A casing anomaly identification method and system based on the local change rate of CCL signal

Abstract

The present invention relates to the field of natural gas exploration and development technology, and discloses a casing anomaly identification method and system based on the local change rate of CCL signals. The method comprises: obtaining the original CCL signals of each perforation section of the oil and gas well, the coupling point data in the CCL signals, and the casing data table; determining the position of the casing pup joint, marking the coupling points of the original CCL signals, and performing depth correction by combining the marked coupling data with the casing data table; calculating the local change rate of the depth-corrected CCL signal; determining the abnormal interval of the CCL signal by the abnormal threshold of the local change rate of the CCL signal; calculating the casing abnormal state score, and determining the casing abnormality level based on the quantile threshold division. The present invention analyzes the changes in the magnetic field distribution law implicit in the CCL signal, establishes a mathematical correlation between the CCL signal and the casing deformation state, and significantly improves the utilization rate of the CCL signal in casing health monitoring.

Description

Technical Field

[0001] The present invention relates to the technical field of natural gas exploration and development, and in particular to a casing anomaly identification method and system based on the local change rate of a CCL signal. Background Art

[0002] In the current development of shale gas in my country, horizontal well casing deformation (casing deformation) has become a key obstacle to efficient development. For example, in southern Sichuan, due to a combination of unfavorable factors, including complex geology, high geostress, and strong fracturing disturbances, casing deformation occurs at a relatively high rate. Plastic deformations such as diameter reduction and ovalization account for a significant proportion of these deformations. This problem leads to a series of serious consequences. On the one hand, it reduces the perforation cluster opening rate; on the other hand, it causes uneven fracturing fluid distribution and reduced proppant delivery efficiency, significantly weakening the effectiveness of individual well fracturing stimulations and resulting in production losses. More seriously, the casing deformation section is prone to forming localized stress concentration areas, which can lead to various accidents such as casing collapse and downhole tool sticking. This not only significantly increases well repair costs but also significantly increases safety risks, posing a serious challenge to the economic and efficient development of shale gas.

[0003] Current methods for detecting casing deformation primarily rely on multi-finger caliper instruments (MIT) and electromagnetic flaw detection (EMI). MIT uses mechanical contact measurement to obtain highly accurate casing inner diameter data. However, in horizontal well sections, its detection effectiveness is limited due to the increasing complexity of the wellbore trajectory and the limited mechanical stability of the tool. EMI, while based on the principle of electromagnetic induction to detect metal loss, is limited in its ability to distinguish between subtle deformations and internal and external wall defects. Furthermore, identifying casing deformation using current technology is expensive.

[0004] Traditionally, magnetic positioning logging, as a magnetic positioning logging technology, has the core function of detecting sudden changes in magnetic field intensity at the casing collar to achieve well depth positioning and correction, and is not directly used for casing deformation diagnosis. However, recent studies have shown that abnormal changes in the CCL (Casing Collar Locator) signal, such as abnormal magnetic field intensity attenuation gradients, waveform distortion, or broadening of the collar characteristic peak, are correlated with deformations such as casing diameter reduction and ovalization. The root cause of this correlation is that casing deformation can cause changes in the magnetic field distribution pattern near the collar. For example, when the casing shrinks, the magnetic coupling effect between the collar and the instrument probe is exacerbated; and ovalization deformation may cause the direction of the magnetic field vector to shift. These changes enable the CCL signal to provide an indirect indication of early casing deformation while completing the depth correction task. However, there is currently no effective method to accurately describe the abnormal status of the casing directly using the CCL signal. Currently, most judgments rely on manual experience, resulting in a low utilization rate of the CCL signal. This problem urgently needs to be effectively solved. Summary of the Invention

[0005] The purpose of the present invention is to provide a casing anomaly identification method and system based on the local change rate of the CCL signal. By analyzing the changes in the magnetic field distribution pattern implicit in the CCL signal, a mathematical correlation between the CCL signal and the casing deformation state is established. This method does not require reliance on other detection tools or manual experience intervention, and significantly improves the utilization rate of the CCL signal in casing health monitoring.

[0006] To achieve the above object, the present invention provides the following solutions:

[0007] A casing anomaly recognition method based on the local change rate of CCL signals, the method comprising the following steps:

[0008] S1, data preparation: obtain the original CCL signal of each perforation section of the oil and gas well, the coupling point data in the CCL signal, and the casing data table of the oil and gas well;

[0009] S2, depth correction: Based on the casing data table, the position of the casing pup joint is determined, and the position information of the CCL signal coupling point is obtained based on the original CCL signal and coupling point data. Combined with the depth information of the corresponding casing in the casing data table, the CCL signal corresponding to each casing is reconstructed and spliced ​​together to achieve depth correction;

[0010] S3, calculating the local change rate: calculating the local change rate of the depth-corrected CCL signal through a sliding window to obtain the local change rate of the CCL signal;

[0011] S4, determining the abnormal interval: normalizing the local change rate of the CCL signal, calculating the abnormal threshold of the normalized local change rate of the CCL signal, and determining the abnormal interval of the CCL signal according to the abnormal threshold of the local change rate of the CCL signal;

[0012] S5, casing anomaly level classification: Based on the CCL signal anomaly interval, determine the casing anomaly status score corresponding to each anomaly window, merge adjacent anomaly segments to obtain the merged casing anomaly status score, and determine the casing anomaly level based on the quantile threshold.

[0013] Furthermore, the depth correction step S2 specifically includes:

[0014] S201, by consulting the casing data table of the oil and gas well, determining the corresponding depth of the casing pup joint in the depth section corresponding to the original CCL signal, and if there is more than one, the one with the smallest depth shall prevail;

[0015] S202, the original CCL signal is converted through the coupling point Cut into several signal segments, each segment corresponds to a casing, and obtain the location information of the CCL signal coupling point;

[0016] According to the position of the casing nipple, find the depth interval of the casing in the casing data table corresponding to the k-th segment signal, and map the original CCL signal to the depth interval to generate a reconstructed CCL signal;

[0017] The reconstructed CCL signals are spliced ​​together according to the depth to obtain a depth-corrected CCL signal, wherein all the collar points in the depth-corrected CCL signal are aligned with the positions in the casing data table.

