Casing abnormality diagnosis method and device based on historical self-comparison

Abstract

This invention proposes a casing anomaly diagnosis method and device based on historical self-comparison, relating to the field of intelligent diagnosis technology for engineering logging and integrity of oil and gas wells. The method includes: acquiring the casing coupling locator signal sequence of the target well during the earliest operational trip measurement phase, establishing a healthy baseline profile for the target well; performing peak detection on the healthy baseline profile to generate a benchmark coupling fingerprint database for the target well; acquiring the casing coupling locator signal sequence of subsequent operations, and systematically correcting the signal depth system to the benchmark depth system of the healthy baseline profile; calculating the historical difference ratio and new mutation energy point by point based on the aligned signal sequence and baseline, generating an inherent structure mask using multiple historical measurement data to identify newly occurring anomaly areas; extracting a specific feature vector of waveform morphology for each newly occurring anomaly area; inputting the specific feature vector into a classification decision model, and outputting the anomaly type label and corresponding confidence level for the newly occurring anomaly area.

Description

Technical Field

[0001] This invention relates to the field of oil and gas well engineering logging and integrity intelligent diagnosis technology, and in particular to a casing anomaly diagnosis method and device based on historical self-comparison. This invention utilizes data measured by a casing collar locator (CCL) in multiple operations, and automatically identifies casing anomalies and trends and diagnoses anomaly types through lateral comparison between different runs. Background Technology

[0002] During oil drilling, completion, and production, various instruments and tools (such as perforating guns, packers, and logging instruments) need to be lowered downhole. These tools are lowered via cables or wires. To determine the precise depth of these tools downhole, a magnetic measuring tool called a "casing coupling locator" (CCL) is typically used. Its principle is as follows: a permanent magnet and an induction coil are installed in the tool string. When the tool passes the junction between two casing sections (called the "coupling"), the sudden change in metal volume causes a change in the magnetic field, generating a voltage spike in the coil. By recording the depth at which these spikes appear, the location of the coupling can be determined, allowing for depth calibration.

[0003] The CCL signal is a voltage curve that varies with depth. A normal coupling will exhibit regular, large-amplitude spikes. Other features of the casing (such as centralizers and perforation orifices) will also leave signals with specific morphologies on the CCL curve.

[0004] In existing technologies, there are some automated tools for analyzing CCL data; for example, one tool works by comparing two CCL curves measured when the instrument is lowered and raised in the same operation, and using unsupervised algorithms (e.g., isolated forest) to find depth points where the two curves are inconsistent in shape, and outputting "anomaly score" or anomaly location.

[0005] However, existing technologies have at least the following drawbacks: 1. They only provide alarms, not diagnoses. They can only tell engineers "the signal is abnormal here," but cannot answer "what is this abnormality." For example, existing technologies cannot determine whether the abnormality is casing deformation, localized damage, perforation hole defects, or distortion of the coupling itself. After receiving an alarm, engineers still need to manually analyze historical data and guess the type of abnormality, causing the automation chain to break at the last step. 2. Single measurement cannot distinguish between newly occurring defects and historical legacies. Existing technologies only use data from the current operation and cannot know whether a certain signal feature is new to this operation or an inherent feature that existed before (e.g., permanent marks left by couplings or historical perforations). This leads to historical perforation sections being repeatedly falsely reported as "damaged," resulting in an extremely high false alarm rate. 3. They rely on external engineering data, which is not feasible for older wells. Some studies have attempted to introduce external data such as perforation design tables and casing coupling tables to assist in judgment, but the reality in oilfields is that a large number of engineering data for older wells are lost or incomplete, data formats from different eras are inconsistent, and manual maintenance costs are extremely high. Such solutions are almost impractical in reality. 4. Misuse of tension signals; Some techniques attempt to use cable tension to correct depth, but in highly deviated and horizontal wells, the forces on the cable are extremely complex (coupling of multiple fields such as gravity, friction, and pumping thrust), and there is no definite mapping relationship between tension and depth. Using tension to calculate depth offset will produce huge errors, leading to incorrect conclusions.

[0006] In summary, there is an urgent need for a technical solution that can overcome the above-mentioned defects, improve the casing anomaly detection mechanism, and achieve anomaly diagnosis. Summary of the Invention

[0007] To address the problems existing in the prior art, this invention proposes a method and device for diagnosing casing anomalies based on historical self-comparison. This invention completely abandons reliance on external engineering data, utilizing only historical CCL data (historical data and the latest measurement data) measured by the target well itself in different operation runs. Through horizontal self-comparison, it achieves automatic detection and type classification of newly occurring anomalies.

[0008] In a first aspect of the present invention, a method for diagnosing cannula anomalies based on historical self-comparison is proposed, the method comprising: Acquire the casing coupling locator signal sequence of the target well during the earliest workable trip measurement stage, and establish it as the health baseline profile of the target well; Peak detection is performed on the health baseline profile to generate a benchmark coupling fingerprint database for the target well; The casing coupling locator signal sequence measured during the subsequent operation of the target well during the lowering stage is obtained. The coupling peak is detected and matched with the reference coupling fingerprint database to determine the matching coupling peak. Based on the matching coupling peak, a depth mapping function is constructed to correct the signal depth system corresponding to the casing coupling locator signal sequence of this operation to the reference depth system of the healthy baseline profile. Based on the aligned signal sequence and baseline, the historical difference ratio and new mutation energy are calculated point by point. The inherent structure mask is generated using multiple historical measurement data to identify new abnormal regions. For newly emerging abnormal areas, a unique feature vector of waveform morphology is extracted. The unique feature vector is determined solely based on the morphology of the sleeve coupling positioner signal sequence. The specific feature vector is input into the classification decision model, and the abnormality type label and corresponding confidence level of the newly emerging abnormal region are output.

