)·x(1)+x2(8)·x(2)+x2(9)·x(3)

[0133] =0

[0134] r xy (7)=x2(8)·x(1)+x2(9)·x(2)+x2(10)·x(3)

[0135] =0

[0136] r xy (8)=x2(9)·x(1)+x2(10)·x(2)+x2(11)·x(3)

[0137] =0

[0138] r xy (9)=x2(10)·x(1)+x2(11)·x(2)+x2(12)·x(3)

[0139] =a·a

[0140] r xy (10)=x2(11)·x(1)+x2(12)·x(2)+x2(13)·x(3)=0

[0141] r xy (11)=x2(12)·x(1)+x2(13)·x(2)+x2(14)·x(3)=0

[0142] r xy =(0,0,0,0,0,0,a 2 ,0,0,0,0,a 2 ,0,0)

[0143] Correlation processing can also obtain detection signals when there is random interference.

[0144] In step 6, all data refer to seismic traces excited in the well, with the source located at a certain depth in the well, receiving direct waves and random interference, and not receiving formation reflection waves.

[0145] In step 7, time shift processing refers to calculating the direct wave time t of the first track, and shifting the correlated data backward by t to obtain the seismic record after cross-correlation filtering; the first track refers to the seismic track closest to the reference track.

[0146] The weak signal extraction method of three-dimensional excitation data in the well in the present invention selects the seismic trace near the well as the reference signal, avoids the need to assume that the seismic reflection coefficient is white noise and use the seismic trace to extract the seismic wavelet, solves the problem that the seismic trace cannot be used to extract the seismic wavelet when the reflection coefficient does not meet the white noise, and the problem that a suitable reference signal cannot be selected. The present invention utilizes the fact that seismic waves propagate in the mud in the well without reflection, and there is no reflected wave but only a direct wave, selects the seismic trace near the well as the reference signal, selects the suitable reference signal and uses the relevant principle to extract the weak signal, which can realize the extraction of weak signals and improve the signal-to-noise ratio of the data. BRIEF DESCRIPTION OF THE DRAWINGS

[0147] Figure 1 It is a flow chart of a weak signal extraction method of three-dimensional excitation data in a well in a specific embodiment 1 of the present invention;

[0148] Figure 2 It is a schematic diagram of data that is not processed by the present invention in a specific embodiment 1 of the present invention;

[0149] Figure 3 This is a schematic diagram of data processed by the present invention in a specific embodiment 1 of the present invention;

[0150] Figure 4 It is a flow chart of a weak signal extraction method of three-dimensional excitation data in a well in a specific embodiment 2 of the present invention;

[0151] Figure 5 A schematic diagram of data not processed by the present invention in specific embodiments 2 and 3 of the present invention;

[0152] Figure 6 It is a schematic diagram of data processed by the present invention in a specific embodiment 2 of the present invention;

[0153] Figure 7 It is a flow chart of a weak signal extraction method of three-dimensional excitation data in a well in a specific embodiment 3 of the present invention;

[0154] Figure 8 It is a schematic diagram of data processed by the present invention in a specific embodiment 3 of the present invention. DETAILED DESCRIPTION

[0155] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

[0156] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprising" and / or "including" are used in this specification, it indicates the presence of features, steps, operations and / or combinations thereof.

[0157] The weak signal extraction method of the three-dimensional excitation data in the well of the present invention comprises the following steps:

[0158] Step 1: Input data; the input data is a seismic signal with certain morphological characteristics.