[0018] Furthermore, in the step S202, the original CCL signal is converted to Cut into X segments of signal, each segment of signal corresponds to a casing, including sampling points. The depth sequence of the kth segment signal after segmentation can be obtained by formula (1):

[0019]

[0020] Where: is the depth sequence corresponding to the kth segment signal after segmentation, is the original CCL signal, is the collar depth information of the CCL signal, X is the number of complete casings contained in the CCL signal, and the Split operator represents the segmentation operation of the original CCL signal. according to The depth in is divided into X segments, is the number of sampling points.

[0021] Furthermore, in said S202, the depth interval [ , ], satisfying the conditions shown in formula (2):

[0022]

[0023] Where: is the corrected starting depth of the casing, is the corrected end depth of the casing, is the starting depth of the kth casing, is the length of the sleeve;

[0024] In S202, mapping the original CCL signal to the depth interval to generate a reconstructed CCL signal specifically includes:

[0025] Map the original CCL signal to the depth interval, and perform linear interpolation on the depth interval to ensure that the new depth sequence is generated. sampling points to generate a new depth sequence, as shown in formula (3):

[0026]

[0027] Where: is the reconstructed depth sequence, is the corrected starting depth of the casing, is the corrected end depth of the casing, is the number of sampling points;

[0028] After calculating the new depth sequence, the original CCL signal value remains unchanged, and the original depth sequence is mapped to the reconstructed depth sequence to generate the reconstructed CCL signal. The calculation is shown in formula (4):

[0029]

[0030] Where: is the CCL signal after reconstructing the jth segment, The operator does not change the CCL value, but only changes the mapping operation of the corresponding depth sequence, transforming the signal from the original depth sequence Map to On the new depth sequence after reconstruction;

[0031] In S202, the reconstructed CCL signals are spliced ​​together according to the depth to obtain the depth-corrected CCL signals, which specifically includes:

[0032] After reconstructing the CCL signal corresponding to each casing according to the process in step S2, the reconstructed signals are spliced ​​together according to the depth to become the depth-corrected signal, as follows:

[0033]

[0034] Where: is the depth-corrected signal, is the CCL signal after reconstructing the jth segment, It represents the splicing of the CCL signals after the segmented X segments are reconstructed;

[0035] When splicing and reconstructing CCL signals, the depth of each connection point is saved to In the sequence, this sequence is the depth sequence corresponding to the tie point after depth correction.

[0036] Furthermore, the step S3, calculating the local change rate, specifically includes:

[0037] S301, split sliding window:

[0038] The window length of the sliding window is N (for example, 200) sampling points, and the sliding step is M (for example, 100 sampling points). When the sliding window is used to extract the signal, when the sliding window contains When the point in the sequence is skipped, the window is skipped. When calculating the local change rate of the CCL signal, when the first window and the last window do not contain the complete front and back windows, the current segment data is used as the data of the missing window; where N and M are both positive integers, and M <N;

[0039] S302, calculate the local change rate of the CCL signal:

[0040] The local change rates include the local signal fluctuation change rate LFSV, the local pulse change rate LPVR, the local mutation density change rate LMDVR, and the local depth association change rate LDAVR. Each local change rate is calculated through a sliding window.

[0041] Furthermore, the S302, calculating each local change rate through a sliding window, specifically includes:

[0042]

[0043] Where: LFSV is the local signal fluctuation rate corresponding to the current window, is the standard deviation of the CCL signal data in the current window, is the standard deviation of the CCL signal data in the previous window, is the standard deviation of the CCL data in the next window;

[0044]

[0045] Where: is the impulse factor of the current window, is the impulse factor of the previous window, is the pulse factor of the next window, LPVR is the local pulse change rate corresponding to the current window, is the maximum value operation, N is the number of sampling points in the window, is the CCL data of the current window, is the CCL data of the next window, It is the CCL data of the previous window;

[0046]

[0047] Where: Represents the number of extreme points in the x sequence, Operator representation Symbols;

[0048]

[0049] Where: LMDVR is the local mutation density change rate corresponding to the current window, N is the number of sampling points in the window, The operator represents the number of extreme points in the x sequence, is the CCL data of the current window, is the CCL data of the previous window, It is the CCL data of the next window;

[0050]

[0051] Where: is the sum of the absolute values ​​of the adjacent differences of CCL data in a window, is the CCL data point, is the depth point corresponding to the CCL data point, It is the sum of the absolute values ​​of the adjacent differences of the corresponding depths of the CCL data in a window;

[0052]

[0053] Where: LDAVR is the local depth correlation change rate corresponding to the current window, is the sum of the absolute values ​​of the adjacent differences of the CCL data in the current window, It is the sum of the absolute values ​​of the adjacent differences of the CCL data corresponding to the depth of the current window. is the sum of the absolute values ​​of the adjacent differences of the CCL data of the previous window, It is the sum of the absolute values ​​of the adjacent differences of the CCL data corresponding to the depth of the previous window. is the sum of the absolute values ​​of the adjacent differences of the CCL data of the next window, It is the sum of the absolute values ​​of the adjacent differences in the depth corresponding to the CCL data of the next window.

[0054] Furthermore, the step S4 of determining the abnormal interval specifically includes:

[0055] S401, respectively normalizing the four local change rates: local signal fluctuation change rate LFSV, local pulse change rate LPVR, local mutation density change rate LMDVR, and local depth association change rate LDAVR;

[0056] S402, respectively calculating the abnormal threshold of each local change rate after normalization;

[0057] S403, filtering each local change rate according to the abnormal threshold of each local change rate, marking the window corresponding to the local change rate exceeding the abnormal threshold as an abnormal window, and setting the local change rate abnormal flag of the corresponding abnormal window to 1;

[0058] When any flag bit of the four local change rates corresponding to a window is 1, the depth segment corresponding to the window is an abnormal segment of the CCL signal, and all abnormal segments constitute the abnormal interval of the CCL signal.