[0009] In a second aspect of the present invention, a cannula anomaly diagnostic device based on historical self-comparison is proposed, the device comprising: The data processing module is used to acquire the casing coupling locator signal sequence of the target well during the earliest workable trip measurement stage and establish it as the health baseline profile of the target well. The peak detection module is used to perform peak detection on the health baseline profile and generate a benchmark coupling fingerprint library for the target well. The depth alignment module is used to acquire the casing coupling locator signal sequence measured during the lowering stage of the target well in subsequent operations. After detection, the coupling peak is obtained and pattern matching is performed with the reference coupling fingerprint database to determine the matching coupling peak. Based on the matching coupling peak, a depth mapping function is constructed to systematically correct the signal depth system corresponding to the casing coupling locator signal sequence of this operation to the reference depth system where the healthy baseline profile is located. The new anomaly detection module is used to calculate the historical difference ratio and new mutation energy point by point based on the aligned signal sequence and baseline, and generate an intrinsic structure mask using multiple historical measurement data to identify new anomaly regions. The feature extraction module is used to extract a unique feature vector of waveform morphology for newly emerging abnormal areas. The unique feature vector is determined solely based on the morphology of the sleeve coupling locator signal sequence. The anomaly type classification module is used to input the specific feature vector into the classification decision model and output the anomaly type label and corresponding confidence level of the newly occurring anomaly region.

[0010] In a third aspect of the present invention, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement a casing anomaly diagnosis method based on historical self-comparison.

[0011] In a fourth aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing a computer program, which, when executed by a processor, implements a casing anomaly diagnosis method based on historical self-comparison.

[0012] In a fifth aspect of the present invention, a computer program product is provided, the computer program product comprising a computer program that, when executed by a processor, implements a casing anomaly diagnosis method based on historical self-comparison.

[0013] The casing anomaly diagnosis method and device proposed in this invention establishes a healthy baseline using the CCL signal from the earliest logging run of the target well, and achieves high-precision depth alignment by using coupling peak matching and segmented depth mapping. It combines historical difference ratio, new mutation energy, and inherent structure masking to accurately shield inherent structure interference. It can achieve efficient identification and automatic classification of new casing anomalies by relying solely on the waveform morphology characteristics of the CCL itself. No external engineering data is required throughout the process, which effectively reduces the false alarm rate and improves the accuracy and reliability of anomaly diagnosis. At the same time, it can support anomaly evolution analysis and activity classification, and is suitable for intelligent monitoring and safety assessment of the entire life cycle of casing based on multiple logging runs of old wells. Attached Figure Description

[0014] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0015] Figure 1 This is a schematic flowchart of a cannula anomaly diagnosis method based on historical self-comparison according to an embodiment of the present invention.

[0016] Figure 2 This is a schematic diagram of the evolutionary analysis and activity classification process according to an embodiment of the present invention.

[0017] Figure 3 This is a schematic flowchart of the perforation aftereffect evaluation according to an embodiment of the present invention.

[0018] Figure 4 This is a schematic diagram of the tension-assisted discrimination process according to an embodiment of the present invention.

[0019] Figure 5 This is a schematic diagram of the architecture of a cannula anomaly diagnosis device based on historical self-comparison according to an embodiment of the present invention.

[0020] Figure 6 This is a schematic diagram of a computer device structure according to an embodiment of the present invention. Detailed Implementation

[0021] The principles and spirit of the invention will now be described with reference to several exemplary embodiments. It should be understood that these embodiments are given merely to enable those skilled in the art to better understand and implement the invention, and are not intended to limit the scope of the invention in any way. Rather, these embodiments are provided to make this disclosure more thorough and complete, and to fully convey the scope of this disclosure to those skilled in the art.

[0022] Those skilled in the art will recognize that embodiments of the present invention can be implemented as a system, apparatus, device, method, or computer program product. Therefore, this disclosure can be specifically implemented in the following forms: entirely hardware, entirely software (including firmware, resident software, microcode, etc.), or a combination of hardware and software.

[0023] According to an embodiment of the present invention, a method and device for diagnosing casing anomalies based on historical self-comparison are proposed, which relates to the field of intelligent diagnosis technology for engineering logging and integrity of oil and gas wells.

[0024] The principles and spirit of the present invention will be explained in detail below with reference to several representative embodiments.

[0025] Figure 1 This is a schematic flowchart of a cannula anomaly diagnosis method based on historical self-comparison according to an embodiment of the present invention. Figure 1 As shown, the method includes: S101, acquire the casing coupling locator signal sequence of the target well during the earliest workable trip measurement stage, and establish it as the health baseline profile of the target well; S102, perform peak detection on the health baseline profile to generate a benchmark coupling fingerprint database for the target well; S103, acquire the casing coupling locator signal sequence measured during the subsequent operation of the target well during the lowering stage, obtain the coupling peak after detection, and perform pattern matching with the reference coupling fingerprint database to determine the matching coupling peak. Based on the matching coupling peak, construct a depth mapping function to correct the signal depth system corresponding to the casing coupling locator signal sequence of this operation to the reference depth system where the healthy baseline profile is located. S104. Based on the aligned signal sequence and baseline, calculate the historical difference ratio and new mutation energy point by point, generate an intrinsic structure mask using multiple historical measurement data, and identify new abnormal regions. S105, For newly emerging abnormal areas, extract the exclusive feature vector of the waveform morphology. The exclusive feature vector is determined only based on the morphology of the sleeve coupling positioner signal sequence itself. S106, input the specific feature vector into the classification decision model, and output the anomaly type label and corresponding confidence level of the newly emerging anomaly region.

[0026] To provide a clearer explanation of the above-mentioned method for diagnosing cannula anomalies based on historical self-comparison, each step will be explained in detail below.

