Seismic data processing amplitude preservation evaluation method based on similarity measurement

Abstract

The invention provides a seismic data processing amplitude preservation evaluation method based on similarity measurement. The seismic data processing amplitude preservation evaluation method comprises the following steps: S1, acquiring stacked sections before and after seismic data processing along a main measuring line direction; s2, calculating the relative variation of Pearson similarity coefficients before and after seismic data processing; s3, calculating and normalizing the relative variation of the Euclidean distance before and after seismic data processing; s4, calculating and normalizing the relative variation of the DTW distance before and after seismic data processing; and S5, calculating the average similarity of the seismic stack sections of all the main measuring lines, and carrying out weighted summation to obtain a comprehensive evaluation value for measuring the amplitude preservation performance of the seismic stack sections before and after seismic data processing. According to the method, the time window is solved in a segmented mode in the travel time direction, information of more local areas is captured, the accuracy of seismic data non-stationary signal processing is improved, the amplitude preservation characteristic of three-dimensional seismic data can be rapidly calculated, and compared with manual evaluation, reliability and efficiency are improved.

Description

Technical Field

[0001] The invention belongs to the field of geophysical technology, and in particular relates to an amplitude preservation evaluation method for seismic data processing based on similarity measurement. Background Art

[0002] In the process of seismic data processing, it is necessary to monitor the amplitude preservation of each step and evaluate the advantages and disadvantages of the processing steps through amplitude preservation. However, due to the lack of sufficient technical means, this process is often completed by processing personnel based on experience and observation, and there is no intuitive way to evaluate amplitude preservation.

[0003] The patent "A method and device for quantitatively evaluating the amplitude preservation of seismic data" (CN117406279A) uses an evaluation method based on wavelet analysis, which transforms the seismic data from the time domain to the frequency domain, performs wavelet analysis, and then calculates its coherence for evaluation. The patent "Amplitude preservation processing analysis and evaluation method" (CN102176055A) subtracts the original SEGY seismic data from the SEGY seismic data after amplitude preservation to obtain differential seismic data, analyzes whether the differential record contains effective signal components and noise components in the differential seismic record, and determines the amplitude preservation of the processed data. The prior art has the problem that it is difficult to quantitatively and accurately evaluate the amplitude preservation in the process of seismic data processing quality control. Summary of the invention

[0004] In order to solve the problem that amplitude preservation is difficult to be quantitatively and accurately evaluated in the process of quality control of seismic data processing, the embodiment of the present application provides a method for evaluating amplitude preservation of seismic data processing based on similarity measurement, which can quantitatively evaluate the amplitude preservation situation in the process of seismic data processing and provide an intuitive basis for processing personnel to adjust processing methods and parameters.

[0005] The embodiment of the present application provides a method for evaluating amplitude preservation of seismic data processing based on similarity metrics, including:

[0006] S1, obtain the stacked sections of seismic data before and after processing along a main survey line;

[0007] S2, respectively calculate the Pearson similarity coefficient of two adjacent seismic traces before and after seismic data processing, and then find the relative change of the Pearson similarity coefficient before and after seismic data processing;

[0008] S3, respectively calculating the Euclidean distances between two adjacent seismic traces before and after the seismic data processing, and then calculating the relative change of the Euclidean distances before and after the seismic data processing and normalizing them;

[0009] S4, respectively calculating the DTW distances of two adjacent seismic traces before and after the seismic data processing, and then calculating the relative change of the DTW distances before and after the seismic data processing and normalizing them;

[0010] S5, calculate the average similarity of the seismic stacking sections of all main survey lines and sum them up by weight, and obtain a comprehensive evaluation value to measure the amplitude preservation of the seismic stacking sections before and after seismic data processing.

[0011] In some embodiments, in step S1, the stacked sections before and after seismic data processing are taken out along the direction of the j-th main survey line j=1, 2, ..., J. The stacked section before seismic data processing is recorded as S(x, t), and the stacked section after seismic data processing is recorded as H(x, t), where x represents the CMP number and t represents the travel time.

[0012] In some embodiments, step S2 includes: S21, in the jth main survey line of the stacked profile before processing, let s i (t) represents the common center point seismic trace data of the ith trace, s i (t) = S(x i ,t), calculate s i (t) The seismic trace s adjacent to it i+1 (t) Pearson similarity coefficient; S22, in the jth main line of the processed stacked profile, let h i (t) = H(x i ,t), represents the seismic data of the i-th track after processing, and calculates the Pearson similarity coefficient between two adjacent seismic tracks after processing; S23, calculates the relative change of the similarity coefficient of the j-th main survey line before processing and the similarity coefficient after processing.