[0159] Step 2: Select the seismic trace near the well site; Select the seismic trace near the well site, which is the closest to the wellhead. The seismic trace selected that is closest to the wellhead refers to the seismic trace that receives direct waves and random interference, but does not receive formation reflection waves. This seismic trace refers to the excitation in the well, where the source is located at a certain depth in the well. When the excited seismic wave propagates in the well, since the medium in the well is mud, which is a uniform medium, according to the principle of seismic reflection, when there is a difference in wave impedance in the underground medium, reflection and transmission will occur. The amplitude of reflection and transmission is related to the reflection and transmission coefficient of the medium. When there is no difference in wave impedance in the underground medium, that is, when it is a uniform medium, no reflection will occur, only transmission. The amplitude of transmission is related to the transmission coefficient of the medium. Therefore, seismic waves propagate in the well without reflection, only transmission. The detector at the wellhead can only receive the direct wave signal and interference signal of the source, but not the reflection signal. When the seismic wave propagates to the formation around the well, according to seismic theory, the medium can be approximated as a layered medium. The layered medium is an inhomogeneous medium with a difference in wave impedance. Therefore, seismic waves propagate in the formation around the well and reflection and transmission will occur. According to Snell's law, the detector at the wellhead cannot receive the reflected wave, while the detector not at the wellhead can receive the seismic trace received by the wellhead detector of the reflected wave and the direct wave.

[0160] Step 3: Select the seismic trace as the reference seismic trace; select the seismic trace in step 2 as the reference seismic trace. The reference seismic trace is a seismic trace that can be used as a seismic wavelet. A seismic wavelet refers to a seismic wavelet whose convolution with the reflection coefficient plus the noise equals the seismic record according to the seismic convolution theory.

[0161] Step 4: Select one of the input data; Select one of the input data. The data track refers to the excitation in the well, the source is located at a certain depth in the well, and the excited seismic waves propagate in the well. Since the medium in the well is mud, which is a uniform medium, according to the principle of seismic reflection, when there is a difference in wave impedance in the underground medium, reflection and transmission will occur. The amplitude of reflection and transmission is related to the reflection and transmission coefficient of the medium. When there is no difference in wave impedance in the underground medium, that is, it is a uniform medium, there will be no reflection, only transmission, and the amplitude of transmission is related to the transmission coefficient of the medium. Therefore, seismic waves propagate in the well without reflection, only transmission. The detector at the wellhead can only receive the direct wave signal and interference signal of the source, but not the reflected signal. When the seismic wave propagates to the formation around the well, according to the seismic theory, the medium can be approximated as a layered medium. The layered medium is an inhomogeneous medium with a difference in wave impedance. Therefore, seismic waves propagate in the formation around the well and reflection and transmission will occur. According to Snell's law, the detector at the wellhead cannot receive the reflected wave, but the detector not at the wellhead can receive the seismic trace of the reflected wave and the direct wave received by the detector not at the wellhead.

[0162] Step 5: The data is cross-correlated with the reference seismic trace; the data in step 4 is cross-correlated with the reference seismic trace, and the cross-correlation formula is:

[0163]

[0164] n=1,2,3...,N

[0165] x(n) is the nth amplitude value of the signal, y(n) is the nth amplitude value of the scan signal, and N is the number of sampling points. The specific steps are to assume that the signal received by the wellhead detector is:

[0166] x=(0,0,a)

[0167] Assume that the signal received by non-wellhead geophone 1 is:

[0168] x1=(0,0,0,0,a,0,0,a)

[0169] Assume that the signal received by non-wellhead geophone 2 is:

[0170] x2=(0,0,0,0,0,0,a,0,0,0,0,a)

[0171] x1 and x can be calculated using the previous related formula:

[0172] r xy (-2) = x1(1) x(3)

[0173] =0

[0174] r xy(-1)=x1(1) x(2)+x1(2) x(3)

[0175] =0

[0176] r xy (0)=x1(1) x(1)+x1(2) x(2)+x1(3) x(3)

[0177] =0

[0178] r xy (1)=x1(2) x(1)+x1(3) x(2)+x1(4) x(3)

[0179] =0

[0180] r xy (2)=x1(3) x(1)+x1(4) x(2)+x1(5) x(3)

[0181] =a·a

[0182] r xy (3)=x1(4) x(1)+x1(5) x(2)+x1(6) x(3)

[0183] =0

[0184] r xy (4)=x1(5) x(1)+x1(6) x(2)+x1(7) x(3)

[0185] =0

[0186] r xy (5)=x1(6) x(1)+x1(7) x(2)+x1(8) x(3)

[0187] =a·a

[0188] r xy (6)=x1(7) x(1)+x1(8) x(2)+x1(9) x(3)