[0027] In one embodiment, for S101, the casing coupling locator signal sequence of the target well during the earliest workable trip measurement phase is acquired, and a healthy baseline profile of the target well is established. The specific process includes: The signal sequence of the sleeve coupling positioner is preprocessed; the preprocessing method includes at least power frequency noise filtering, baseline drift correction and amplitude normalization. The earliest operation with acceptable signal quality is automatically identified and used as the health baseline; among them, the signal sequence of the decentralization phase is used first.

[0028] In practical applications, CCL data for all operations throughout the target well's entire lifecycle (including data from the lowering and raising phases) and cable tension data during pumping operations (if applicable) are collected. Basic data processing, such as filtering, noise reduction, and baseline correction, is then applied to the data.

[0029] In one embodiment, for S102, peak detection is performed on the healthy baseline profile to generate a benchmark coupling fingerprint database for the target well. The specific process includes: Adaptive threshold peak detection is performed on the healthy baseline profile, wherein the threshold is set as a predetermined proportion of the maximum amplitude of the healthy baseline profile; All detected local maxima points are used as coupling peaks, and the depth sequence based on the coupling peaks is used to construct the benchmark coupling fingerprint database, which serves as the permanent anchor point for depth alignment in all subsequent passes.

[0030] In practical applications, the earliest operation with acceptable signal quality is automatically identified (prioritizing data from the downsizing phase) and used as the "health baseline." Adaptive peak detection is performed on the health baseline to extract the depth of all coupling peaks, forming the well's "baseline coupling fingerprint database." The baseline coupling fingerprint database serves as the permanent anchor point for all subsequent depth alignments, derived entirely from the well's own historical data, without requiring any external coupling tables.

[0031] In the preprocessing and baseline pass identification process, the adaptive threshold coupling peak detection uses the following calculation formula:

[0032] In the formula, The CCL signal representing the baseline pass is a function of depth d and is measured in millivolts (mV).

[0033] This indicates an adaptive threshold.

[0034] This represents the maximum amplitude of the entire reference signal.

[0035] 0.4 is an empirical coefficient used to adaptively determine the threshold, avoiding the applicability problem of a fixed threshold in different wells.

[0036] The detected coupling peak depth sequence will serve as a permanent depth benchmark (baseline coupling fingerprint database) for the well.

[0037] Physical significance: Automatically identifies the location of all couplings from the baseline run, without the need for any external coupling table. These coupling peaks are permanent features that are permanently fixed downhole, and all subsequent runs must align to them.

[0038] In one embodiment, for S103, the casing coupling locator signal sequence measured during the lowering stage of the target well in subsequent operations is acquired. The coupling peak is detected and pattern matched with the reference coupling fingerprint database to determine the matching coupling peak. A depth mapping function is constructed based on the matching coupling peak to systematically correct the signal depth system corresponding to the casing coupling locator signal sequence of this operation to the reference depth system of the healthy baseline profile. The specific process includes: For the casing coupling positioner signal sequence during the lowering stage of each subsequent operation, coupling peak detection is performed, and pattern matching is performed with the benchmark coupling fingerprint database to determine at least three pairs of matching coupling peaks. A depth mapping function is constructed using the matching points corresponding to the coupling peaks. The depth mapping function uses piecewise polynomial interpolation and forces alignment at the matching points to map the depth of all sampling points in this operation onto the reference depth system.

[0039] In practical applications, for each subsequent CCL (Cable Clamping) data delivery, clamp peak detection is performed, and then pattern matching is conducted with the baseline clamp fingerprint database to identify at least three matching clamp peaks. Using these matching points, a depth mapping function is constructed to map the depth of all sampling points in the current data onto the baseline depth system. This eliminates depth deviations between different deliveries caused by cable stretching, instrument differences, etc. The depth mapping function employs piecewise polynomial interpolation and enforces precise alignment at the matching points.

[0040] Depth mapping function (piecewise cubic spline interpolation): ,satisfy (Force precise alignment at the matching coupling point) In the formula, Indicates the depth in the reference depth system; This represents the depth of the i-th matching coupling in the reference depth system.

[0041] Indicates the measured depth; This indicates the measured depth of the same coupling in the current trip.

[0042] This represents the depth mapping function, which maps the baseline depth to the depth of the current pass. It is obtained by performing piecewise cubic spline interpolation on the matching points, ensuring precise alignment at the matching points and a smooth, continuous curve throughout.

[0043] In this embodiment of the invention, forced alignment is performed because, considering that the couplings are unique, permanently fixed, have significant signals, and can be automatically identified, using them as anchor points can reliably eliminate depth errors caused by cable stretching, instrument differences, etc. Segmented interpolation can handle non-uniform stretching (e.g., different stretching rates at different cable depths).

[0044] In one embodiment, for S104, based on the aligned signal sequence and baseline, the historical difference ratio and the energy of newly occurring mutations are calculated point by point. An intrinsic structure mask is generated using multiple historical measurement data to identify newly occurring anomalous regions. The specific process includes: Based on multiple historical measurement data, signal features that consistently exist in all measurement operations are identified. An inherent structure mask is established based on these signal features and marked as a permanent structure. In subsequent analysis, these signal features will no longer be regarded as potential anomalies. The aligned casing coupling locator signal sequence of the current trip is compared point by point with the baseline to calculate the historical difference ratio and the energy of new mutations. When the historical difference ratio and the energy of new mutations both exceed the corresponding preset thresholds, and the comparison point is not covered by the inherent structure mask, it is determined to be a new anomaly.

[0045] In practical applications, multiple historical measurements (including the baseline run and subsequent runs) are used to identify signal features that consistently exist across all measurements (e.g., couplings, historical perforations, permanent stabilizers). These depth points are then mapped using an "inherent structure mask," marking them as permanent structures so they are no longer considered potential anomalies in subsequent analyses.