[0013] In some embodiments, in step S21, s i (t) = S(x i ,t), using the sliding time window algorithm to calculate s segment by segment i (t) The seismic trace s adjacent to it i+1 Pearson similarity coefficient P(s) between (t) i ,s i+1 )(i=1,2,...,m-1), assuming that the time window length is Δτ, the sliding step is Δτ / 2, and a total of K time windows are set to divide the seismic traces. The calculation formula of K is:

[0014] The time range of the kth time window is [τ k ,τ k +Δτ], where The formula for calculating the Pearson similarity coefficient in the kth time window is:

[0015]

[0016] Where E(s i,k ) and E(s i+1,k ) represent the two adjacent seismic traces si(t) and si +1 (t) the average amplitude in the kth time window;

[0017] The Pearson similarity coefficient P(s) of all adjacent two tracks of the previous main survey line S(x,t) is processed. i ,s i+1 ) is expressed as:

[0018]

[0019] In step S22, the Pearson similarity coefficient P(h i ,h i+1 ) is also calculated in sections using the sliding time window algorithm, i = 1, 2, ..., m-1, and the time window division is consistent with step S21. The Pearson similarity coefficient between two adjacent seismic traces after processing in the kth time window is P(h i,k ,h i+1,k ), the calculation formula is:

[0020]

[0021] Where E(h i,k ) and E(h i+1,k ) represent the two adjacent seismic traces hi(t) and hi after processing. +1 (t) the average amplitude in the kth time window;

[0022] In step S23, the change degree P of the similarity coefficient between the i-th adjacent seismic traces before and after the j-th main survey line is processed is obtained. i,j , the formula is:

[0023]

[0024] Where i=1,2,...,m-1.

[0025] In some embodiments, step S3 includes: S31, in the jth main survey line of the stacked profile before processing, calculating two adjacent seismic traces s i (t) and s i+1 (t) Euclidean distance between two adjacent seismic traces; S32, i (t) and s i+1 (t) is transformed into the range [0,1]; S33, the h of two adjacent seismic traces after calculation is i (t) and hi+1 (t) Euclidean distance between two adjacent seismic traces after processing; S34, i (t) and h i+1 The metric value of the Euclidean distance between (t) is transformed to [0,1]; S35, calculating the relative change of the Euclidean distance of the j-th main survey line before processing and the Euclidean distance after processing.

[0026] In some embodiments, in step S31, two adjacent seismic traces s are calculated. i (t) and s i+1 (t) The Euclidean distance d(s i ,s i+1 )The formula is:

[0027]

[0028] Where i = 1, 2, ..., m-1;

[0029] In step S32, two adjacent seismic traces s i (t) and s i+1 (t) The Euclidean distance d(s i ,s i+1 ) is transformed to [0,1] to obtain D(s i ,s i+1 ), the transformation formula is:

[0030]

[0031] Where A is a positive real number, which is used to control the degree of expansion and contraction of the horizontal coordinate of the function graph;

[0032] In step S33, the h of two adjacent seismic traces after seismic data processing is calculated. i (t) and h i+1 (t) between the Euclidean distance d(h i ,h i+1 )The formula is:

[0033]

[0034] Where i = 1, 2, ..., m-1;

[0035] After processing in step S34, the two adjacent seismic traces h i (t) and h i+1 (t) between the Euclidean distance d(h i ,h i+1 ) is transformed to [0,1] to obtain D(h i ,h i+1 ), the transformation formula is:

[0036]

[0037] Wherein A is a positive real number, which is consistent with A in step S32;

[0038] In step S35, the change degree D of the Euclidean distance between the i-th adjacent seismic traces before and after the j-th main survey line is processed is obtained. i,j , the formula is:

[0039]

[0040] Where i=1,2,...,m-1.

[0041] In some embodiments, step S4 includes: S41, calculating the adjacent two seismic traces s before processing i (t) and s i+1 (t) is the DTW distance between two adjacent seismic traces, which measures the similarity of the two seismic traces; S42, the two adjacent seismic traces s i (t) and s i+1 (t) is transformed into the range [0,1]; S43, the DTW distance between two adjacent seismic traces h after calculation is processed. i (t) and h i+1 (t) to measure the similarity of two seismic traces; S44, the two adjacent seismic traces h i (t) and h i+1 (t) is transformed into the range of [0,1]; S45, calculating the relative change of the DTW distance before and after the processing of the j-th main measurement line.