[0189] =0

[0190] r xy (7)=x1(8) x(1)+x1(9) x(2)+x1(10) x(3)

[0191] =0

[0192] r xy =(0,0,0,0,a 2 ,0,0,a 2 ,0,0)

[0193] x2 and x can be calculated using the previous correlation formula:

[0194] r xy (-2) = x2(1) x(3)

[0195] =0

[0196] r xy (-1)=x2(1)·x(2)+x2(2)·x(3)

[0197] =0

[0198] r xy (0)=x2(1)·x(1)+x2(2)·x(2)+x2(3)·x(3)

[0199] =0

[0200] r xy (1)=x2(2)·x(1)+x2(3)·x(2)+x2(4)·x(3)

[0201] =0

[0202] r xy (2)=x2(3)·x(1)+x2(4)·x(2)+x2(5)·x(3)

[0203] =0

[0204] r xy (3)=x2(4)·x(1)+x2(5)·x(2)+x2(6)·x(3)

[0205] =0

[0206] r xy (4)=x2(5)·x(1)+x2(6)·x(2)+x2(7)·x(3)

[0207] =a·a

[0208] r xy (5)=x2(6)·x(1)+x2(7)·x(2)+x2(8)·x(3)

[0209] =0

[0210] r xy (6)=x2(7)·x(1)+x2(8)·x(2)+x2(9)·x(3)

[0211] =0

[0212] r xy(7)=x2(8)·x(1)+x2(9)·x(2)+x2(10)·x(3)

[0213] =0

[0214] r xy (8)=x2(9)·x(1)+x2(10)·x(2)+x2(11)·x(3)

[0215] =0

[0216] r xy (9)=x2(10)·x(1)+x2(11)·x(2)+x2(12)·x(3)

[0217] =a·a

[0218] r xy (10)=x2(11)·x(1)+x2(12)·x(2)+x2(13)·x(3)=0

[0219] r xy (11)=x2(12)·x(1)+x2(13)·x(2)+x2(14)·x(3)=0

[0220] r xy =(0,0,0,0,0,0,a 2 ,0,0,0,0,a 2 ,0,0)

[0221] The correlation processing can obtain the detection signal.

[0222] When the received signal has random interference, then:

[0223] x=(0,0,a)+y1

[0224] Assume that the signal received by non-wellhead geophone 1 is:

[0225] x1=(0,0,0,0,a,0,0,a)+y2

[0226] Assume that the signal received by non-wellhead geophone 2 is:

[0227] x2=(0,0,0,0,0,0,a,0,0,0,0,a)+y3

[0228] x1 and x are calculated using the previous correlation formula. Due to random interference, the correlation result is 0, which can be obtained:

[0229] r xy (-2) = x1(1) x(3)

[0230] =0

[0231] r xy (-1)=x1(1) x(2)+x1(2) x(3)

[0232] =0

[0233] r xy (0)=x1(1) x(1)+x1(2) x(2)+x1(3) x(3)

[0234] =0

[0235] r xy (1)=x1(2) x(1)+x1(3) x(2)+x1(4) x(3)

[0236] =0

[0237] r xy (2)=x1(3) x(1)+x1(4) x(2)+x1(5) x(3)

[0238] =a·a

[0239] r xy (3)=x1(4) x(1)+x1(5) x(2)+x1(6) x(3)

[0240] =0

[0241] r xy (4)=x1(5) x(1)+x1(6) x(2)+x1(7) x(3)

[0242] =0

[0243] r xy (5)=x1(6) x(1)+x1(7) x(2)+x1(8) x(3)

[0244] =a·a

[0245] r xy (6)=x1(7) x(1)+x1(8) x(2)+x1(9) x(3)

[0246] =0

[0247] r xy (7)=x1(8) x(1)+x1(9) x(2)+x1(10) x(3)

[0248] =0

[0249] r xy =(0,0,0,0,a 2 ,0,0,a 2,0,0)

[0250] x2 and x are calculated using the previous correlation formula. Due to random interference, the correlation result is 0, which can be obtained:

[0251] r xy (-2) = x2(1) x(3)

[0252] =0

[0253] r xy (-1)=x2(1)·x(2)+x2(2)·x(3)

[0254] =0

[0255] r xy (0)=x2(1)·x(1)+x2(2)·x(2)+x2(3)·x(3)

[0256] =0

[0257] r xy (1)=x2(2)·x(1)+x2(3)·x(2)+x2(4)·x(3)

[0258] =0

[0259] r xy (2)=x2(3)·x(1)+x2(4)·x(2)+x2(5)·x(3)

[0260] =0

[0261] r xy (3)=x2(4)·x(1)+x2(5)·x(2)+x2(6)·x(3)

[0262] =0

[0263] r xy (4)=x2(5)·x(1)+x2(6)·x(2)+x2(7)·x(3)

[0264] =a·a

[0265] r xy (5)=x2(6)·x(1)+x2(7)·x(2)+x2(8)·x(3)

[0266] =0

[0267] r xy (6)=x2(7)·x(1)+x2(8)·x(2)+x2(9)·x(3)

[0268] =0

[0269] rxy (7)=x2(8)·x(1)+x2(9)·x(2)+x2(10)·x(3)

[0270] =0

[0271] r xy (8)=x2(9)·x(1)+x2(10)·x(2)+x2(11)·x(3)

[0272] =0

[0273] r xy (9)=x2(10)·x(1)+x2(11)·x(2)+x2(12)·x(3)

[0274] =a·a

[0275] r xy (10)=x2(11)·x(1)+x2(12)·x(2)+x2(13)·x(3)=0

[0276] r xy (11)=x2(12)·x(1)+x2(13)·x(2)+x2(14)·x(3)=0

[0277] r xy =(0,0,0,0,0,0,a 2 ,0,0,0,0,a 2 ,0,0)

[0278] Correlation processing can also obtain detection signals when there is random interference.

[0279] Step 6: All data are processed according to step 5; All data are processed according to step 5. All data refer to the excitation in the well, the source is located at a certain depth in the well, and the excited seismic waves propagate in the well. Since the medium in the well is mud, which is a uniform medium, according to the principle of seismic reflection, when there is a difference in wave impedance in the underground medium, reflection and transmission will occur. The amplitude of reflection and transmission is related to the reflection and transmission coefficient of the medium. When there is no difference in wave impedance in the underground medium, that is, it is a uniform medium, no reflection will occur, only transmission, and the amplitude of transmission is related to the transmission coefficient of the medium. Therefore, seismic waves propagate in the well without reflection, only transmission. The detector at the wellhead can only receive the direct wave signal and interference signal of the source, but not the reflected signal. When the seismic wave propagates to the formation around the well, according to seismic theory, the medium can be approximated as a layered medium. The layered medium is an inhomogeneous medium with a difference in wave impedance. Therefore, seismic waves propagate in the formation around the well and reflection and transmission will occur. According to Snell's law, the detector at the wellhead cannot receive the reflected wave, while the detector not at the wellhead can receive the reflected wave and the direct wave, including the seismic traces received by the detector not at the wellhead and the seismic traces received by the detector at the wellhead.

[0280] Step 7: Time shift processing: Time shift processing refers to calculating the direct wave time t of the first track, and shifting the correlated data backward by t to obtain the seismic record after cross-correlation filtering. The first track refers to the seismic track closest to the reference track.

[0281] Step 8: Output the result profile.

[0282] In a specific embodiment 1 of the present invention, Figure 1 As shown, Figure 1 The present invention is a flow chart of the weak signal extraction method of the three-dimensional excitation data in the well.

[0283] Step 101, input data;

[0284] Step 102, selecting a seismic trace near the well site;

[0285] Step 103, selecting the seismic trace as a reference seismic trace;

[0286] Step 104, select a channel of input data;

[0287] Step 105, cross-correlation processing is performed on the data and the reference seismic trace;

[0288] Step 106, process all data according to step 5;

[0289] Step 107, time shift processing;

[0290] Step 108, output the result profile.