[0046] Inherent structure mask The definition of is: If for all historical passes i=1,…,N, then...

[0047] Otherwise, it is 0.

[0048] In the formula, N represents the total number of historical trips involved in the analysis (at least 3 trips yield better results).

[0049] This represents the signal value at depth d for the i-th historical measurement (all signals have been aligned to the reference depth system).

[0050] This represents the median of all historical signals at that depth point. The median is more robust than the mean and is more resistant to interference from individual outliers.

[0051] This represents the standard deviation of the baseline signal within a window near depth d, indicating the normal fluctuation range of the signal at that point under healthy conditions. The window size is typically 20-30 sampling points (approximately 1-1.5 meters).

[0052] This represents the threshold parameter, with a recommended value of 2.0 to 3.0. It indicates the allowable fluctuation factor. If the signal at a certain depth point remains stable near the median throughout all historical data (with fluctuations less than 2 to 3 times the normal fluctuation amplitude), it is considered a permanent structure.

[0053] Physical meaning: This indicates that depth d belongs to an inherent structure (e.g., couplings, historical perforations, permanent stabilizers), and these points are permanently masked in subsequent analyses and will no longer be considered as new anomalies.

[0054] This indicates that the signal at this point has historically been unstable or has undergone new changes, requiring further analysis.

[0055] The current pass's downlinked CCL signal (already aligned) is compared point-by-point with the baseline, and two metrics are calculated: historical difference ratio and new mutation energy. Only when both metrics exceed the threshold, and the point is not covered by the inherent structure mask, is it identified as a new anomalous point.

[0056] For historical difference ratio :

[0057] In the formula, This represents the historical difference ratio corresponding to depth d; This represents the signal value (in mV) after depth alignment and amplitude normalization of the previous CCL drop.

[0058] This represents the baseline signal value (mV).

[0059] This represents the standard deviation of the baseline signal within a local window near depth d, reflecting the normal fluctuation range at that location. The window size is typically 20-30 sampling points (1-1.5 meters).

[0060] To represent a very small positive number, to prevent the denominator from being zero, for example... .

[0061] Physical meaning: It measures how much the current signal deviates from the baseline's normal fluctuations. For example... =5 indicates that the difference between the current signal and the baseline is 5 times the normal fluctuation of the baseline itself, which is a significant deviation.

[0062] For newly emerging mutation energy :

[0063] In the formula, This represents the energy of the new mutation corresponding to depth d.

[0064] L represents the half-width of the integral, typically taken as 10-20 sampling points (0.5-1 meter). The integration interval is [dL, d+L]).

[0065] The integral represents the sum of the energy of the difference signals within a local window. The squaring operation highlights large differences and downplays small fluctuations.

[0066] Physical meaning: The total energy of the newly emerging signal was quantified. For abrupt events such as localized damage, the energy of the differential signal increases dramatically.

[0067] New anomaly determination rules: if , and If the depth d is then marked as a newly discovered anomaly.

[0068] in, This represents the historical difference ratio threshold, with a recommended value of 3.0 (i.e., the difference exceeds 3 times the normal fluctuation).

[0069] The recommended threshold for the energy of newly emerging mutations is the 90th percentile of the energy distribution in the non-coiled area in all historical measurements of the well. Alternatively, a fixed empirical value (such as 0.5 times the average energy of the baseline signal) can be used as the initial value.

[0070] Logical explanation: A new anomaly can only be identified if it simultaneously meets the following criteria: significant difference, strong energy, and does not belong to the inherent structure. This dual-indicator approach prevents misjudgment.

[0071] In one embodiment, for S105, for newly emerging abnormal areas, a unique feature vector of waveform morphology is extracted, and the unique feature vector is determined solely based on the morphology of the sleeve coupling locator signal sequence.

[0072] The specific feature vector includes at least one of the following: deformation feature group, damage feature group, perforation feature group, and coupling feature group.

[0073] In one embodiment, for S106, the specific feature vector is input into the classification decision model, and the anomaly type label and corresponding confidence level of the newly occurring anomaly region are output.

[0074] The classification decision model employs a rule engine and a machine learning classifier; wherein, the rule engine is used to identify the abnormal types of coupling distortion, and the machine learning classifier is used to identify the abnormal types of deformation, damage, and perforation.

[0075] The following features are calculated for newly emerging abnormal regions and rely solely on the morphology of the CCL signal itself, without requiring any external data.

[0076] 1. For the casing "deformation" characteristic group:

[0077] Window half width Typically, 15 to 25 sampling points (0.75 to 1.25 meters) are used, and the median value can resist noise. This feature measures local baseline shift, and the value is significantly greater than 0 in deformed areas.

[0078] (Hilbert transform for envelope)

[0079] Envelope smoothness: The deformed area has a smooth envelope with a large mean / standard deviation ratio (usually >5.0); the damaged area has a small ratio.

[0080] (Discrete wavelet transform, db4 wavelet basis)

[0081] D1 represents the highest frequency detail coefficient. High-frequency energy percentage: The distorted signal is predominantly low-frequency. F def3 <0.2; high frequency of the damaged signal.

[0082] Deformation determination: F def1 >0.1·A global ,and F def2 >5.0, and F def3 <0.2.

[0083] 2. For the "damaged or perforated" characteristic group of the casing:

[0084] W L W R : Full width at half maximum (FWHM) of the left and right sides of the pulse. Asymmetry: The damaged pulse is asymmetrical due to magnetic field distortion, |F dam1 |>0.2.

[0085] (Continuous wavelet transform, Mexican hat wavelet basis)

[0086] scale a From 1 to 5, This represents the peak position. The singularity index measures the severity of abrupt changes; for a broken signal, this value is significantly higher than that of the coupling.