[0042] In some embodiments, in step S41, the DTW distance dtw(si,si) between two adjacent seismic traces before processing is obtained. +1 ), dtw(s i ,s i+1 )=dtw(s i,N ,s i+1,N );

[0043] In step S42, two adjacent seismic traces s i (t) and s i+1 (t) between the DTW distance dtw(s i ,s i+1 ) is transformed to [0,1] to obtain dtw'(s i ,s i+1 ), the transformation formula is:

[0044]

[0045] Where A is a positive real number, which is used to control the degree of expansion and contraction of the horizontal coordinate of the function graph;

[0046] In step S43, the DTW distance dtw (h i ,h i+1 ), dtw(h i ,h i+1 )=dtw(h i,N ,h i+1,N ); In step S44, the two adjacent seismic traces hi(t) and hi +1 (t) between the DTW distance dtw(h i ,h i+1 ) is transformed to [0,1] to obtain dtw'(h i ,h i+1 ), the transformation formula is:

[0047]

[0048] Where A is a positive real number, which is used to control the degree of expansion and contraction of the horizontal coordinate of the function graph;

[0049] In step S45, the change degree dtw' of the DTW distance between the i-th adjacent seismic traces before and after the j-th main survey line is obtained. i,j , the formula is:

[0050]

[0051] Where i=1,2,...,m-1.

[0052] In some embodiments, in step S5, a comprehensive evaluation value Comb for measuring the amplitude preservation of the seismic stacking section before and after the seismic data processing is obtained, and the calculation formula is:

[0053] Comb=α·P mean +β·D mean +γ·dtw' mean

[0054]

[0055]

[0056] Among them, α, β and γ are weights, all of which are positive real numbers.

[0057] The amplitude preservation evaluation method of seismic data processing based on similarity metric in the embodiment of the present application has the following beneficial effects:

[0058] This application directly calculates the similarity measurement values ​​of adjacent channels in the time domain, and then analyzes the changes in the similarity measurement values ​​before and after processing, and then quantitatively evaluates the amplitude preservation in the process of seismic data processing, providing an intuitive basis for processing personnel to adjust processing methods and parameters. It can quickly calculate the amplitude preservation characteristics of three-dimensional seismic data, and compared with manual evaluation, it improves reliability and efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 This is a flow chart of the amplitude preservation evaluation method for seismic data processing based on similarity metrics in this application;

[0060] Figure 2 This is a schematic diagram of the main survey line and adjacent tracks in an embodiment of the present application;

[0061] Figure 3 This is a schematic diagram of the division of the sliding time window in an embodiment of the present application;

[0062] Figure 4 This is a schematic diagram of a conversion function curve according to an embodiment of the present application;

[0063] Figure 5 Schematic diagram of DTW distance calculation between two adjacent seismic traces according to an embodiment of the present application. DETAILED DESCRIPTION

[0064] The present application is further described below in conjunction with the accompanying drawings and embodiments.

[0065] In the following introduction, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance. The following introduction provides multiple embodiments of the present invention, and different embodiments can be replaced or combined, so this application can also be considered to include all possible combinations of the same and / or different embodiments recorded. Therefore, if one embodiment includes features A, B, and C, and another embodiment includes features B and D, then this application should also be considered to include embodiments containing one or more of all other possible combinations of features A, B, C, and D, even though the embodiment may not be clearly recorded in the following text.

[0066] Embodiment 1

[0067] The present application is a method for quantitatively obtaining amplitude preservation before and after processing by calculating similarity metric values ​​of adjacent common center point gathers in seismic data stacking gathers.

[0068] like Figure 1As shown, the present application provides a method for evaluating the amplitude preservation of seismic data processing based on similarity measurement, including: S1, obtaining stacked sections before and after seismic data processing along a main survey line; S2, respectively calculating the Pearson similarity coefficient of two adjacent seismic traces before and after seismic data processing, and then calculating the relative change of the Pearson similarity coefficient before and after the seismic data processing; S3, respectively calculating the Euclidean distance of two adjacent seismic traces before and after the seismic data processing, and then calculating the relative change of the Euclidean distance before and after the seismic data processing and normalizing it; S4, respectively calculating the DTW distance of two adjacent seismic traces before and after the seismic data processing, and then calculating the relative change of the DTW distance before and after the seismic data processing and normalizing it; S5, calculating the average similarity of the seismic stacked sections of all main survey lines and weighted summing them to obtain a comprehensive evaluation value for measuring the amplitude preservation of the seismic stacked sections before and after the seismic data processing.

