Constant false alarm rate-based downhole casing nondestructive testing method and device and medium

Abstract

The present disclosure provides a constant false alarm rate-based downhole casing nondestructive testing method, device and medium. The method can include: obtaining a detection threshold according to a preset false alarm probability and background noise power of a target pulsed eddy current array detection signal, wherein the false alarm probability refers to the probability that the detection system incorrectly determines the presence of a damaged target when a non-damaged target is present in the downhole casing; screening the target pulsed eddy current array detection signal based on the detection threshold to obtain a downhole casing damage detection result signal; and determining the position of the downhole casing damage according to the downhole casing damage detection result signal. Through this technical solution, the detection threshold can be dynamically adjusted, and the position of the downhole casing damage can be accurately located, thereby improving the efficiency and accuracy of downhole casing damage detection.

Description

Technical Field

[0001] This disclosure relates to the field of oil extraction safety assurance technology, and in particular to a method, device and medium for non-destructive testing of downhole casing based on constant false alarm rate. Background Technology

[0002] To ensure the safety of oil and gas wells, extend the service life of downhole casing, and improve the efficiency of oil and gas extraction, it is necessary to detect downhole casing damage, thereby ensuring the reliability and sustainability of oil and gas extraction.

[0003] Currently, the detection method using a Uniform Linear Multicoil Array (ULMA) to generate transient electromagnetic signals is commonly used to detect damage in downhole casing. Although ULMA has proven effective in non-destructive testing of downhole casing, the radial dimension of the detection probe cannot be too large due to the limitation of the wellbore radius. This prevents the ULMA from achieving the required degrees of freedom within the downhole casing, creating a significant technical bottleneck in its detection performance. Therefore, the signal strength can only be improved by increasing the number of coil turns in the probe's longitudinal direction. However, as the longitudinal dimension of the probe increases, the detection resolution and accuracy of the surrounding medium decrease.

[0004] Furthermore, non-destructive testing of downhole casing typically employs fixed-threshold detection technology to determine the presence of damage targets. This involves setting a fixed threshold based on prior information; when the power amplitude of the signal to be tested exceeds the threshold, a damage target is considered to exist at the corresponding location, while signals that do not reach the threshold are considered interference signals. However, the complex and variable downhole environment, with its diverse clutter types and the influence of internal instrument noise, can easily cause the detection threshold amplitude to rise, leading to a sharp increase in the false alarm probability and a decline in detection performance. Summary of the Invention

[0005] In view of this, this disclosure aims to provide a method, device and medium for non-destructive testing of downhole casing based on a constant false alarm rate, which can dynamically adjust the detection threshold and accurately locate the damage position of the downhole casing, thereby improving the efficiency and accuracy of downhole casing damage detection.

[0006] The technical solution disclosed herein is implemented as follows:

[0007] In a first aspect, this disclosure provides a method for non-destructive testing of downhole casing based on a constant false alarm rate, the method comprising:

[0008] The detection threshold is obtained based on the preset false alarm probability and the background noise power of the target pulse eddy current array detection signal. The false alarm probability refers to the probability that the detection system incorrectly judges the presence of a damaged target when there is no damaged target in the downhole casing.

[0009] The target pulse eddy array detection signal is filtered based on the detection threshold to obtain the damage detection result signal of the downhole casing;

[0010] The location of the downhole casing damage is determined based on the damage detection result signal of the downhole casing.

[0011] Secondly, this disclosure provides a non-destructive testing device for downhole casing based on a constant false alarm rate. The device includes: an acquisition section, a screening section, and a determination section; wherein,

[0012] The acquisition section is configured to acquire a detection threshold based on a preset false alarm probability and the background noise power of the target pulse eddy current array detection signal. The false alarm probability refers to the probability that the detection system incorrectly judges the presence of a damaged target when there is no damaged target in the downhole casing.

[0013] The filtering section is configured to filter the target pulse eddy array detection signal based on the detection threshold to obtain the damage detection result signal of the downhole casing;

[0014] The determining component is configured to determine the location of damage to the downhole casing based on the damage detection result signal of the downhole casing.