[0087]

[0088] W 50 W represents the full width at half maximum (FWHM) of the current pulse. collar,50 This indicates the average half-width of the normal coupling in the well. The damaged pulse is narrower, (F dam3 <0.6.

[0089] Damage assessment: |F dam1 |>0.2 and F dam2 >1.5 × (mean singularity index of coupling) and F dam3 <0.6.

[0090] 3. For the "perforation hole" feature group (without designed perforation density): (Peak spacing coefficient) { } represents the depth difference between consecutive peak values. The perforation spacing is uniform, with a coefficient of variation <0.1.

[0091] i≠j (waveform consistency index)

[0092] P i Let be a local waveform segment of the i-th peak, and corr be the Pearson correlation coefficient. The perforation waveforms are highly consistent, F... perf3 >0.85.

[0093] Perforation detection: F perf1 <0.1 and Fperf3 >0.85. (Completely abandoning the need to design spectral characteristics with dense apertures)

[0094] 4. For the "coupling" feature group (no coupling table): (Peak amplitude ratio) A peak This represents the current peak amplitude; A collar,std This represents the average amplitude of the normal coupling in this well (extracted from the baseline). Normal coupling F col1 ≈1.0.

[0095] (Peak symmetry)

[0096] A left A right The depth (or amplitude) of the left and right valleys. Symmetry close to 1 indicates good condition.

[0097] Coupling determination: F col1 >0.7 and F col2 >0.85. (Discard the depth matching feature that requires a coupling table)

[0098] In practical applications, for newly emerging abnormal areas, extract their waveform morphology-specific feature vectors (e.g., four major categories and twelve items), input these features into a pre-trained classifier (which can be a rule engine plus a machine learning model), and output the abnormality type label (deformation, damage, perforation, coupling distortion) and confidence level.

[0099] It should be noted that the features analyzed in this embodiment (e.g., waveform asymmetry, symmetry, high-frequency energy ratio, etc.) do not involve directional information. Since CCL currently operates with a single probe and lacks directionality, all information in this embodiment is extracted from the time-series signal of the single probe and does not involve spatial direction information.

[0100] In practical applications, multi-probe scenarios may occur (i.e., using directional CCL instruments). Since this invention relies only on the time-depth sequence characteristics of the CCL signal and does not depend on whether the probe is directional, the method of this invention is applicable regardless of whether a single-probe CCL or a directional multi-probe CCL is used. If a directional CCL is used, spatial features can be further expanded based on the current morphological features.

[0101] In one embodiment, reference Figure 2 The method also includes: S201, For well sections with more than three operations, analyze the evolution trend of abnormal conditions over time; S202, determine the anomaly level according to the evolution trend, and determine the maintenance priority recommendation according to the anomaly level.

[0102] In practical applications, for well sections with more than three historical measurements, the evolution trend of anomalies over time is analyzed. The absolute rate of change, acceleration exponent, and the time of the first occurrence of abrupt changes are calculated. Anomalies are classified into three levels: "rapid deterioration," "slow development," and "long-term stability," and maintenance priority recommendations are given according to different levels.

[0103] (Absolute rate of change)

[0104] t N For the current time, t 1 represents the time when the anomaly first occurred (or the earliest historical time). Unit: amplitude / year.

[0105] (Acceleration Index)

[0106] At least three consecutive measurements are required. Positive values ​​indicate accelerated deterioration, while negative values ​​indicate a slowdown.

[0107] (Moment of sudden change)

[0108] Find the moment when the change is greatest between two consecutive measurements, i.e. the time when the anomaly first appears significantly.

[0109] Activity levels are classified as follows: Level I (High-risk activity): F time1 >0.3 base / year and F time2 >0; Level II (Slow Development): 0.1 base / Year <F time1 ≤0.3 base / Year; Level III (Stable): F time1 ≤0.1 base / years ago, and the mutation occurred much earlier than the present.

[0110] For different operating modes, the present invention further provides methods for analysis, evaluation, and auxiliary judgment.

[0111] For details, please refer to Figure 3 This is a schematic diagram of the perforation aftereffect evaluation process according to an embodiment of the present invention. (Reference) Figure 3 Specific methods include: S301, after the perforation operation, acquire the casing coupling positioner signal sequence during the lifting phase of this operation; S302, using the reference coupling fingerprint database as an anchor point, align the casing coupling locator signal sequence of the lifting stage of the current operation with the casing coupling locator signal sequence of the lifting stage before any perforation in the historical operation to the same reference depth system, and determine the signal characteristics of the perforation section. S303 outputs the confidence level of the perforation effect and an early warning of unexpected damage based on the signal characteristics of the perforation section.

[0112] In practical applications, after a perforation operation, the CCL (Center Clamping Line) data of the lifting phase of that operation is acquired and compared with the lifting CCL of a previous perforation operation (again, depth alignment is performed first). Signal features of the perforated section are extracted, such as signal clutter level (baseline thickening index), perforation peak detection rate, and coupling distortion index, and the perforation effect confidence score and unexpected damage warning are output. This module is completely decoupled from the core diagnostics, and the results are displayed side by side.

[0113] Specifically, after perforation, CCL measurement data that reflects the state of the perforated section is acquired. This data can be obtained through any of the following methods: In vertical wells, after perforation, the instrument string is lowered below the perforation section and then raised back up through the perforation section (i.e., "post-perforation logging"). If an instrument design with the CCL located below the perforation gun is adopted, the data can be directly obtained by lifting the instrument after perforation. Alternatively, other CCL measurement methods that can obtain the state of the casing after perforation can be used.