[0069] This application directly calculates the similarity measurement values ​​of adjacent channels in the time domain, and then analyzes the changes in the similarity measurement values ​​before and after processing, and then quantitatively evaluates the amplitude preservation in the process of seismic data processing, providing an intuitive basis for processing personnel to adjust processing methods and parameters. It can quickly calculate the amplitude preservation characteristics of three-dimensional seismic data, and compared with manual evaluation, it improves reliability and efficiency.

[0070] Embodiment 2

[0071] The present application provides a method for evaluating amplitude preservation of seismic data processing based on similarity metrics, including:

[0072] S1, the stacked sections of seismic data before and after processing are obtained along a main survey line.

[0073] S2, calculate the Pearson similarity coefficient of two adjacent seismic traces before and after seismic data processing, and then find the relative change of the Pearson similarity coefficient before and after seismic data processing.

[0074] This step includes: S21, in the jth main survey line of the stacked profile before processing, let s i (t) represents the common center point seismic trace data of the ith trace, s i (t) = S(x i ,t), calculate s i (t) The seismic trace s adjacent to it i+1 (t) Pearson similarity coefficient; S22, in the jth main line of the processed stacked profile, let h i (t) = H(x i,t), represents the seismic data of the i-th track after processing, and calculates the Pearson similarity coefficient between two adjacent seismic tracks after processing; S23, calculates the relative change of the similarity coefficient of the j-th main survey line before processing and the similarity coefficient after processing.

[0075] S3, respectively calculate the Euclidean distances between two adjacent seismic traces before and after the seismic data are processed, and then find the relative change of the Euclidean distances before and after the seismic data are processed and normalized.

[0076] This step includes: S31, in the jth main survey line of the stacked profile before processing, calculate the two adjacent seismic traces s i (t) and s i+1 (t) Euclidean distance between two adjacent seismic traces; S32, i (t) and s i+1 (t) is transformed into the range [0,1]; S33, the h of two adjacent seismic traces after calculation is i (t) and h i+1 (t) Euclidean distance between two adjacent seismic traces after processing; S34, i (t) and h i+1 The metric value of the Euclidean distance between (t) is transformed to [0,1]; S35, calculating the relative change of the Euclidean distance of the j-th main survey line before processing and the Euclidean distance after processing.

[0077] S4, respectively calculating the DTW distances of two adjacent seismic traces before and after the seismic data processing, and then calculating the relative change of the DTW distances before and after the seismic data processing and normalizing them.

[0078] This step includes: S41, calculating the two adjacent seismic traces s before processing i (t) and s i+1 (t) is the DTW distance between two adjacent seismic traces, which measures the similarity of the two seismic traces; S42, the two adjacent seismic traces s i (t) and s i+1 (t) is transformed into the range [0,1]; S43, the DTW distance between two adjacent seismic traces h after calculation is processed. i (t) and h i+1 (t) to measure the similarity of two seismic traces; S44, the two adjacent seismic traces h i (t) and h i+1 (t) is transformed into the range of [0,1]; S45, calculating the relative change of the DTW distance before and after the processing of the j-th main measurement line.

[0079] S5, calculate the average similarity of the seismic stacking sections of all main survey lines and sum them up by weight, and obtain a comprehensive evaluation value to measure the amplitude preservation of the seismic stacking sections before and after seismic data processing.

[0080] This application uses the similarity metric value of two adjacent channels of seismic data stacking data, and then measures whether the metric value changes before and after processing to determine the amplitude preservation of data processing. Compared with directly comparing the data before and after processing, this method is more consistent with the actual situation of the underground geological body, and can better avoid the abnormal influence on the calculation of similarity metric value caused by changes in frequency, phase, and amplitude after seismic data processing.

[0081] Embodiment 3

[0082] This application uses adjacent channels to calculate similarity metrics. The data of adjacent channels on the superimposed channel profile record the reflection waves of two nearby points in the underground geological layer with zero shot offset, which means that the travel time data between them are very close, which is more consistent with the actual situation of the underground geological layer. Therefore, the calculation of the similarity metric value is more accurate.

[0083] In a specific embodiment, the amplitude preservation evaluation method for seismic data processing based on similarity measurement includes:

[0084] Step S1, such as Figure 2 As shown, the stacked sections before and after seismic data processing are obtained, and the data are taken out along the jth main survey line direction (j = 1, 2, ..., J), which are respectively recorded as S (x, t) and H (x, t) (x has m sampling points and t has n sampling points), where x represents the CMP number and t represents the travel time.