[0059] Furthermore, the above S5, casing abnormality classification, specifically includes:

[0060] S501, calculating the casing abnormal state score corresponding to each abnormal window :

[0061]

[0062] Where: is the casing abnormal state score, is the normalized local signal fluctuation rate, It is the abnormal flag of the local signal fluctuation rate. is the normalized local pulse change rate, It is the abnormal flag of the local pulse change rate. is the normalized local mutation density change rate, is the abnormal flag of the local mutation density change rate, is the local depth correlation change rate, It is the abnormal flag of the local depth correlation change rate;

[0063] S502: Merge adjacent abnormal segments and calculate the abnormal status score of the merged casing:

[0064] When the abnormal windows have depth overlap, the two windows are merged and the abnormal state score of the merged casing is calculated according to formula (15): :

[0065]

[0066] Where: is the casing abnormal state score corresponding to the merged abnormal segment, is the set of abnormal state scores of all original windows in the merged window, gather The maximum value in Operator Representation Set The average value of n is the number of overlapping segment windows; sort out the merged abnormal segments to obtain complete abnormal segment information;

[0067] S503, based on the combined casing abnormality score in step S502 According to the conditions shown in formula (16), the casing abnormality level is divided into three levels, namely severe abnormality, moderate abnormality, and slight abnormality:

[0068]

[0069] Where: Indicates the p-quantile of the data; in the above formula, corresponding to different levels, the value of p is 95%, 80%;

[0070] By classifying the abnormal state scores of the CCL signal in step S503, the abnormal state of the casing is identified.

[0071] The present invention further provides a casing anomaly identification system based on the local change rate of the CCL signal, which is used to execute the above-mentioned casing anomaly identification method based on the local change rate of the CCL signal, comprising:

[0072] The data preparation module is used to obtain the original CCL signal of each perforation section of the oil and gas well, the coupling point data in the CCL signal, and the casing data table of the oil and gas well;

[0073] The depth correction module is used to determine the position of the casing pup joint based on the casing data table, and obtain the position information of the CCL signal coupling point based on the original CCL signal and coupling point data. Combined with the depth information of the corresponding casing in the casing data table, the CCL signal corresponding to each casing is reconstructed and spliced ​​together to achieve depth correction;

[0074] A local change rate calculation module is used to calculate the local change rate of the CCL signal after depth correction through a sliding window to obtain the local change rate of the CCL signal;

[0075] an abnormal interval determination module, configured to normalize the local change rate of the CCL signal, calculate an abnormal threshold of the normalized local change rate of the CCL signal, and determine the abnormal interval of the CCL signal according to the abnormal threshold of the local change rate of the CCL signal;

[0076] The casing anomaly level classification module is used to determine the casing anomaly status score corresponding to each anomaly window based on the CCL signal anomaly interval, merge adjacent anomaly segments to obtain the merged casing anomaly status score, and determine the casing anomaly level based on the quantile threshold.

[0077] The present invention also provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed, the casing anomaly identification method based on the local change rate of the CCL signal as described above is implemented.

[0078] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects: the casing anomaly identification method and system based on the local change rate of the CCL signal provided by the present invention obtains the location information of the CCL signal coupling point based on the original CCL signal and the coupling point data, combines the depth information of the corresponding casing in the casing data table, reconstructs the CCL signal corresponding to each casing and splices it together to achieve depth correction, calculates the local change rate of the corrected CCL signal, determines the abnormal quantile of the local change rate, combines the historical data at the same depth, calculates the abnormal state score, and determines the casing state. It can be seen that the present invention realizes numerical feature extraction by analyzing the changes in the magnetic field distribution law implicit in the CCL signal, solving the problem in the prior art of relying on manual experience to interpret the signal and being unable to directly establish a mathematical correlation between the CCL signal and the casing deformation state.

[0079] The present invention proposes a CCL signal depth correction method based on signal reconstruction and splicing, which makes the CCL signal more consistent with the actual depth of the project and more accurately corresponds to the casing. In view of the correlation between the CCL signal and the casing status, four different CCL signal local change rate calculation methods are proposed, and a quantitative relationship between the casing status and the CCL signal is established. By determining the abnormal threshold value of different CCL signal local change rates, the depth of the casing corresponding to the abnormal segment of the CCL signal is identified. By classifying all abnormal segments, the abnormal status of the casing corresponding to the CCL signal is divided, which has guiding significance for on-site construction. The present invention realizes quantitative determination of the abnormal status of the casing based on the original CCL signal, without relying on other detection tools or manual experience intervention, and significantly improves the utilization rate of the CCL signal in casing health monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0081] Figure 1 This is a flow chart of a casing anomaly identification method based on the local change rate of the CCL signal according to the present invention;

[0082] Figure 2 The comparison diagram of the CCL signal before and after depth correction of the present invention, wherein (a) is the CCL signal before depth correction, and (b) is the CCL signal after depth correction;

[0083] Figure 3 This is a schematic diagram showing the original CCL signal according to an embodiment of the present invention;

[0084] Figure 4 This is a schematic diagram of the corresponding depths of the 1st to 137th signal segments according to an embodiment of the present invention;

[0085] Figure 5 This is a comparison diagram of the CCL signal before and after depth correction of the Xth perforation section of an oil and gas well according to an embodiment of the present invention, wherein (a) is the CCL signal before depth correction, and (b) is the CCL signal after depth correction;

[0086] Figure 6 This is a comparison chart of the casing anomaly level identification results of the Xth section of an oil and gas well in an embodiment of the present invention and the MIT24 multi-arm wellbore measurement results. DETAILED DESCRIPTION

[0087] The following are detailed examples of the present invention. These examples are intended to explain the present invention and are not to be construed as limiting the present invention. Where specific techniques or conditions are not specified in the examples, the techniques or conditions described in the literature in the art or in the product specifications were used. Materials or instruments used, where the manufacturer is not specified, are commercially available conventional products.

[0088] The present invention uses CCL signals collected by perforating guns to predict abnormal conditions of downhole casing. Parameters such as the CCL signal's magnetic field intensity attenuation gradient and waveform distortion are quantified through mathematical modeling. Four local change rates of the signals and abnormal quantile thresholds are defined to provide a quantitative determination criterion for casing abnormal conditions based solely on the original CCL signal. This method eliminates the need for other detection tools or manual intervention, significantly improving the utilization of CCL signals in casing health monitoring.