[0291] The schematic diagram of weak signal extraction and processing data using the present invention is as follows Figure 3 The schematic diagram of data processing without using the present invention is shown in Figure 2 As shown. It can be seen from the figure that the signal-to-noise ratio of the data not processed by the present invention is low, while the signal-to-noise ratio of the imaging data processed by the present invention is high. A method for extracting weak signals of the present invention is correct. Since the conventional correlation detection method uses an approximate detection signal as a reference signal that is different from the real reference signal, the effect is not ideal.

[0292] The actual weak signal extraction and processing effect of the present invention is ideal. This is because the present invention realizes weak signal extraction and processing, and by using wellhead data as a reference signal, the correlation detection method in weak signal extraction is superior to the conventional correlation detection method.

[0293] In a specific embodiment 2 of the present invention, as Figure 4 As shown, Figure 4 Flow chart of the weak signal extraction method of three-dimensional excitation data in a well according to the present embodiment.

[0294] Step 101, input data;

[0295] Step 102, selecting a seismic trace far from the well site;

[0296] Step 103, selecting the seismic trace as a reference seismic trace;

[0297] Step 104, select a channel of input data;

[0298] Step 105, cross-correlation processing is performed on the data and the reference seismic trace;

[0299] Step 106, process all data according to step 5;

[0300] Step 107, time shift processing;

[0301] Step 108, output the result profile.

[0302] The schematic diagram of weak signal extraction and processing data using the present invention is as follows Figure 6 The schematic diagram of data processing without using the present invention is shown in Figure 5 As shown. It can be seen from the figure that the signal-to-noise ratio of the original data is low, and the signal-to-noise ratio of the imaging data without the data processing of the present invention is not high. A method of extracting weak signals of the present invention is correct. Since the seismic trace far away from the wellhead is used as the reference signal in this embodiment, the effect is not ideal because the influence of the propagation path is different from the real reference signal.

[0303] The actual weak signal extraction and processing effect of the present invention is ideal. This is because the present invention realizes weak signal extraction and processing, and by using wellhead data as a reference signal, the correlation detection method in weak signal extraction is superior to the conventional correlation detection method.

[0304] In a specific embodiment 3 of the present invention, as Figure 7 As shown, Figure 7 The present invention is a flow chart of the weak signal extraction method of the three-dimensional excitation data in the well.

[0305] Step 101, inputting data after random noise is removed by fx domain deconvolution method;

[0306] Step 102, selecting a seismic trace near the well site;

[0307] Step 103, selecting the seismic trace as a reference seismic trace;

[0308] Step 104, select a channel of input data;

[0309] Step 105, cross-correlation processing is performed on the data and the reference seismic trace;

[0310] Step 106, process all data according to step 5;

[0311] Step 107, time shift processing;

[0312] Step 108, output the result profile.

[0313] The schematic diagram of weak signal extraction and processing data using the present invention is as follows Figure 8 The schematic diagram of data processing without using the present invention is shown in Figure 5 As shown. It can be seen from the figure that the signal-to-noise ratio of the original data is low, while the imaging signal-to-noise ratio of the data processed by the present invention is high, and the signal-to-noise ratio of the denoised data is high in the shallow, medium and deep layers compared with the original data. A method for extracting weak signals of the present invention is correct. Since the data processed by the fx domain deconvolution method is used as input data in this embodiment, the effect is more ideal because the input data has a certain signal-to-noise ratio.

[0314] The actual weak signal extraction and processing effect of the present invention is ideal. This is because the present invention realizes weak signal extraction and processing, and by using wellhead data as a reference signal, the correlation detection method in weak signal extraction is superior to the conventional correlation detection method.

[0315] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions recorded in the aforementioned embodiments or replace some of the technical features therein with equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

[0316] Except for the technical features described in the specification, all other technical features are known technologies to those skilled in the art.