[0015] Thirdly, this disclosure provides a computing device, the computing device comprising: a memory and a processor; wherein,

[0016] The memory is used to store computer programs that can run on the processor;

[0017] The processor is configured to execute the downhole casing non-destructive testing method based on constant false alarm rate as described in the first aspect when running the computer program.

[0018] Fourthly, this disclosure provides a computer storage medium storing at least one instruction, which is executed by a processor to implement the non-destructive testing method for downhole casing based on constant false alarm rate as described in the first aspect.

[0019] This disclosure provides a method, apparatus, and medium for non-destructive testing of downhole casing based on a constant false alarm rate. By pre-setting the false alarm probability, the detection system maintains a stable false alarm level under different noise environments, meaning the probability of incorrectly identifying a damaged target when there is none remains constant. The detection threshold is dynamically adjusted based on the background noise power, enabling the detection system to adapt to different working environments and improving the flexibility and accuracy of detection. By comparing the signal power of the detected unit in the target pulse eddy current array detection signal with the detection threshold to filter the signal, the detection result signal that truly indicates damage is effectively identified, ensuring that only truly abnormal signals are further analyzed and processed, reducing false alarms and missed alarms, and improving the accuracy of damage detection. Attached Figure Description

[0020] Figure 1 This is a downhole pulsed eddy current detection system based on ULMA;

[0021] Figure 2 A flowchart of a non-destructive testing method for downhole casing based on a constant false alarm rate is provided in this disclosure;

[0022] Figure 3 This is a schematic diagram of the signal processing flow of a downhole transient electromagnetic detection system provided in this disclosure;

[0023] Figure 4 A schematic diagram illustrating the detection principle of the mean constant false alarm rate provided in this disclosure;

[0024] Figure 5 This is a schematic diagram of a sleeve structure provided in this disclosure;

[0025] Figure 6 A waveform diagram of the original induced electromotive force of an early signal from a receiving coil provided in this disclosure;

[0026] Figure 7 A waveform diagram of the original induced electromotive force of a receiving coil for a later signal provided in this disclosure;

[0027] Figure 8 For the purposes of this disclosure and Figure 6 The corresponding induced electromotive force weighted waveform of the early signal of the receiving coil based on the LCMV criterion;

[0028] Figure 9 For the purposes of this disclosure and Figure 7 The corresponding induced electromotive force weighted waveform of the late signal of the receiving coil based on the LCMV criterion;

[0029] Figure 10 This is a diagram showing the detection results of an early signal using the CA-CFAR algorithm provided in this disclosure;

[0030] Figure 11 This is a diagram showing the detection results of an early signal using the SO-CFAR algorithm provided in this disclosure;

[0031] Figure 12 This is a diagram showing the detection results of an early signal using the GO-CFAR algorithm provided in this disclosure;

[0032] Figure 13 This is a diagram showing the detection results of late-stage signals using the CA-CFAR algorithm provided in this disclosure;

[0033] Figure 14 This is a diagram showing the detection results of late-stage signals using the SO-CFAR algorithm provided in this disclosure;

[0034] Figure 15 This is a diagram showing the detection results of late-stage signals using the GO-CFAR algorithm provided in this disclosure;

[0035] Figure 16 This is a schematic diagram of the downhole casing non-destructive testing device based on constant false alarm rate provided in this disclosure;

[0036] Figure 17 This is a schematic diagram of the hardware structure of a computing device provided in this disclosure. Detailed Implementation

[0037] The technical solutions in this disclosure will now be clearly and completely described with reference to the accompanying drawings.

[0038] Downhole casing is an important component of oil and gas well structures. Its function is to maintain wellbore stability, prevent oil and gas leakage, and ensure the safety of oil and gas extraction. Therefore, regular damage inspection of downhole casing is crucial.