[0089] like Figure 1As shown, the casing abnormality identification method based on the local change rate of the CCL signal provided by the embodiment of the present invention includes the following steps:

[0090] Step 1: Data preparation:

[0091] Prepare the raw CCL signals for each perforation section of the oil and gas well, the information about the collar points in the CCL signals, and the casing data table for the well. When preparing the raw CCL signals, note that the signal sampling rate is 1%, meaning that one meter contains 100 sampling points, with a sampling interval of 0.01m.

[0092] Step 2: Depth Correction

[0093] Based on the casing data table, the position of the casing pup joint (short casing) is determined, and the position information of the CCL signal coupling point is obtained based on the original CCL signal and coupling point data. Combined with the depth information of the corresponding casing in the casing data table, the CCL signal corresponding to each casing is reconstructed and spliced ​​together to achieve depth correction; specifically, it includes:

[0094] S201, determine the position of the casing nipple:

[0095] By consulting the casing data table of the oil and gas well, determine the corresponding depth of the casing short section (i.e., short casing) in the depth section corresponding to the original CCL signal. If there is more than one, the one with the smallest depth shall prevail.

[0096] Among them, when perforating to determine the depth, the short casing is called a positioning pup joint. This is the most accurate method to determine the depth of each casing. The positioning pup joints do not appear continuously. Each well generally has only 2-3 positioning pup joints. Each casing in the subsequent steps refers to the normal casing.

[0097] S202, reconstructing the CCL signal according to the depth:

[0098] a. After the position of the special length casing is determined, the original CCL signal is cut into X segments through the collar point according to formula (1). Each segment corresponds to a casing, including sampling points, the depth sequence of the k-th segment signal after segmentation can be obtained by formula (1):

[0099]

[0100] Where: is the depth sequence corresponding to the kth segment signal after segmentation, is the original CCL signal, is the collar depth information of the CCL signal, X is the number of complete casings contained in the CCL signal, and the Split operator represents the segmentation operation of the original CCL signal. according to The depth in is divided into X segments, is the number of sampling points.

[0101] b. After dividing the signal, according to the position of the special casing, find the corresponding k-th segment signal corresponding to the depth interval of the casing in the casing table [ , ], and the corresponding casing length , satisfying the conditions shown in formula (2):

[0102]

[0103] Where: is the corrected starting depth of the casing, is the corrected end depth of the casing, is the starting depth of the kth casing, is the length of the sleeve;

[0104] In S202, mapping the original CCL signal to the depth interval to generate a reconstructed CCL signal specifically includes:

[0105] c. After determining the depth interval, map the original CCL signal to the depth interval and perform linear interpolation on the depth interval to ensure that the new depth sequence is generated. sampling points to generate a new depth sequence, as shown in formula (3):

[0106]

[0107] Where: is the reconstructed depth sequence, is the corrected starting depth of the casing, is the corrected end depth of the casing, is the number of sampling points;

[0108] After calculating the new depth sequence, the original CCL signal value remains unchanged, and the original depth sequence is mapped to the reconstructed depth sequence to generate the reconstructed CCL signal. The calculation is shown in formula (4):

[0109]

[0110] Where: is the CCL signal after reconstructing the jth segment, The operator does not change the CCL value, but only changes the mapping operation of the corresponding depth sequence, transforming the signal from the original depth sequence Map to On the new depth sequence after reconstruction;

[0111] d. Concatenate the reconstructed CCL signals according to depth to obtain a depth-corrected CCL signal, specifically including:

[0112] After reconstructing the CCL signal corresponding to each casing according to the process in steps ac, the reconstructed signals are spliced ​​together according to the depth to become the depth-corrected signal, as follows:

[0113]

[0114] Where: is the depth-corrected signal, is the CCL signal after reconstructing the jth segment, It represents the splicing of the CCL signals after the segmented X segments are reconstructed;

[0115] When splicing and reconstructing CCL signals, the depth of each connection point is saved to In the sequence, this sequence is the depth sequence corresponding to the tie point after depth correction.

[0116] After depth correction of the CCL signal, all coupling points are aligned with the positions in the casing data table, such as Figure 2 shown.

[0117] Step 3: Calculate the local rate of change:

[0118] S301, split sliding window:

[0119] The local change rate of the CCL signal after depth correction is calculated through a sliding window. The window length of the sliding window is N (for example, 200) sampling points, and the sliding step is M (for example, 100 sampling points). When the sliding window is used to extract the signal, when the sliding window contains When the point in the sequence is missing, the window is skipped. When calculating the local change rate of the CCL signal, when the first window and the last window do not contain the complete front and back windows, the current segment data is used as the data of the missing window;

[0120] S302, calculate the local change rate of the CCL signal:

[0121] The local change rates include the local signal fluctuation change rate LFSV, the local pulse change rate LPVR, the local mutation density change rate LMDVR, and the local depth association change rate LDAVR. Each local change rate is calculated through a sliding window, specifically including:

[0122]

[0123] Where: LFSV is the local signal fluctuation rate corresponding to the current window, is the standard deviation of the CCL signal data in the current window, is the standard deviation of the CCL signal data in the previous window, is the standard deviation of the CCL data in the next window;

[0124]

[0125] Where: is the impulse factor of the current window, is the impulse factor of the previous window, is the pulse factor of the next window, LPVR is the local pulse change rate corresponding to the current window, is the maximum value operation, N is the number of sampling points in the window, is the CCL data of the current window, is the CCL data of the next window, It is the CCL data of the previous window;

[0126]

[0127] Where: Represents the number of extreme points in the x sequence, Operator representation Symbols;

[0128]

[0129] Where: LMDVR is the local mutation density change rate corresponding to the current window, N is the number of sampling points in the window, The operator represents the number of extreme points in the x sequence, is the CCL data of the current window, is the CCL data of the previous window, It is the CCL data of the next window;

[0130]

[0131] Where: is the sum of the absolute values ​​of the adjacent differences of CCL data in a window, is the CCL data point, is the depth point corresponding to the CCL data point, It is the sum of the absolute values ​​of the adjacent differences of the corresponding depths of the CCL data in a window;

[0132]

[0133] Where: LDAVR is the local depth correlation change rate corresponding to the current window, is the sum of the absolute values ​​of the adjacent differences of the CCL data in the current window, It is the sum of the absolute values ​​of the adjacent differences of the CCL data corresponding to the depth of the current window. is the sum of the absolute values ​​of the adjacent differences of the CCL data of the previous window, It is the sum of the absolute values ​​of the adjacent differences of the CCL data corresponding to the depth of the previous window. is the sum of the absolute values ​​of the adjacent differences of the CCL data of the next window, It is the sum of the absolute values ​​of the adjacent differences in the depth corresponding to the CCL data of the next window.