[0114] It should be noted that currently, the CCL (Cyclic Cordless Tracking) is primarily positioned above the perforating gun and connected via cable. This arrangement is mainly to prevent cable damage and data transmission interruption should the CCL be placed below the perforating gun. Therefore, in vertical wells, after perforation, the instrument string is typically lowered manually below the perforation point and then pulled back up to measure the post-perforation effect. In horizontal wells, because pumping is required, the CCL cannot be directly lowered, and the data is not actually passed through the perforated hole during the pull-up operation, making post-perforation effect measurement currently impossible. Therefore, in practical applications, a wireless transmission method could be considered. Using an instrument with the CCL positioned below the perforating gun, data can be collected wirelessly during the pull-up operation. This avoids data transmission failure due to cable damage and meets the requirements for post-perforation effect measurement in different types of wells.

[0115] For perforation-specific features, at least the following should be included: perforation section baseline thickening index, perforation peak detection rate, and coupling distortion index.

[0116] (Baseline thickening index of perforation section)

[0117] Sperf For the signal within the perforation section, S ref This represents the signal from a nearby non-perforation section. After perforation, the signal becomes cluttered, with a variance ratio > 1.

[0118] (Perforation peak detection rate)

[0119] N expected Based on the perforation section length and typical perforation density (e.g., 16 perforations / meter), precise design is not required; N detected The number of perforation peaks for automatic identification.

[0120] (Harness distortion index) To ensure the symmetry of the coupling after perforation, This represents the symmetry of the same coupling before the perforation. A calculation result > 0.3 indicates that the coupling may be damaged.

[0121] For details, please refer to Figure 4 This is a schematic flowchart illustrating the tension-assisted determination process according to an embodiment of the present invention. (See reference...) Figure 4 Specific methods include: S401, during pumping operations, monitors the signal sequence of the sleeve coupling positioner and the cable tension in real time; S402, When the signal sequence of the sleeve coupling positioner disappears and the cable tension drops instantaneously by more than a preset ratio, the obstruction and jamming detection is triggered; If the device is in a lowering state, it is considered to have encountered resistance; if it is in an upward state and the tension rise rate exceeds the preset slope threshold, it is considered to have encountered a jam.

[0122] In practical applications, the results of tension-assisted discrimination are for on-site reference only and are not used for any depth correction or anomaly classification.

[0123] For tension-assisted features, at least the following should be included: descent depth upon encountering resistance and upward climbing slope.

[0124] (Depth of descent upon encountering resistance)

[0125] T base T represents the average tension (kN) in the stable pumping section before encountering resistance. drop This represents the tension trough value (kN) at the instant of encountering resistance. A calculation result > 0.2 indicates significant resistance.

[0126] (Increase in elevation, unit: kN / m)

[0127] T pickup\_start Indicates the initial tension; T max This indicates the peak tension during the lifting process; This indicates the upward distance from the starting point to the peak value. A slope exceeding 0.5~1.5 kN / m indicates a jam (adjustment required based on cable type).

[0128] It should be noted that the following "three no's" principle applies: tension signals are not used for any form of depth offset calculation, are not used for depth alignment correction, and are not included in anomaly type classification.

[0129] This invention determines the "health changes" of the sheath by comparing it to itself using its "historical records".

[0130] Specifically, the CCL signal measured in the earliest operation of the well (such as cementing quality logging of a new well or the first perforation operation of an old well) is used as the "health baseline". This baseline implies all the inherent structural features of the well (couplings, centralizers, formation background magnetic field, etc.).

[0131] In subsequent operations, the CCL signals measured are depth-aligned and compared point by point with the baseline to identify areas where the signal changes significantly. These areas are the "new anomalies".

[0132] For newly emerging anomalies, extract the mathematical characteristics of their waveform morphology (e.g., whether it is a slow drift or a sudden pulse, whether it is symmetrical or asymmetrical, whether it is periodic or isolated), and determine the type of anomaly based on these characteristics: deformation, damage, perforation, coupling distortion, etc.

[0133] At the same time, by using multiple historical measurements, we can analyze the trend of anomalies over time (whether they are deteriorating at an accelerated pace) and provide recommendations on maintenance priorities.

[0134] For perforation operations, a separate "perforation post-effect evaluation" module is designed to automatically evaluate the perforation effect and potential damage by comparing the CCL after perforation with the CCL before historical perforation.

[0135] For cable tension signals, their sole purpose is clearly defined: to be used only in pumping operations for real-time identification of obstruction / jamming, and strictly prohibited from being used for depth correction or as a basis for anomaly classification.

[0136] This invention is innovative in the following aspects: 1. Zero external data dependence: Completely abandons all external engineering data such as coupling table, perforation design, casing program, wellbore trajectory, tension model, etc., and completes diagnosis using only the target well's own CCL historical data.

[0137] 2. Earliest Trip Forced Benchmark Anchoring: The earliest CCL (Continuous Collapsed Limit) of the target well is used as the health baseline, which includes all static characteristics to achieve a true "individual health record".

[0138] 3. Self-generated benchmark coupling fingerprint library: Automatically extracts coupling peaks from baseline data to establish a permanent depth reference system without any manual marking.

[0139] 4. Automatic generation of inherent structure mask: Through multiple historical measurements, it automatically identifies and permanently masks permanent features such as couplings and historical perforations, completely avoiding false alarms.

[0140] 5. A dedicated morphological feature system based on physical mechanisms: Calculateable mathematical features are designed for deformation, damage, perforation, and coupling distortion to achieve refined automatic classification of abnormal types.

[0141] 6. Independent evaluation of post-perforation effects: For the first time, the comparison of CCL lifting after perforation is incorporated into the automated system, realizing a quantitative evaluation of perforation effect and damage.

[0142] 7. Precise positioning of tension signals: The tension signal is clearly defined to be used only for resistance / jamming detection and is strictly prohibited from being used for depth correction, thus solving the industry problem of misuse of tension in horizontal wells.