Claims

1. A method for extracting weak signals from three-dimensional excitation data in a well, characterized in that: The weak signal extraction method of the three-dimensional excitation data in the well includes: Step 1: Collect seismic data; Step 2: Select the seismic trace near the well site; Step 3: Select the seismic trace in step 2 as the reference seismic trace; Step 4: Select a seismic trace from the seismic data collected in step 1; Step 5: Perform cross-correlation processing on a seismic trace in step 4 and the reference seismic trace; Step 6: Process all seismic traces of seismic data collected in step 1 according to step 5; Step 7: Perform time shift processing; Step 8: Output the result profile; In step 1, the collected seismic data is a seismic signal with certain morphological regular characteristics; In step 2, the seismic trace near the well site is selected to be the closest to the wellhead. The seismic trace is excited in the well, the source is located at a certain depth in the well, and the direct wave and random interference are received, but the formation reflection wave is not received. In step 3, the seismic trace in step 2 is selected as a reference seismic trace, where the reference seismic trace is a seismic wavelet, and the seismic wavelet refers to a seismic wavelet whose convolution with the reflection coefficient plus the noise equals the seismic record according to the seismic convolution theory; In step 4, the selected seismic trace refers to a seismic trace that is excited in the well, has a source located at a certain depth in the well, receives direct waves and random interference, and does not receive formation reflection waves; In step 5, a seismic trace in step 4 is cross-correlated with the reference seismic trace. The cross-correlation formula is: n=1,2,3...,N x(n) is the nth amplitude value of the signal, y(n) is the nth amplitude value of the scanned signal, and N is the number of sampling points; r xy (m) is the mth amplitude value of the cross-correlation signal; The specific steps are to set the signal received by the wellhead detector as: x=(0,0,a) Assume that the signal received by non-wellhead geophone 1 is: x1=(0,0,0,0,a,0,0,a) Assume that the signal received by non-wellhead geophone 2 is: x2=(0,0,0,0,0,0,a,0,0,0,0,a) a is the amplitude value of the seismic wavelet; In step 6, all the seismic traces for collecting seismic data in step 1 are seismic traces that are excited in the well, with the source located at a certain depth in the well, receiving direct waves and random interference, and not receiving formation reflection waves; In step 7, time shift processing refers to calculating the direct wave time t of the first track, and shifting the correlated data backward by t to obtain the seismic record after cross-correlation filtering; the first track refers to the seismic track closest to the reference track.

2. The weak signal extraction method of three-dimensional excitation data in a well according to claim 1, characterized in that: In step 5, x1 and x are calculated using the previous correlation formula: r xy (-2)=x1(1) x(3)=0 r xy (-1)=x1(1) x(2)+x1(2) x(3)=0 r xy (0)=x1(1) x(1)+x1(2) x(2)+x1(3) x(3)=0 r xy (1)=x1(2) x(1)+x1(3) x(2)+x1(4) x(3)=0 r xy (2)=x1(3)·x(1)+x1(4)·x(2)+x1(5)·x(3)=a·a r xy (3)=x1(4) x(1)+x1(5) x(2)+x1(6) x(3)=0 r xy (4)=x1(5) x(1)+x1(6) x(2)+x1(7) x(3)=0 r xy (5)=x1(6)·x(1)+x1(7)·x(2)+x1(8)·x(3)=a·a r xy (6)=x1(7) x(1)+x1(8) x(2)+x1(9) x(3)=0 r xy (7)=x1(8) x(1)+x1(9) x(2)+x1(10) x(3)=0 r xy =(0,0,0,0,a 2 ,0,0,a 2 ,0,0) x2 and x are calculated using the previous correlation formula: xy (-2) = x2(1) x(3) = 0 r xy (-1)=x2(1)·x(2)+x2(2)·x(3)=0 r xy (0)=x2(1)·x(1)+x2(2)·x(2)+x2(3)·x(3)=0 r xy (1)=x2(2)·x(1)+x2(3)·x(2)+x2(4)·x(3)=0 r xy (2)=x2(3)·x(1)+x2(4)·x(2)+x2(5)·x(3)=0 r xy (3)=x2(4)·x(1)+x2(5)·x(2)+x2(6)·x(3)=0 r xy (4)=x2(5)·x(1)+x2(6)·x(2)+x2(7)·x(3)=a·a r xy (5)=x2(6)·x(1)+x2(7)·x(2)+x2(8)·x(3)=0 r xy (6)=x2(7)·x(1)+x2(8)·x(2)+x2(9)·x(3)=0 r xy (7)=x2(8)·x(1)+x2(9)·x(2)+x2(10)·x(3)=0 r xy (8)=x2(9)·x(1)+x2(10)·x(2)+x2(11)·x(3)=0 r xy (9)=x2(10)·x(1)+x2(11)·x(2)+x2(12)·x(3)=a·a r xy (10)=x2(11)·x(1)+x2(12)·x(2)+x2(13)·x(3)=0 r xy (11)=x2(12)·x(1)+x2(13)·x(2)+x2(14)·x(3)=0 r xy =(0,0,0,0,0,0,a 2 ,0,0,0,0,a 2 ,0,0) The detection signal is obtained by correlation processing.