[0134] The calculation results of all local change rates involved in this step are shown in Table 1:

[0135] Table 1 Calculation results of local change rate of CCL signal

[0136]

[0137] Step 4: Determine the abnormal interval:

[0138] Normalize the local change rate of the CCL signal, calculate the abnormal threshold of the normalized local change rate of the CCL signal, and determine the abnormal range of the CCL signal according to the abnormal threshold of the local change rate of the CCL signal; specifically, include:

[0139] S401, respectively normalizing the four local change rates: local signal fluctuation change rate LFSV, local pulse change rate LPVR, local mutation density change rate LMDVR, and local depth association change rate LDAVR;

[0140] S402, respectively calculating the abnormal threshold of each local change rate after normalization;

[0141] S403, filtering each local change rate according to the abnormal threshold of each local change rate, marking the window corresponding to the local change rate exceeding the abnormal threshold as an abnormal window, and setting the local change rate abnormal flag of the corresponding abnormal window to 1, the flag bit defaults to 0;

[0142] When any flag bit of the four local change rates corresponding to a window is 1, the depth segment corresponding to the window is an abnormal segment of the CCL signal, and all abnormal segments constitute the abnormal interval of the CCL signal.

[0143] Taking the local signal fluctuation rate (LFSV) as an example, step 1 demonstrates the LFSV normalization process. Step 2 demonstrates the calculation of the abnormality threshold for the CCL signal's LFSV. Step 3 demonstrates how to determine the abnormal range of the CCL signal using the abnormality threshold calculated in step 2. (Note: Each local signal fluctuation rate must go through the following three steps.)

[0144] ① Normalize the local change rate of the CCL signal

[0145] The local change rate of the CCL signal is normalized by formula (12) to generate the normalized local change rate;

[0146]

[0147] Where: is the normalized local change rate of the CCL signal, is the local signal fluctuation change rate of the CCL signal, is the mean of LFSV, is the standard deviation of LFSV.

[0148] ②Calculate the local change rate anomaly threshold

[0149] Normalized local signal fluctuation rate The abnormal threshold of the local signal fluctuation rate LFSV is calculated according to formula (13);

[0150]

[0151] Where: is the abnormal threshold of the local fluctuation change rate of the CCL signal, The operator is the infimum (that is, the smallest value that satisfies the conditions), is the set of local signal fluctuation change rates of the standardized CCL signal, is the normalized local change rate of the CCL signal.

[0152] ③Determine the abnormal range of CCL signal

[0153] The abnormal threshold of the local fluctuation change rate of the CCL signal calculated in step ② is used to filter the local fluctuation change rate of the CCL signal, and the window corresponding to the local change rate exceeding the threshold is marked as an abnormal window, and the local fluctuation change rate flag of the corresponding abnormal window is set to Set to 1, the flag bit defaults to 0.

[0154] When the flag bit of any window among the four local change rates corresponding to a window is 1, the depth segment corresponding to the window is an abnormal segment of the CCL signal. All abnormal segments are saved in a table, such as shown in Table 2.

[0155] Table 2 CCL signal abnormal segment results

[0156]

[0157] Step 5: Classification of casing abnormality levels:

[0158] Based on the abnormal interval of the CCL signal, the casing abnormal status score corresponding to each abnormal window is determined. Adjacent abnormal segments are merged to obtain the merged casing abnormal status score. The casing abnormality level is determined based on the quantile threshold. Specifically, the following are included:

[0159] S501, combining the data in Table 2, the casing abnormal state score corresponding to each abnormal window can be calculated :

[0160]

[0161] Where: is the casing abnormal state score, is the normalized local signal fluctuation rate, It is the abnormal flag of the local signal fluctuation rate. is the normalized local pulse change rate, It is the abnormal flag of the local pulse change rate. is the normalized local mutation density change rate, is the abnormal flag of the local mutation density change rate, is the local depth correlation change rate, It is the abnormal flag of the local depth correlation change rate;

[0162] S502: Merge adjacent abnormal segments and calculate the abnormal status score of the merged casing:

[0163] When the abnormal windows have depth overlap, the two windows are merged and the abnormal state score of the merged casing is calculated according to formula (15): :

[0164]

[0165] Where: is the casing abnormal state score corresponding to the merged abnormal segment, is the set of abnormal state scores of all original windows in the merged window, gather The maximum value in Operator Representation Set The average value of n is the number of overlapping segment windows; the merged abnormal segments are sorted to obtain complete abnormal segment information, as shown in Table 3;

[0166] Table 3 Abnormal state scores of abnormal segments of CCL signals

[0167]

[0168] S503, based on the combined casing abnormality score in step S502 According to the conditions shown in formula (16), the casing abnormality level is divided into three levels, namely severe abnormality, moderate abnormality, and slight abnormality:

[0169]

[0170] Where: Indicates the p-quantile of the data; in the above formula, corresponding to different levels, the value of p is 95%, 80%;

[0171] in, The calculation formula is as follows:

[0172]

[0173] Where: Represents the p-quantile of the data; pos represents the position of the p-quantile in the original data; Indicates that pos is rounded down.

[0174] By classifying the abnormal state scores of the CCL signal in step S503, the abnormal state of the casing is identified.