[0085] Step S2: In the jth main survey line of the stacked profile before seismic data processing, let s i (t) represents the common center point seismic trace data of the ith trace, then s i (t) is expressed as: s i (t) = S(x i ,t)

[0086] Calculate i (t) The seismic trace s adjacent to it i+1 Pearson similarity coefficient P(s) between (t) i ,s i+1 )(i=1,2,...,m-1), when calculating the Pearson similarity coefficient, the sliding time window algorithm is used for segmented calculation. Figure 3 As shown, to divide the time window, first set the time window length to Δτ, the sliding step length to Δτ / 2, and set K time windows to divide the seismic traces. The calculation formula of K is:

[0087]

[0088] Therefore, the time range of the kth time window is [τ k ,τ k +Δτ], where

[0089]

[0090] The formula for calculating the Pearson similarity coefficient in the kth time window is:

[0091]

[0092] Where E(s i,k ) and E(s i+1,k ) represent the two adjacent seismic traces si(t) and si +1 (t) the average amplitude in the kth time window;

[0093] Therefore, the Pearson similarity coefficient P(s i ,s i+1 ), expressed as:

[0094]

[0095] Generally, the obtained P(s i ,s i+1 ) is in the range of [0,1], and the larger its value is, the stronger the similarity between two adjacent seismic traces is.

[0096] Step S3 is similar to step S2. In the jth main survey line of the stacked profile after seismic data processing, let h i (t) = H(x i ,t) represents the processed seismic data of the ith channel, and the Pearson similarity coefficient P(h i ,h i+1 )(i=1,2,...,m-1), the sliding time window algorithm is also used for segmented calculation. The division of the time window is consistent with step S2. Let the Pearson similarity coefficient between two adjacent seismic traces after processing in the kth time window be P(h i,k ,h i+1,k ), the calculation formula is:

[0097]

[0098] Where E(h i,k ) and E(h i+1,k ) represent the two adjacent seismic traces hi(t) and hi after processing. +1 (t) is the average amplitude in the kth time window.i ,h i+1 ) is in the range of [0,1]. The larger its value is, the more similar the two time series are, that is, the larger the similarity measure value of the two adjacent seismic traces after processing is.

[0099] Step S4, calculate the similarity coefficient P(s) of the jth main survey line before processing i ,s i+1 ) and the similarity coefficient after processing P(h i ,h i+1 ) to obtain the change degree P of the similarity coefficient between the ith adjacent seismic traces before and after the jth main survey line processing i,j , the formula is:

[0100]

[0101] where i=1,2,...,m-1,P i,j The smaller the value, the stronger the similarity between the seismic traces before and after processing, and the better the amplitude preservation of seismic data processing.

[0102] Step S5, in the jth main survey line of the stacked profile before processing, calculate the two adjacent seismic traces s i (t) and s i+1 (t) The Euclidean distance d(s i ,s i+1 ). Its calculation formula is:

[0103]

[0104] where i=1,2,...,m-1,d(s i ,s i+1 ) value is smaller, indicating that the adjacent two seismic traces s i (t) and s i+1 (t)The stronger the similarity.

[0105] Step S6: combine two adjacent seismic traces i (t) and s i+1 (t) The Euclidean distance d(s i ,s i+1 ) is transformed to [0,1] to obtain D(s i ,s i+1 ), the transformation formula is:

[0106]

[0107] Where A is a positive real number used to control the degree of expansion and contraction of the horizontal coordinate of the function graph. i ,s i+1) value, it can be limited to [0,1] through transformation, which can avoid the special abnormal values ​​from having a great impact on the measurement of seismic trace similarity. When A=1, the schematic diagram of the transformation function curve is as follows Figure 4 shown.

[0108] Step S7, similar to step S5, calculates the h of two adjacent seismic traces after seismic data processing. i (t) and h i+1 (t) between the Euclidean distance d(h i ,h i+1 ), and its calculation formula is:

[0109]

[0110] Where i=1,2,...,m-1.

[0111] Step S8, similar to step S6, is to process the two adjacent seismic traces h i (t) and h i+1 (t) between the Euclidean distance d(h i ,h i+1 ) is transformed to [0,1] to obtain D(h i ,h i+1 ), the transformation formula is:

[0112]

[0113] Wherein A is a positive real number, which is consistent with A in step S6.

[0114] Step S9, calculate the Euclidean distance D(s) of the jth main survey line before processing i ,s i+1 ) and the processed Euclidean distance D(h i ,h i+1 ) to obtain the relative change of the Euclidean distance between the i-th adjacent seismic traces before and after the j-th main survey line is processed. i,j , the formula is:

[0115]

[0116] where i = 1, 2, ..., m-1, D i,j The smaller the value, the stronger the similarity between the seismic traces before and after processing, and the better the amplitude preservation of seismic data processing.