[0039] Currently, the detection method for non-destructive testing of bushings typically employs a uniform linear multi-coil array to generate pulsed eddy current (PEC) signals. See also... Figure 1 It shows a structural diagram of a downhole pulsed eddy current detection system 100 based on ULMA. Note that... Figure 1 The illustrated detection system 100 is only one example of a possible system, and embodiments of this disclosure can be implemented in any of various systems as needed. The detection system 100 comprises, from the inside out, a probe core, an air layer, a tool housing, well fluid, pipelines, a cement sheath, and a formation. The detection system 100 is coaxially cylindrical and has a total of [missing information - likely a number of components]. J The layers, from the inside out, are numbered layer 1, 2, ... J To measure the wall thickness of the downhole casing, the electrical parameters of all media layers and the inner diameter of the casing can be set to be fixed. For any layer, for example, the first... jThe electrical and geometric parameters of the layer, , , and They represent the first j The relative permittivity, relative permeability, conductivity, and layer radius of the layer, among which, .

[0040] For example, such as Figure 1 The detection system 100 shown has a probe core including a transmitting coil and There are 10 receiving coils, and the spacing between each receiving coil, i.e., the receiving array element, is 1 / 2. Each receiving coil contains a coaxial soft magnetic core. The transmitting coil and... Each receiving coil is located in the second layer, i.e., the air layer, and all receiving coils have the same number of turns. Assuming that the diameter of all coils is small enough, the active region only contains the second layer, and the induced electromagnetic force (EMF) of the receiving coils is related to the vertical component of the magnetic field (core) of the first layer; all other layers, such as well fluid, cement sheath, casing, and formation, are considered passive regions.

[0041] It's important to clarify that active and passive regions refer to the different areas of interaction between the transmitting coil and the surrounding medium in a downhole pulsed eddy current detection system. The active region is the area where the transmitting coil is located; that is, the area where the magnetic field generated by the transmitting coil can significantly influence and induce eddy currents. In downhole casing non-destructive testing, the active region typically refers to the transmitting coil and the casing area within a certain range around it. The magnetic field changes within this region are sufficient to generate detectable eddy currents in the casing. In other words, the signal generated by the transmitting coil in the active region directly induces an electromagnetic response, thereby generating a primary alternating magnetic field used to detect the integrity and damage of the downhole casing. The passive region refers to the area that does not directly participate in the interaction with the electromagnetic field generated by the transmitting coil, or whose response to the magnetic field generated by the transmitting coil is weak and insufficient to generate a detectable signal. That is, there is no direct excitation source in the passive region. However, in some examples, a secondary magnetic field can be generated by the electromagnetic induction between the medium and the eddy currents. In downhole casing non-destructive testing, the passive region can include well fluid, pipelines, cement sheaths, and formations. Passive regions are typically considered background or signal interference areas. The purpose of non-destructive testing (NDT) of downhole casing is to distinguish signals from the background in order to identify and locate damage or anomalies in the casing. Specifically, in this disclosure, signals within passive regions are not considered invalid information. NDT technology utilizes the electromagnetic response of the metal casing, while the layer containing the casing, i.e., the pipeline layer, is located in the passive region. Based on the valid information obtained from the pipeline layer, the wall thickness variation at different locations of the casing is inferred, ultimately determining the damage and achieving NDT. Therefore, distinguishing between active and passive regions is crucial for improving the accuracy and reliability of testing. By optimizing the electromagnetic characteristics of the active region, damage or defects in downhole casing can be detected more effectively.

[0042] for Figure 1 The detection system 100 shown works by applying a bipolar step signal or a slanted step signal to the transmitting coil. During the interval between the primary pulse magnetic field shutdowns, the receiving coil observes the secondary eddy current field. By measuring the change of the secondary field over time at various points after the power outage, the geoelectric characteristics at different depths or radial locations downhole can be obtained. During downhole detection, since the conductivity of metals such as casing is much greater than that of air, cement sheath, and formation, abnormal damage to the downhole casing can be interpreted by inverting the induced electromotive force received by the receiving coil.

[0043] It should be noted that pulsed eddy current technology and its derivatives can detect targets without physical contact. Furthermore, they can identify cracks, pits, and corrosion in metallic infrastructure without surface preparation. However, unlike surface measurements, the non-destructive testing performance of downhole casing is significantly affected by the high-temperature and high-pressure environment downhole, resulting in a substantial decrease in the performance of the detection system.