[0175] Specifically, the specific implementation of the present invention is described by taking the CCL signal of a perforation section of a casing-change well in a certain area as an example:

[0176] (1) Data preparation

[0177] Prepare the original CCL signal (4660m, 6195.7m) of a perforation section of an oil and gas well, the depth sequence of the collar point in the CCL signal, and the casing data table of the well. Confirm that the sampling interval of the CCL signal in this section is 0.01m. The collected data are as follows: Figure 3 And shown in Table 4.

[0178] Table 4 Sample table of raw data preparation

[0179]

[0180] (2) Depth correction

[0181] ①Determine the position of the casing nipple

[0182] The corresponding depth of the short casing was determined by consulting the casing data table of the oil and gas well. The short casing of the well was located at (5378.107m, 5380.317m).

[0183] ②Reconstruct CCL signal according to depth

[0184] After the position of the special length casing is determined, it can be seen from the prepared data that there are 138 coupling points in this section of CCL signal, with a total of 137 casings. According to formula (1), the original CCL signal is converted into .

[0185] According to the CCL coupling depth sequence, it is cut into 137 sections, each section corresponds to a casing. The actual depth of each casing is determined by the position of the short casing, and then the actual starting depth and ending depth of each section after segmentation are determined according to formula (2), as follows: Figure 4 shown.

[0186] Based on the above depth information, each depth segment is reconstructed using formula (3); formula (4) stitches all the reconstructed segments together.

[0187] The signal corresponding to each casing is reconstructed and spliced ​​in accordance with the above steps, and the spliced ​​signal becomes the depth-corrected signal as shown in the figure. Figure 5 shown.

[0188] (3) Calculate the local rate of change

[0189] ① Split sliding window

[0190] The local change rate of the depth-corrected CCL signal is calculated using a sliding window. The window length of the sliding window is 200 sampling points, which is subsequently represented by N, and the sliding step is 100 sampling points. When using a sliding window to extract the signal, when the window contains When the point in the sequence is reached, the window is skipped.

[0191] ②Calculate the local change rate of the CCL signal

[0192] Take the second window of the segmentation as an example to calculate its local change rate. If the data of the window before the first window is missing, the data of the first window is used as the previous window, and the window after the first window is the second window. Substitute the data of the first and second windows into formulas (6)-(11) to calculate the local change rates, including the local signal fluctuation change rate LFSV, the local pulse change rate LPVR, the local mutation density change rate LMDVR, and the local depth correlation change rate LDAVR.

[0193] The variables contained in , , ; , , ; , , ; , , .

[0194] Substitute the variables corresponding to the first window into formulas (6)-(11) to obtain formulas (17)-(20) to calculate the local change rate of the CCL signal of the first window. The local change rate of each window is calculated in turn, and the final results are shown in Table 5.

[0195]

[0196]

[0197]

[0198]

[0199] Table 5 Calculation results of local change rate of CCL signal

[0200]

[0201] (4) Determine the abnormal interval

[0202] The local change rate data in the above calculation result table are standardized respectively to obtain the corresponding standardized abnormal threshold value.

[0203] ① Normalize the local change rate of the CCL signal

[0204] Calculate the mean and standard deviation of each CCL signal's local change rate, and substitute them into formula (12) to standardize each local change rate. Based on the data in Table 5, the mean and standard deviation of each CCL signal's local change rate are calculated as follows:

[0205] , ; , ; , ; , The standardized results are shown in Table 6.

[0206] Table 6. Standardized results of local change rate of CCL signal in section X of an oil and gas well

[0207]

[0208] ②Calculate the local change rate anomaly threshold

[0209] Substitute the above standardized data into formula (13) to obtain the abnormal threshold of each standardized CCL signal local change rate. The abnormal threshold of the local signal fluctuation change rate LFSV is: , the abnormal threshold of the local pulse change rate LPVR is , the abnormal threshold of the local mutation density change rate LMDVR is , the abnormal threshold of the local depth correlation change rate LDAVR is .

[0210] ③Determine the abnormal range of CCL signal

[0211] The abnormal interval of the CCL signal is determined by calculating the abnormal threshold, as shown in Table 7.

[0212] Table 7 Results of abnormal CCL signal section in section X of an oil and gas well

[0213]

[0214] (5) Classification of casing abnormality

[0215] The casing abnormality score is calculated using the result table calculated in step (4), and the adjacent abnormal segments are merged to obtain the information of the final abnormal segment. The casing abnormality level is determined based on the abnormality score and the quantile threshold.

[0216] Calculate the casing abnormal state score corresponding to each abnormal window

[0217] Using the parameters in Table 7, the casing abnormality score of each window is calculated according to formula (14): Taking the first window as an example, , , , , , , , Substitute into formula (14) to obtain formula (21), and calculate the casing abnormal state score of the first abnormal window: is 2.48; then, the corresponding casing abnormal state score is calculated for each abnormal window in Table 7:

[0218]

[0219] Merge the adjacent abnormal segments and calculate the abnormal status score of the merged casing

[0220] Take the result of step ① and merge the two windows when the abnormal windows have depth overlap. Calculate the merged abnormal segment according to formula (15): ; Arrange the merged abnormal segments to obtain complete abnormal segment information, as shown in Table 8.

[0221] Table 8 Casing abnormal status score table

[0222]

[0223] Classification of casing abnormality levels

[0224] Based on the abnormal segment in step ② According to condition (16), the casing abnormality level is divided into three levels, namely severe abnormality, moderate abnormality, and slight abnormality. Columns, calculated of , , put it into condition (16), and get conditional formula (22);

[0225]

[0226] According to the classification results of condition (22), the abnormal grade result table of casing in section X of a certain oil and gas well was obtained, as shown in Table 9.

[0227] Table 9 Casing abnormality level results for the Xth section of an oil and gas well

[0228]

[0229] Compare the casing abnormality results identified by this method with the MIT24 multi-arm wellbore results of the well. Figure 6 As shown in the figure, it shows that the casing abnormality status result identified by this method is very close to the MIT24 multi-arm wellbore result of the well, and this method has guiding significance for the field.