[0117] Step S10, calculate the two adjacent seismic traces s before seismic data processing i (t) and s i+1 (t) between the DTW distance dtw(s i ,s i+1) measures the similarity of two seismic traces. Figure 5 As shown, it is known that s i,k =S(x i ,t k ), then dtw(s i,k ,s i+1,k )=d(s i,k ,s i+1,k )+min{dtw(s i,k-1 ,s i+1,k ),dtw(s i,k-1 ,s i+1,k-1 ),dtw(s i,k-1 ,s i+1,k-1 )}, where d(s i,k ,s i+1,k ) is s i,k and i+1,k The Euclidean distance between two points, the distance at the first time point is dtw(s i,1 ,s i+1,1 )=d(s i,1 ,s i+1,1 )=(s i,1 -s i+1,1 ) 2 , dtw(s i,k ,s i+1,k ) is calculated from k = 1, k = 2, ..., to k = N, and the DTW distance between two adjacent seismic traces before seismic data processing is obtained.

[0118] dtw(s i ,s i+1 )=dtw(s i,N ,s i+1,N )

[0119] dtw(s i ,s i+1 ) value is smaller, indicating that the adjacent two seismic traces s i (t) and s i+1 (t)The stronger the similarity.

[0120] Step S11: combine two adjacent seismic traces i (t) and s i+1 (t) between the DTW distance dtw(s i ,s i+1 ) is transformed to [0,1] to obtain dtw'(s i ,s i+1 ), the transformation formula is as follows

[0121]

[0122] Wherein A is a positive real number, which is consistent with A in step S6.

[0123] Step S12, calculate the h of two adjacent seismic traces after seismic data processing i (t) and h i+1 (t) between the DTW distance dtw(h i ,h i+1 ), which measures the similarity of two seismic traces. i,k =H(x i ,t k ), then dtw(h i,k, h i+1,k )=d(h i,k ,h i+1,k )+min{dtw(h i,k-1 ,h i+1,k ),dtw(h i,k-1 ,h i+1,k-1 ),dtw(h i,k-1 ,h i+1,k-1 )}, where d(h i,k ,h i+1,k ) is h i,k and h i+1,k The Euclidean distance between two points, the distance at the first time point is dtw(h i,1 ,h i+1,1 )=d(h i,1 ,h i+1,1 )=(h i,1 -h i+1,1 ) 2 , dtw(h i,k ,h i+1,k ) is calculated from k = 1, k = 2, ..., to k = N, and the DTW distance between two adjacent seismic traces after seismic data processing is obtained.

[0124] dtw(h i ,h i+1 )=dtw(h i,N ,h i+1,N )

[0125] dtw(s i ,s i+1 ) value is smaller, indicating that the h i (t) and h i+1 (t)The stronger the similarity.

[0126] Step S13: combine two adjacent seismic traces h i (t) and h i+1 (t) between the DTW distance dtw(h i ,h i+1 ) is transformed to [0,1] to obtain dtw'(h i ,hi+1 ), the transformation formula is as follows

[0127]

[0128] Wherein A is a positive real number, which is consistent with A in step S6.

[0129] Step S14, calculate the DTW distance dtw'(s) before processing the jth main measurement line i ,s i+1 ) and the processed DTW distance dtw'(h i ,h i+1 ) to obtain the change degree dtw' of the DTW distance between the i-th adjacent seismic traces before and after the j-th main survey line is processed i,j , the formula is as follows

[0130]

[0131] where i = 1, 2, ..., m-1, dtw' i,j The smaller the value, the stronger the similarity between the seismic traces before and after processing, and the better the amplitude preservation of seismic data processing.

[0132] Step S15, calculate the average similarity P of the seismic stacked sections of all main survey lines mean , D mean and dtw' mean , and weighted sum to obtain the comprehensive evaluation value Comb that measures the amplitude preservation of the seismic stacking profile before and after seismic data processing. The calculation formula is as follows

[0133] Comb=α·P mean +β·D mean +γ·dtw' mean

[0134]

[0135] Among them, α, β and γ are weights, all of which are positive real numbers, P i,j It represents the change degree of similarity coefficient between the ith adjacent seismic traces before and after the processing of the jth main survey line, which is the calculation result of S4. i,j Indicates the change in the Euclidean distance between the i-th adjacent seismic traces before and after the j-th main survey line is processed. It is the calculation result of S9. dtw' i,j It indicates the change degree of the DTW distance between the i-th adjacent seismic traces before and after the j-th main survey line is processed, which is the calculation result of S12. The smaller the comprehensive evaluation value Comb is, the higher the similarity of the stacked profiles before and after the seismic data is processed, and the stronger the amplitude preservation is.