[0044] Combination Figure 1The detection system 100 shown, although effective in non-destructive testing of casing using a uniform linear multi-coil array, suffers from limitations due to the borehole radius. This restricts the radial dimension of the detection probe, preventing it from meeting the required degrees of freedom within the casing and hindering the breakthrough of its detection performance. Therefore, this disclosure aims to provide a technical solution for non-destructive testing of downhole casing based on a constant false alarm rate (CFAR). Utilizing adaptive threshold detection technology, this solution maintains a constant false alarm probability while improving the accuracy and efficiency of downhole casing damage detection. Specifically, see [link to relevant documentation]. Figure 2 The diagram shows a flowchart of a non-destructive testing method for downhole casing based on a constant false alarm rate, which may include steps S201 to S203.

[0045] S201: Obtain the detection threshold based on the preset false alarm probability and the background noise power of the target pulse eddy current array detection signal.

[0046] False alarm probability refers to the probability that a detection system incorrectly identifies a damaged target as present when no damaged target exists in the downhole casing. The target pulse eddy current array detection signal is obtained by phase compensation of the initial pulse eddy current array detection signal based on the transmit and receive distances of the transmitting coil and each receiving coil. The initial pulse eddy current array detection signal is acquired by a downhole pulse eddy current detection system based on ULMA. Background noise power is the total power of all signals other than the target pulse eddy current array detection signal. This signal can be internal noise of electronic equipment, environmental interference, or reflections or scattering of other non-target signals. The background noise power level directly affects the performance of the detection system, such as the detection probability and false alarm probability. Therefore, in complex and noisy environments, background noise power is crucial for improving the accuracy and reliability of the detection system. The detection threshold is a pre-set or pre-calculated threshold used to determine whether a damaged target or anomaly in the target pulse eddy current array detection signal has been detected. It can dynamically change according to the background noise power of the target pulse eddy current array detection signal.

[0047] S202: Based on the detection threshold, the target pulse eddy array detection signal is filtered to obtain the damage detection result signal of the downhole casing.

[0048] Based on the background noise power and the preset false alarm probability, the detection threshold is determined by comparing the target pulse eddy current array detection signal received by the uniform linear multi-coil array, and filtering out the signal that exceeds the detection threshold. This signal is considered as the damage detection result signal, that is, the signal containing the damage target, thereby realizing the identification of downhole casing damage.

[0049] S203: Determine the location of downhole casing damage based on the damage detection results signal.

[0050] For example, the initial pulsed eddy current array detection signal acquired by each receiving coil of the uniform multi-coil array is phase-compensated according to the transmit and receive distances of the transmitting coil and each receiving coil to construct a standard detection curve characterizing the undamaged downhole casing. When damage exists in the downhole casing, the wall thickness at the damaged location will show a significant increase in amplitude or width. Combined with CFAR technology, the specific location and duration of damage in the damage detection result signal can be accurately identified, thereby accurately determining the location and severity of the downhole casing damage.

[0051] As described in the above scheme, by pre-setting the false alarm probability, the detection system maintains a stable false alarm level under different noise environments, meaning the probability of incorrectly identifying a damaged target when there is none remains constant. The detection threshold is dynamically adjusted based on the background noise power, enabling the detection system to adapt to different working environments and improving detection flexibility and accuracy. By comparing the signal power of the detected unit in the target pulse eddy current array detection signal with the detection threshold, signal filtering is performed, effectively identifying the detection result signal that truly indicates damage. This ensures that only truly abnormal signals are further analyzed and processed, reducing false alarms and missed alarms, and improving the accuracy of damage detection.