[0230] The present invention further provides a casing anomaly identification system based on the local change rate of the CCL signal, which is used to execute the above-mentioned casing anomaly identification method based on the local change rate of the CCL signal, comprising:

[0231] The data preparation module is used to obtain the original CCL signal of each perforation section of the oil and gas well, the coupling point data in the CCL signal, and the casing data table of the oil and gas well;

[0232] The depth correction module is used to determine the position of the casing pup joint based on the casing data table, and obtain the position information of the CCL signal coupling point based on the original CCL signal and coupling point data. Combined with the depth information of the corresponding casing in the casing data table, the CCL signal corresponding to each casing is reconstructed and spliced ​​together to achieve depth correction;

[0233] A local change rate calculation module is used to calculate the local change rate of the CCL signal after depth correction through a sliding window to obtain the local change rate of the CCL signal;

[0234] an abnormal interval determination module, configured to normalize the local change rate of the CCL signal, calculate an abnormal threshold of the normalized local change rate of the CCL signal, and determine the abnormal interval of the CCL signal according to the abnormal threshold of the local change rate of the CCL signal;

[0235] The casing anomaly level classification module is used to determine the casing anomaly status score corresponding to each anomaly window based on the CCL signal anomaly interval, merge adjacent anomaly segments to obtain the merged casing anomaly status score, and determine the casing anomaly level based on the quantile threshold.

[0236] The present invention also provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed, the casing anomaly identification method based on the local change rate of the CCL signal as described above is implemented.

[0237] Matters not covered by the present invention are known technologies.

[0238] While various embodiments of the present invention have been described above, the above descriptions are intended to be illustrative, non-exhaustive, and not limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is selected to best explain the principles of the embodiments, their practical applications, or improvements to existing technologies, or to enable others skilled in the art to understand the embodiments disclosed herein.

Claims

1. A casing anomaly identification method based on the local change rate of CCL signal, characterized in that: The following steps are involved: S1, data preparation: obtain the original CCL signal of each perforation section of the oil and gas well, the coupling point data in the CCL signal, and the casing data table of the oil and gas well; S2, depth correction: Based on the casing data table, the position of the casing pup joint is determined, and the position information of the CCL signal coupling point is obtained based on the original CCL signal and coupling point data. Combined with the depth information of the corresponding casing in the casing data table, the CCL signal corresponding to each casing is reconstructed and spliced ​​together to achieve depth correction; S3, calculating the local change rate: calculating the local change rate of the depth-corrected CCL signal through a sliding window to obtain the local change rate of the CCL signal; S4, determining the abnormal interval: normalizing the local change rate of the CCL signal, calculating the abnormal threshold of the normalized local change rate of the CCL signal, and determining the abnormal interval of the CCL signal according to the abnormal threshold of the local change rate of the CCL signal; S5, casing anomaly classification: Based on the CCL signal anomaly interval, determine the casing anomaly status score corresponding to each anomaly window, merge adjacent anomaly segments to obtain the merged casing anomaly status score, and determine the casing anomaly level based on the quantile threshold; The step S3, calculating the local rate of change, specifically includes: S301, split sliding window: The window length of the sliding window is N sampling points, and the sliding step is M sampling points. When the sliding window is used to extract the signal, when the sliding window contains When the point in the sequence is skipped, the window is skipped. When calculating the local change rate of the CCL signal, when the first window and the last window do not contain the complete front and back windows, the current segment data is used as the data of the missing window; where N and M are both positive integers, and M <N; The sequence is the depth sequence corresponding to the tie point after depth correction; S302, calculate the local change rate of the CCL signal: The local change rates include the local signal fluctuation change rate LFSV, the local pulse change rate LPVR, the local mutation density change rate LMDVR, and the local depth association change rate LDAVR. Each local change rate is calculated through a sliding window.

2. The casing anomaly identification method based on the local change rate of the CCL signal according to claim 1 is characterized in that: S2, depth correction, specifically includes: S201, by consulting the casing data table of the oil and gas well, determining the corresponding depth of the casing pup joint in the depth section corresponding to the original CCL signal, and if there is more than one, the one with the smallest depth shall prevail; S202, the original CCL signal is converted through the coupling point Cut into several signal segments, each segment corresponds to a casing, and obtain the location information of the CCL signal coupling point; According to the position of the casing nipple, find the depth interval of the casing in the casing data table corresponding to the k-th segment signal, and map the original CCL signal to the depth interval to generate a reconstructed CCL signal; The reconstructed CCL signals are spliced ​​together according to the depth to obtain a depth-corrected CCL signal, wherein all the collar points in the depth-corrected CCL signal are aligned with the positions in the casing data table.

3. The casing anomaly identification method based on the local change rate of the CCL signal according to claim 2 is characterized in that: In the step S202, the original CCL signal is converted to Cut into X segments of signal, each segment of signal corresponds to a casing, including sampling points, the depth sequence of the kth segment signal after segmentation is obtained by formula (1): Where: is the depth sequence corresponding to the kth segment signal after segmentation, is the original CCL signal, is the collar depth information of the CCL signal, X is the number of complete casings contained in the CCL signal, and the Split operator represents the segmentation operation of the original CCL signal. according to The depth in is divided into X segments, is the number of sampling points.

4. The casing anomaly identification method based on the local change rate of the CCL signal according to claim 2 is characterized in that: In the above S202, the depth interval is expressed as [ , ], satisfying the conditions shown in formula (2): Where: is the corrected starting depth of the casing, is the corrected end depth of the casing, is the starting depth of the kth casing, is the length of the sleeve; In S202, mapping the original CCL signal to the depth interval to generate a reconstructed CCL signal specifically includes: Map the original CCL signal to the depth interval, perform linear interpolation on the depth interval, and ensure that the new depth sequence is generated. sampling points to generate a new depth sequence, as shown in formula (3): Where: is the reconstructed depth sequence, is the corrected starting depth of the casing, is the corrected end depth of the casing, is the number of sampling points; After calculating the new depth sequence, the original CCL signal value remains unchanged, and the original depth sequence is mapped to the reconstructed depth sequence to generate the reconstructed CCL signal. The calculation is shown in formula (4): Where: is the CCL signal after reconstructing the jth segment, The operator does not change the CCL value, but only changes the mapping operation of the corresponding depth sequence, transforming the signal from the original depth sequence Map to On the new depth sequence after reconstruction; In S202, the reconstructed CCL signals are spliced ​​together according to the depth to obtain the depth-corrected CCL signals, which specifically includes: After reconstructing the CCL signal corresponding to each casing according to the process in step S2, the reconstructed signals are spliced ​​together according to the depth to become the depth-corrected signal, as follows: Where: is the depth-corrected signal, is the CCL signal after reconstructing the jth segment, It represents the splicing of the CCL signals after the segmented X segments are reconstructed; When splicing and reconstructing CCL signals, the depth of each connection point is saved to In the sequence, this sequence is the depth sequence corresponding to the tie point after depth correction.