[0143] Referring to Table 1, compared with the prior art, the present invention has at least the following technical effects.

[0144] Table 1. Technical effects of the present invention compared to the prior art

[0145] It should be noted that although the operation of the method of the present invention has been described in a specific order in the above embodiments and figures, this does not require or imply that the operations must be performed in that specific order, or that all the operations shown must be performed to achieve the desired result. Additionally or alternatively, certain steps may be omitted, multiple steps may be combined into one step, and / or one step may be broken down into multiple steps.

[0146] After introducing the method of exemplary embodiments of the present invention, the following references are made. Figure 5 This invention provides an exemplary embodiment of a cannula anomaly diagnostic device based on historical self-comparison.

[0147] The implementation of the cannula anomaly diagnostic device based on historical self-comparison can refer to the implementation of the above method, and the repeated parts will not be described again. The term "module" or "unit" used below can refer to a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0148] Based on the same inventive concept, this invention also proposes a cannula anomaly diagnostic device based on historical self-comparison, such as... Figure 5 As shown, the device includes: Data processing module 510 is used to acquire the casing coupling locator signal sequence of the target well during the earliest workable trip measurement stage and establish it as the health baseline profile of the target well. Peak detection module 520 is used to perform peak detection on the health baseline profile and generate a benchmark coupling fingerprint library for the target well; The depth alignment module 530 is used to acquire the casing coupling locator signal sequence measured during the lowering stage of the target well in subsequent operations, obtain the coupling peak after detection, and perform pattern matching with the reference coupling fingerprint database to determine the matching coupling peak. Based on the matching coupling peak, a depth mapping function is constructed to systematically correct the signal depth system corresponding to the casing coupling locator signal sequence of this operation to the reference depth system where the healthy baseline profile is located. The new anomaly detection module 540 is used to calculate the historical difference ratio and new mutation energy point by point based on the aligned signal sequence and baseline, generate an intrinsic structure mask using multiple historical measurement data, and identify new anomaly regions. The feature extraction module 550 is used to extract a unique feature vector of waveform morphology for newly emerging abnormal areas. The unique feature vector is determined solely based on the morphology of the sleeve coupling locator signal sequence. The anomaly type classification module 560 is used to input the exclusive feature vector into the classification decision model and output the anomaly type label and corresponding confidence level of the newly occurring anomaly region.

[0149] It should be noted that although several modules of the cannula anomaly diagnostic device based on historical self-comparison are mentioned in the detailed description above, this division is merely exemplary and not mandatory. In fact, according to embodiments of the present invention, the features and functions of two or more modules described above can be embodied in a single module. Conversely, the features and functions of a single module described above can be further divided and embodied by multiple modules.

[0150] Based on the aforementioned inventive concept, such as Figure 6 As shown, the present invention also proposes a computer device 600, including a memory 610, a processor 620, and a computer program 630 stored in the memory 610 and executable on the processor 620. When the processor 620 executes the computer program 630, it implements the aforementioned method for diagnosing cannula anomalies based on historical self-comparison.

[0151] Based on the aforementioned inventive concept, the present invention proposes a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the aforementioned sleeve anomaly diagnosis method based on historical self-comparison.

[0152] Based on the aforementioned inventive concept, the present invention proposes a computer program product, which includes a computer program that, when executed by a processor, implements a casing anomaly diagnosis method based on historical self-comparison.

[0153] The casing anomaly diagnosis method and device proposed in this invention establishes a healthy baseline using the CCL signal from the earliest logging run of the target well, and achieves high-precision depth alignment by using coupling peak matching and segmented depth mapping. It combines historical difference ratio, new mutation energy, and inherent structure masking to accurately shield inherent structure interference. It can achieve efficient identification and automatic classification of new casing anomalies by relying solely on the waveform morphology characteristics of the CCL itself. No external engineering data is required throughout the process, which effectively reduces the false alarm rate and improves the accuracy and reliability of anomaly diagnosis. At the same time, it can support anomaly evolution analysis and activity classification, and is suitable for intelligent monitoring and safety assessment of the entire life cycle of casing based on multiple logging runs of old wells.

[0154] The acquisition, storage, use, and processing of data in the technical solution of this invention all comply with the relevant provisions of laws and regulations.

[0155] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0156] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0157] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0158] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0159] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for diagnosing cannula anomalies based on historical self-comparison, characterized in that, The method includes: Acquire the casing coupling locator signal sequence of the target well during the earliest workable trip measurement stage, and establish it as the health baseline profile of the target well; Peak detection is performed on the health baseline profile to generate a benchmark coupling fingerprint database for the target well; The casing coupling locator signal sequence measured during the subsequent operation of the target well during the lowering stage is obtained. The coupling peak is detected and matched with the reference coupling fingerprint database to determine the matching coupling peak. Based on the matching coupling peak, a depth mapping function is constructed to correct the signal depth system corresponding to the casing coupling locator signal sequence of this operation to the reference depth system of the healthy baseline profile. Based on the aligned signal sequence and baseline, the historical difference ratio and new mutation energy are calculated point by point. The inherent structure mask is generated using multiple historical measurement data to identify new abnormal regions. For newly emerging abnormal areas, a unique feature vector of waveform morphology is extracted. The unique feature vector is determined solely based on the morphology of the sleeve coupling positioner signal sequence. The specific feature vector is input into the classification decision model, and the abnormality type label and corresponding confidence level of the newly emerging abnormal region are output.

2. The method for diagnosing cannula anomalies based on historical self-comparison according to claim 1, characterized in that, Acquire the casing coupling locator signal sequence of the target well during the earliest workable trip measurement phase, and establish it as the healthy baseline profile of the target well, including: The signal sequence of the sleeve coupling positioner is preprocessed; the preprocessing method includes at least power frequency noise filtering, baseline drift correction and amplitude normalization. The earliest operation with acceptable signal quality is automatically identified and used as the health baseline; among them, the signal sequence of the decentralization phase is used first.