[0052] against Figure 2 The technical solution shown, exemplarily, can be implemented in the following manner: Figure 3 The signal processing flow of the detection system shown is used to achieve this. Specifically, as... Figure 3As shown, firstly, a downhole columnar layered geological model is constructed to obtain the structural and transmission characteristics of the pulsed eddy current signal, and the geological model parameters are output. These parameters include structural parameters, such as the number of layers in the columnar model, and electromagnetic parameters, such as the conductivity, permeability, and relative permittivity of each layer. Secondly, as shown in steps S301 to S304, specifically, as in step S301, a ULMA-based downhole pulsed eddy current array signal model is constructed based on the geological model parameters and the detection system parameters. Specifically, the geological model parameters and the detection system parameters are input, where the detection system parameters can include structural and electromagnetic parameters, such as the position of the transmitting coil, the position of the receiving coil, the number of detection array elements, the coil radius, the number of coil turns, and the magnetic core parameters, to construct the ULMA-based downhole pulsed eddy current signal model. The received vector representation of the induced electromotive force of the M receiving coils is output. Then, as in step S302, the initial pulsed eddy current array detection signal is phase-compensated using the Linearly Constrained Minimum Variance (LCMV) criterion to improve the reliability of target detection. Specifically, based on the weighting principle of array signals, phase compensation is performed on the initial pulse eddy current array detection signal according to the LCMV criterion to eliminate errors caused by different transmitting and receiving distances between different receiving array elements, thereby improving the accuracy of the detection signal. Finally, as in steps S303 to S304, the phase-compensated array signal, i.e., the target pulse eddy current array detection signal, is used to filter the signal using the CFAR detection principle to achieve the detection of the number of downhole casing damages and the precise location of the damage.

[0053] In some examples, the target pulsed eddy current array detection signal is obtained by phase compensation of the initial pulsed eddy current array detection signal based on the transmit and receive distances of the transmitting coil and each receiving coil. For example, combined with... Figure 1 The detection system 100 shown first acquires the initial pulse eddy array detection signal of the downhole casing by using each receiving coil of the uniform multi-coil array; then, it performs phase compensation on the initial pulse eddy array detection signal according to the transmission and reception distance between each receiving coil and the transmitting coil in the uniform multi-coil array to obtain the target pulse eddy array detection signal.

[0054] In some examples, each receiving coil of a uniform multi-coil array is used to acquire the initial pulsed eddy current array detection signal of the downhole casing. Specifically, the process of acquiring the initial pulsed eddy current array detection signal is to derive the expression of the electromagnetic response of each receiving element based on the constructed ULMA-based downhole pulsed eddy current signal model; for the detection signal of each receiving element, the cumulative summation form is transformed into a vector product form to form the array detection signal, and the electromagnetic response process is characterized using matrix form.

[0055] For the example above, specifically, in order to measure the wall thickness of the downhole metal casing, assuming that the electrical parameters of each layer and the inner diameter of the metal casing are fixed, the induced electromotive force of the m-th receiving coil in the time domain can be expressed as:

[0056] (1)

[0057] in, ; ; t and t of These refer to the observation time and the shutdown time, respectively. z m and d m The first m The transmit / receive distance of each receiving array element and the wall thickness of the metal sleeve at the observation position; S and D s These are the order and integral coefficients of the Gaver-Sstehfest inverse Laplace transform, respectively. P and Q , respectively, are the Legendre polynomial orders of the integral coefficients A and the zero B; C1 is the reflection coefficient of the innermost layer; x j and λ The introduced variable, and satisfies , It is a zeroth-order variant of the Bessel function of the first kind.

[0058] As can be seen from equation (1), the term containing casing wall thickness information and , and Related, and Only with Therefore, the cumulative summation of the induced electromotive force shown in equation (1) can be transformed into a vector multiplication form, which can be expressed as:

[0059] (2)

[0060] in, ; ; ; It is the transpose of a vector or matrix; This refers to the shutdown time.

[0061] Because the length of the receiving coil array is much smaller than the distance between adjacent thickness variations, the thickness of the metal casing at each receiving coil observation position is the same along the wellbore axis. ,in, By combining the noise generated by each receiving coil, it can be... All EMFs in a receiver coil are represented as a single receiver vector:

[0062] (3)

[0063] in, ; The vector is Gaussian white noise. , and The number of turns of the receiving coil and the transmitting coil. The magnitude of the excitation current; , These are the parameters introduced during the equivalent transformation.

[0064] It should be noted that the above equation (3) is the initial pulse eddy array detection signal obtained based on the ULMA downhole pulse eddy signal model.