[0136] In the present application, the Pearson similarity coefficient may be replaced by the Spearman correlation coefficient or the cosine similarity, and the Euclidean distance may be replaced by the Mahalanobis distance.

[0137] In the time domain, the present application scheme uses similarity metrics to evaluate the amplitude preservation before and after processing. In addition, the amplitude preservation is measured by first calculating the similarity metrics of adjacent channels and then analyzing the difference in the similarity metric values ​​before and after processing. This application provides three methods for calculating similarity metrics: similarity coefficient, Euclidean distance, and DTW distance to adapt to the amplitude preservation measurement of different processing methods, and designs a segmented solution time window along the travel time direction to capture more information in local areas, improve the accuracy of non-stationary signal processing of seismic data, and can quickly calculate the amplitude preservation characteristics of three-dimensional seismic data. Compared with manual evaluation, it improves reliability and efficiency.

[0138] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. 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.

Claims

1. A method for evaluating amplitude preservation of seismic data processing based on similarity metrics, characterized in that: include: S1, obtain the stacked sections of seismic data before and after processing along a main survey line; S2, respectively calculate the Pearson similarity coefficient of two adjacent seismic traces before and after seismic data processing, and then find the relative change of the Pearson similarity coefficient before and after seismic data processing; S3, respectively calculating the Euclidean distances between two adjacent seismic traces before and after the seismic data processing, and then calculating the relative change of the Euclidean distances before and after the seismic data processing and normalizing them; S4, respectively calculating the DTW distances of two adjacent seismic traces before and after the seismic data processing, and then calculating the relative change of the DTW distances before and after the seismic data processing and normalizing them; S5, calculate the average similarity of the seismic stacking sections of all main survey lines and sum them up by weight, and obtain a comprehensive evaluation value to measure the amplitude preservation of the seismic stacking sections before and after seismic data processing.

2. The amplitude preservation evaluation method for seismic data processing based on similarity measurement according to claim 1, characterized in that: In step S1, the stacked sections before and after seismic data processing are taken out along the direction of the j-th main survey line j=1,2,...,J. The stacked section before seismic data processing is recorded as S(x,t), and the stacked section after seismic data processing is recorded as H(x,t), where x represents the CMP number and t represents the travel time.

3. The amplitude preservation evaluation method for seismic data processing based on similarity metric according to claim 1 or 2, characterized in that: Step S2 includes: S21, in the jth main survey line of the stacked profile before processing, let s i (t) represents the common center point seismic trace data of the ith trace, s i (t) = S(x i ,t), calculate s i (t) The seismic trace s adjacent to it i+1 (t) Pearson similarity coefficient; S22, in the jth main line of the processed stacked profile, let h i (t) = H(x i ,t), represents the seismic data of the i-th track after processing, and calculates the Pearson similarity coefficient between two adjacent seismic tracks after processing; S23, calculates the relative change of the similarity coefficient of the j-th main survey line before processing and the similarity coefficient after processing.

4. The method for evaluating amplitude preservation of seismic data processing based on similarity metrics according to claim 3, characterized in that: In step S21, s i (t) = S(x i ,t), using the sliding time window algorithm to calculate s segment by segment i (t) The seismic trace s adjacent to it i+1 Pearson similarity coefficient P(s) between (t) i ,s i+1 )(i=1,2,...,m-1), assuming that the time window length is Δτ, the sliding step is Δτ / 2, and a total of K time windows are set to divide the seismic traces. The calculation formula of K is: The time range of the kth time window is [τ k ,τ k +Δτ], where The formula for calculating the Pearson similarity coefficient in the kth time window is: Where E(s i,k ) and E(s i+1,k ) represent the two adjacent seismic traces si(t) and si +1 (t) the average amplitude in the kth time window; The Pearson similarity coefficient P(s) of all adjacent two tracks of the previous main survey line S(x,t) is processed. i ,s i+1 ) is expressed as: In step S22, the Pearson similarity coefficient P(h i ,h i+1 ) is also calculated in sections using the sliding time window algorithm, i = 1, 2, ..., m-1, and the time window division is consistent with step S21. The Pearson similarity coefficient between two adjacent seismic traces after processing in the kth time window is P(h i,k ,h i+1,k ), the calculation formula is: Where E(h i,k ) and E(h i+1,k ) represent the two adjacent seismic traces hi(t) and hi after processing. +1 (t) the average amplitude in the kth time window; In step S23, the change degree P of the similarity coefficient between the i-th adjacent seismic traces before and after the j-th main survey line is processed is obtained. i,j , the formula is: Where i=1,2,...,m-1.