5. The casing anomaly identification method based on the local change rate of the CCL signal according to claim 4 is characterized in that: The step S302, calculating each local change rate through a sliding window, specifically includes: Where: LFSV is the local signal fluctuation rate corresponding to the current window, is the standard deviation of the CCL signal data in the current window, is the standard deviation of the CCL signal data in the previous window, is the standard deviation of the CCL data in the next window; Where: is the impulse factor of the current window, is the impulse factor of the previous window, is the pulse factor of the next window, LPVR is the local pulse change rate corresponding to the current window, is the maximum value operation, N is the number of sampling points in the window, is the CCL data of the current window, is the CCL data of the next window, It is the CCL data of the previous window; Where: Represents the number of extreme points in the x sequence, Operator representation Symbols; Where: LMDVR is the local mutation density change rate corresponding to the current window, N is the number of sampling points in the window, The operator represents the number of extreme points in the x sequence, is the CCL data of the current window, is the CCL data of the previous window, It is the CCL data of the next window; Where: is the sum of the absolute values ​​of the adjacent differences of CCL data in a window, is the CCL data point, is the depth point corresponding to the CCL data point, It is the sum of the absolute values ​​of the adjacent differences of the corresponding depths of the CCL data in a window; Where: LDAVR is the local depth correlation change rate corresponding to the current window, is the sum of the absolute values ​​of the adjacent differences of the CCL data in the current window, It is the sum of the absolute values ​​of the adjacent differences of the CCL data corresponding to the depth of the current window. is the sum of the absolute values ​​of the adjacent differences of the CCL data of the previous window, It is the sum of the absolute values ​​of the adjacent differences of the CCL data corresponding to the depth of the previous window. is the sum of the absolute values ​​of the adjacent differences of the CCL data of the next window, It is the sum of the absolute values ​​of the adjacent differences in the depth corresponding to the CCL data of the next window.

6. The casing anomaly identification method based on the local change rate of the CCL signal according to claim 5 is characterized in that: The step S4, determining the abnormal interval, specifically includes: S401, respectively normalizing the four local change rates: local signal fluctuation change rate LFSV, local pulse change rate LPVR, local mutation density change rate LMDVR, and local depth association change rate LDAVR; S402, respectively calculating the abnormal threshold of each local change rate after normalization; S403, filtering each local change rate according to the abnormal threshold of each local change rate, marking the window corresponding to the local change rate exceeding the abnormal threshold as an abnormal window, and setting the local change rate abnormal flag of the corresponding abnormal window to 1; When any flag bit of the four local change rates corresponding to a window is 1, the depth segment corresponding to the window is an abnormal segment of the CCL signal, and all abnormal segments constitute the abnormal interval of the CCL signal.

7. The casing anomaly identification method based on the local change rate of the CCL signal according to claim 6 is characterized in that: S5, casing abnormality classification, specifically includes: S501, calculating the casing abnormal state score corresponding to each abnormal window : Where: is the casing abnormal state score, is the normalized local signal fluctuation rate, It is the abnormal flag of the local signal fluctuation rate. is the normalized local pulse change rate, It is the abnormal flag of the local pulse change rate. is the normalized local mutation density change rate, is the abnormal flag of the local mutation density change rate, is the local depth correlation change rate, It is the abnormal flag of the local depth correlation change rate; S502: Merge adjacent abnormal segments and calculate the abnormal status score of the merged casing: When the abnormal windows have depth overlap, the two windows are merged and the abnormal state score of the merged casing is calculated according to formula (15): : Where: is the casing abnormal state score corresponding to the merged abnormal segment, is the set of abnormal state scores of all original windows in the merged window, gather The maximum value in Operator Representation Set The average value of n is the number of overlapping segment windows; sort out the merged abnormal segments to obtain complete abnormal segment information; S503, based on the combined casing abnormality score in step S502 According to the conditions shown in formula (16), the casing abnormality level is divided into three levels, namely severe abnormality, moderate abnormality, and slight abnormality: Where: Indicates the p-quantile of the data; in the above formula, corresponding to different levels, the value of p is 95%, 80%; By classifying the abnormal state scores of the CCL signal in step S503, the abnormal state of the casing is identified.

8. A casing anomaly recognition system based on the local change rate of CCL signals, applied to execute the casing anomaly recognition method based on the local change rate of CCL signals according to any one of claims 1 to 7, characterized in that: include: The data preparation module is used to obtain the original CCL signal of each perforation section of the oil and gas well, the coupling point data in the CCL signal, and the casing data table of the oil and gas well; The depth correction module is used to determine the position of the casing pup joint based on the casing data table, and obtain the position information of the CCL signal coupling point based on the original CCL signal and coupling point data. Combined with the depth information of the corresponding casing in the casing data table, the CCL signal corresponding to each casing is reconstructed and spliced ​​together to achieve depth correction; A local change rate calculation module is used to calculate the local change rate of the CCL signal after depth correction through a sliding window to obtain the local change rate of the CCL signal; an abnormal interval determination module, configured to normalize the local change rate of the CCL signal, calculate an abnormal threshold of the normalized local change rate of the CCL signal, and determine the abnormal interval of the CCL signal according to the abnormal threshold of the local change rate of the CCL signal; The casing anomaly level classification module is used to determine the casing anomaly status score corresponding to each anomaly window based on the CCL signal anomaly interval, merge adjacent anomaly segments to obtain the merged casing anomaly status score, and determine the casing anomaly level based on the quantile threshold.

Images