[0065] After obtaining the initial pulsed eddy current array detection signal, due to... Figure 1 In the detection system 100 shown, the receiving coils and transmitting coils are at different transmission and reception distances, and the phase changes with the transmission and reception distance. Directly adding the induced electromotive forces of all receiving coils will produce errors. In order to eliminate the errors caused by the different transmission and reception distances between different receiving array elements, phase compensation and noise suppression are required. Based on the weighting principle of array signals, phase compensation is performed on the initial pulsed eddy current array detection signal according to the LCMV criterion.

[0066] For example, phase compensation is performed on the initial pulsed eddy current array detection signal based on the transmit and receive distances between each receiving coil and transmitting coil in a uniform multi-coil array to obtain the target pulsed eddy current array detection signal. Specifically, the LCMV algorithm is used to perform phase compensation on the initial pulsed eddy current array detection signal by calculating the optimal weighting factor. First, the initial pulsed eddy current array detection signal is optimized by minimizing the variance of the formula, and the vector is adjusted according to the LCMV criterion. V Apply constraints to vectors V The weighted term corresponds to the case where the transmit / receive distance is 0. Specifically, construct the LCMV function:

[0067] (4)

[0068] in, , ; st These are the constraints that the system of equations must satisfy.

[0069] Secondly, the Lagrange multiplier method is adopted, that is, the optimal weights are obtained by introducing Lagrange operators. These optimal weights, i.e., weighting factors, can be used to constrain vectors.V This ensures that the corresponding transmit / receive distance is zero and minimizes the variance of the minimized formula. Specifically, the optimal weights are solved using the Lagrange multiplier method:

[0070] (5)

[0071] in, These are the weighting coefficients; for u The autocorrelation matrix; f It is a column vector with all elements equal to 1; V The above items contain the axial position information of the array elements.

[0072] It should be noted that the optimal weights obtained by using the Lagrange multiplier method, i.e. the weighting factors, are used to correct the phase shift of the received signal caused by different transmission and reception distances between the receiving coil and the transmitting coil, so that signals from different transmission and reception distances can be aligned in phase, thereby improving the output signal-to-noise ratio and improving the overall signal quality.

[0073] Due to the limited downhole detection space, the degrees of freedom of ULMA-based downhole detection systems or instruments are restricted, resulting in a large weighting factor error and consequently inaccurate damage target detection results. Therefore, after obtaining the target pulsed eddy current array detection signal, the constant false alarm rate (CFAR) target detection principle is combined to further improve the performance of non-destructive testing of downhole casing. It should be noted that the CFAR algorithm, as a target detection technology, can automatically adjust the detection threshold according to changes in background noise power levels to maintain a constant false alarm probability, thereby achieving rapid and accurate target detection in complex background environments.

[0074] To maintain a constant false alarm probability under varying noise conditions, the CFAR algorithm's implementation divides the target pulsed eddy current array detection signal into different units based on signal processing strategies, such as preset unit sizes, search windows, or sliding windows. These units can be individual coils in a uniform linear multi-coil array or combinations of coil signals. For example, the target pulsed eddy current array detection signal is divided into reference units, detection units, and protection units. The reference unit is a signal unit used to estimate the background noise power level, typically located around the detection unit, forming forward and backward search windows, and is considered to include only background noise, excluding casing damage targets. The selection of the reference unit is usually based on an understanding of the signal environment and prior knowledge to ensure that the selected reference unit represents the characteristics of the background noise. The detection unit is the signal unit to be detected for the presence of casing damage targets; it is spatially different from the reference unit to avoid mutual interference between the reference unit and the detection unit. To reduce the probability of false alarms and minimize missed alarms, a protection unit needs to be installed between the detection unit and the reference unit to isolate the reference unit and the unit under test, thereby preventing power leakage between the reference unit and the detection unit and ensuring the accuracy of the detection results. In addition, the protection unit also helps to eliminate noise interference around the target signal and improve detection performance.