5. The amplitude preservation evaluation method for seismic data processing based on similarity metrics according to claim 1 or 2, characterized in that: Step S3 includes: S31, in the jth main survey line of the stacked profile before processing, calculating two adjacent seismic traces s i (t) and s i+1 (t) Euclidean distance between two adjacent seismic traces; S32, i (t) and s i+1 (t) is transformed into the range [0,1]; S33, the h of two adjacent seismic traces after calculation is i (t) and h i+1 (t) Euclidean distance between two adjacent seismic traces after processing; S34, i (t) and h i+1 The metric value of the Euclidean distance between (t) is transformed to [0,1]; S35, calculating the relative change of the Euclidean distance of the j-th main survey line before processing and the Euclidean distance after processing.

6. The amplitude preservation evaluation method for seismic data processing based on similarity measurement according to claim 5 is characterized in that: In step S31, two adjacent seismic traces s are calculated. i (t) and s i+1 (t) The Euclidean distance d(s i ,s i+1 )The formula is: Where i = 1, 2, ..., m-1; In step S32, two adjacent seismic traces s i (t) and s i+1 (t) The Euclidean distance d(s i ,s i+1 ) is transformed to [0,1] to obtain D(s i ,s i+1 ), the transformation formula is: Where A is a positive real number, which is used to control the degree of expansion and contraction of the horizontal coordinate of the function graph; In step S33, the h of two adjacent seismic traces after seismic data processing is calculated. i (t) and h i+1 (t) between the Euclidean distance d(h i ,h i+1 )The formula is: Where i = 1, 2, ..., m-1; After processing in step S34, the two adjacent seismic traces h i (t) and h i+1 (t) between the Euclidean distance d(h i ,h i+1 ) is transformed to [0,1] to obtain D(h i ,h i+1 ), the transformation formula is: Wherein A is a positive real number, which is consistent with A in step S32; In step S35, the change degree D of the Euclidean distance between the i-th adjacent seismic traces before and after the j-th main survey line is processed is obtained. i,j , the formula is: Where i=1,2,...,m-1.

7. The amplitude preservation evaluation method for seismic data processing based on similarity metrics according to claim 1 or 2, characterized in that: Step S4 includes: S41, calculating the two adjacent seismic traces s before processing i (t) and s i+1 (t) is the DTW distance between two adjacent seismic traces, which measures the similarity of the two seismic traces; S42, the two adjacent seismic traces s i (t) and s i+1 (t) is transformed into the range [0,1]; S43, the DTW distance between two adjacent seismic traces h after calculation is processed. i (t) and h i+1 (t) to measure the similarity of two seismic traces; S44, the two adjacent seismic traces h i (t) and h i+1 (t) is transformed into the range of [0,1]; S45, calculating the relative change of the DTW distance before and after the processing of the j-th main measurement line.

8. The method for evaluating amplitude preservation of seismic data processing based on similarity metrics according to claim 7, characterized in that: In step S41, the DTW distance dtw (s i ,s i+1 ), dtw(s i ,s i+1 )=dtw(s i,N ,s i+1,N ); In step S42, two adjacent seismic traces s i (t) and s i+1 (t) between the DTW distance dtw(s i ,s i+1 ) is transformed to [0,1] to obtain dtw'(s i ,s i+1 ), the transformation formula is: Where A is a positive real number, which is used to control the degree of expansion and contraction of the horizontal coordinate of the function graph; In step S43, the DTW distance dtw (h i ,h i+1 ), dtw(h i ,h i+1 )=dtw(h i,N ,h i+1,N ); In step S44, the two adjacent seismic traces hi(t) and hi +1 (t) between the DTW distance dtw(h i ,h i+1 ) is transformed to [0,1] to obtain dtw'(h i ,h i+1 ), the transformation formula is: Where A is a positive real number, which is used to control the degree of expansion and contraction of the horizontal coordinate of the function graph; In step S45, the change degree dtw' of the DTW distance between the i-th adjacent seismic traces before and after the j-th main survey line is obtained. i,j , the formula is: Where i=1,2,...,m-1.

9. The method for evaluating amplitude preservation of seismic data processing based on similarity metrics according to claim 1 or 2, characterized in that: In step S5, a comprehensive evaluation value Comb for measuring the amplitude preservation of the seismic stacking profile before and after the seismic data processing is obtained, and the calculation formula is: Comb=α·P mean +β·D mean +γ·dtw' mean Among them, α, β and γ are weights, all of which are positive real numbers.

Images