3. The weak signal extraction method of three-dimensional excitation data in a well according to claim 1, characterized in that: In step 5, when the received signal has random interference, then: x=(0,0,a)+y1 Assume that the signal received by non-wellhead geophone 1 is: x1=(0,0,0,0,a,0,0,a)+y2 Assume that the signal received by non-wellhead geophone 2 is: x2=(0,0,0,0,0,0,a,0,0,0,0,a)+y3 y1, y2, y3 are the amplitude values ​​of the random signal; x1 and x are calculated using the previous correlation formula. Due to random interference, the correlation result is 0: r xy (-2)=x1(1) x(3)=0 r xy (-1)=x1(1) x(2)+x1(2) x(3)=0 r xy (0)=x1(1) x(1)+x1(2) x(2)+x1(3) x(3)=0 r xy (1)=x1(2) x(1)+x1(3) x(2)+x1(4) x(3)=0 r xy (2)=x1(3)·x(1)+x1(4)·x(2)+x1(5)·x(3)=a·a r xy (3)=x1(4) x(1)+x1(5) x(2)+x1(6) x(3)=0 r xy (4)=x1(5) x(1)+x1(6) x(2)+x1(7) x(3)=0 r xy (5)=x1(6)·x(1)+x1(7)·x(2)+x1(8)·x(3)=a·a r xy (6)=x1(7) x(1)+x1(8) x(2)+x1(9) x(3)=0 r xy (7)=x1(8) x(1)+x1(9) x(2)+x1(10) x(3)=0 r xy =(0,0,0,0,a 2 ,0,0,a 2 ,0,0) x2 and x are calculated using the previous correlation formula. Due to random interference, the correlation result is 0: r xy (-2)=x2(1)·x(3)=0 r xy (-1)=x2(1)·x(2)+x2(2)·x(3)=0 r xy (0)=x2(1)·x(1)+x2(2)·x(2)+x2(3)·x(3)=0 r xy (1)=x2(2)·x(1)+x2(3)·x(2)+x2(4)·x(3)=0 r xy (2)=x2(3)·x(1)+x2(4)·x(2)+x2(5)·x(3)=0 r xy (3)=x2(4)·x(1)+x2(5)·x(2)+x2(6)·x(3)=0 r xy (4)=x2(5)·x(1)+x2(6)·x(2)+x2(7)·x(3)=a·a r xy (5)=x2(6)·x(1)+x2(7)·x(2)+x2(8)·x(3)=0 r xy (6)=x2(7)·x(1)+x2(8)·x(2)+x2(9)·x(3)=0 r xy (7)=x2(8)·x(1)+x2(9)·x(2)+x2(10)·x(3)=0 r xy (8)=x2(9)·x(1)+x2(10)·x(2)+x2(11)·x(3)=0 r xy (9)=x2(10)·x(1)+x2(11)·x(2)+x2(12)·x(3)=a·a r xy (10)=x2(11)·x(1)+x2(12)·x(2)+x2(13)·x(3)=0 r xy (11)=x2(12)·x(1)+x2(13)·x(2)+x2(14)·x(3)=0 r xy =(0,0,0,0,0,0,a 2 ,0,0,0,0,a 2 ,0,0) When there is random interference, the detection signal is obtained by correlation processing.

Images