3. The cannula anomaly diagnosis method based on historical self-comparison according to claim 1, characterized in that, Peak detection is performed on the healthy baseline profile to generate a benchmark coupling fingerprint database for the target well, including: Adaptive threshold peak detection is performed on the healthy baseline profile, wherein the threshold is set as a predetermined proportion of the maximum amplitude of the healthy baseline profile; All detected local maxima points are used as coupling peaks, and the depth sequence based on the coupling peaks is used to construct the benchmark coupling fingerprint database, which serves as the permanent anchor point for depth alignment in all subsequent passes.

4. The cannula anomaly diagnosis method based on historical self-comparison according to claim 1, characterized in that, The casing coupling locator signal sequence measured during the lowering phase of the target well in subsequent operations is acquired. The coupling peak is detected and pattern matched with the baseline coupling fingerprint database to determine the matching coupling peak. Based on the matching coupling peak, a depth mapping function is constructed to systematically correct the signal depth system corresponding to the casing coupling locator signal sequence of this operation to the baseline depth system of the healthy baseline profile, including: For the casing coupling positioner signal sequence during the lowering stage of each subsequent operation, coupling peak detection is performed, and pattern matching is performed with the benchmark coupling fingerprint database to determine at least three pairs of matching coupling peaks. A depth mapping function is constructed using the matching points corresponding to the coupling peaks. The depth mapping function uses piecewise polynomial interpolation and forces alignment at the matching points to map the depth of all sampling points in this operation onto the reference depth system.

5. The method for diagnosing cannula anomalies based on historical self-comparison according to claim 1, characterized in that, Based on the aligned signal sequence and baseline, the historical difference ratio and new mutation energy are calculated point by point. An intrinsic structure mask is generated using multiple historical measurement data to identify newly occurring anomalous regions, including: Based on multiple historical measurement data, signal features that consistently exist in all measurement operations are identified. An inherent structure mask is established based on these signal features and marked as a permanent structure. In subsequent analysis, these signal features will no longer be regarded as potential anomalies. The aligned casing coupling locator signal sequence of the current trip is compared point by point with the baseline to calculate the historical difference ratio and the energy of new mutations. When the historical difference ratio and the energy of new mutations both exceed the corresponding preset thresholds, and the comparison point is not covered by the inherent structure mask, it is determined to be a new anomaly.

6. The method for diagnosing cannula anomalies based on historical self-comparison according to claim 1, characterized in that, The specific feature vector includes at least one of the following: deformation feature group, damage feature group, perforation feature group, and coupling feature group; The specific feature vector is input into the classification decision model, which outputs the anomaly type label and corresponding confidence score of the newly discovered anomaly region, including: The classification decision model employs a rule engine and a machine learning classifier; wherein, the rule engine is used to identify the abnormal types of coupling distortion, and the machine learning classifier is used to identify the abnormal types of deformation, damage, and perforation.

7. The method for diagnosing cannula anomalies based on historical self-comparison according to claim 1, characterized in that, The method also includes: For well sections with more than three operations, analyze the evolution trend of abnormal conditions over time; The anomaly level is determined according to the evolution trend, and maintenance priority recommendations are made according to the anomaly level.

8. The method for diagnosing cannula anomalies based on historical self-comparison according to claim 1, characterized in that, The method also includes: After the perforation operation, obtain the casing coupling positioner signal sequence during the lifting phase of this operation; Using the reference coupling fingerprint database as an anchor point, the casing coupling locator signal sequence of the lifting stage in the current operation is aligned with the casing coupling locator signal sequence of the lifting stage before any perforation in the historical operation to the same reference depth system to determine the signal characteristics of the perforation section. Based on the signal characteristics of the perforation section, output the confidence level of the perforation effect and the early warning of unexpected damage.

9. The method for diagnosing cannula anomalies based on historical self-comparison according to claim 1, characterized in that, The method also includes: During pumping operations, the signal sequence of the sleeve coupling positioner and the cable tension are monitored in real time. When the signal sequence of the sleeve coupling positioner disappears and the cable tension drops instantaneously by more than a preset ratio, the obstruction and jamming judgment is triggered. If the sleeve coupling positioner is in a lowering state, it is judged as obstruction. If the sleeve coupling positioner is in an upward state and the tension rise rate exceeds the preset slope threshold, it is judged as jamming.

10. A casing anomaly diagnostic device based on historical self-comparison, characterized in that, The device includes: The data processing module is used to acquire the casing coupling locator signal sequence of the target well during the earliest workable trip measurement stage and establish it as the health baseline profile of the target well. The peak detection module is used to perform peak detection on the health baseline profile and generate a benchmark coupling fingerprint library for the target well. The depth alignment module is used to acquire the casing coupling locator signal sequence measured during the lowering stage of the target well in subsequent operations. After detection, the coupling peak is obtained and pattern matching is performed with the reference coupling fingerprint database to determine the matching coupling peak. Based on the matching coupling peak, a depth mapping function is constructed to systematically correct the signal depth system corresponding to the casing coupling locator signal sequence of this operation to the reference depth system where the healthy baseline profile is located. The new anomaly detection module is used to calculate the historical difference ratio and new mutation energy point by point based on the aligned signal sequence and baseline, and generate an intrinsic structure mask using multiple historical measurement data to identify new anomaly regions. The feature extraction module is used to extract a unique feature vector of waveform morphology for newly emerging abnormal areas. The unique feature vector is determined solely based on the morphology of the sleeve coupling locator signal sequence. The anomaly type classification module is used to input the specific feature vector into the classification decision model and output the anomaly type label and corresponding confidence level of the newly occurring anomaly region.

Images