[0075] In some examples, the detection threshold is obtained based on a preset false alarm probability and the background noise power of the target pulsed eddy current array detection signal. This may include obtaining a threshold factor based on the preset false alarm probability; estimating the power spectral density of the target pulsed eddy current array detection signal to obtain the background noise power; and obtaining the detection threshold based on the product of the threshold factor and the background noise power.

[0076] In the above example, in target detection based on the CFAR algorithm, the false alarm probability and the detection probability are two important performance indicators, reflecting the system's false alarm rate and detection capability, respectively. Assume the false alarm probability of detecting a target when it does not exist is... P fa The detection probability of detecting the presence of a target when it exists is: P d Then there is P fa + P d =1. The choice of the threshold factor depends on the statistical characteristics of the noise and the required detection performance, such as the false alarm probability. P fa and detection probability P dAccordingly, the relationship between false alarm probability and threshold factor differs depending on the different classifications of CFAR algorithms. Specifically, CFAR algorithms can include Constant False Alarm Rate Detection (CA-CFAR), Smallest False Alarm Rate Detection (SO-CFAR), and Largest False Alarm Rate Detection (GO-CFAR). All three algorithms use the mean statistical method. Based on the relationship between false alarm probability and threshold factor, when the preset false alarm probability is constant, the threshold factor corresponding to the above three algorithms can be obtained.

[0077] The relationship between the false alarm probability and the threshold factor in the CA-CFAR algorithm is as follows:

[0078] (6)

[0079] The relationship between the false alarm probability and the threshold factor in the GO-CFAR algorithm is as follows:

[0080] (7)

[0081] The relationship between the false alarm probability and the threshold factor in the SO-CFAR algorithm is as follows:

[0082] (8)

[0083] in, The threshold factor; N The size of the reference cell; This represents the probability of a false alarm.

[0084] In some examples, the power spectral density of the target pulsed eddy current array probe signal is estimated to obtain the background noise power. Specifically, the target pulsed eddy current array probe signal is segmented for each selected reference cell. The length of each segment is determined based on the signal characteristics and the required frequency and resolution. An autocorrelation function is calculated for each segment, and a Fast Fourier Transform is applied to the result to obtain the power spectral density estimate for each segment. To improve the stability of the estimation, the power spectral density estimates for all segments are averaged to obtain the power spectral density of the entire signal. Frequency bands representing background noise are identified within the power spectral density. Within these noise frequency bands, the power spectral densities of multiple reference cells are summed to obtain the total background noise power.

[0085] The detection threshold is obtained by multiplying the threshold factor obtained in the above steps with the background noise power.

[0086] For the above example, a specific embodiment is used to illustrate the target detection principle based on the CFAR algorithm. As explained above, the false alarm probability and detection probability in target detection are two important performance indicators, reflecting the system's false alarm rate and detection capability, respectively. Correspondingly, the decision space can also be divided into two parts: in H Under the assumption of 0, the received signal u ( t The signal contains only noise. n ( t ); while H Under the assumption 1, the received signal u ( t The target signal exists simultaneously in ) s ( t and noise signals n ( t ).Right now:

[0087] (9)

[0088] See Figure 4 It illustrates the detection principle diagram of the mean constant false alarm rate provided in this disclosure. Figure 4 middle, P Represents the protection unit. D Represents the detection unit, and the two sides of the protection unit. x i ( )and x i ( () represents the reference unit. N Indicates the size of the reference cell. X This represents the estimated background noise power of the front half window. Y This represents the estimated background noise power value for the rear half window. Z The final calculated power estimate of the background noise is obtained. As a threshold factor, T For the detection threshold, where, X and Y The expressions are respectively , .

[0089] Combination Figure 4The detection principle diagram shown illustrates that the CA-CFAR, GO-CFAR, and SO-CFAR algorithms employ different selection strategies for the front and rear sliding window calculations of the reference cell when acquiring background noise power. Specifically, based on the target pulsed eddy current array detection signal and the size of the reference cell, data is segmented to obtain the front half-window data and the rear half-window data. Background noise power estimates are calculated for each half-window separately. CA-CFAR selects the average of the front and rear half-window power estimates as the final calculation result. This, along with its threshold factor, forms the detection threshold; GO-CFAR selects the largest of the power estimates for the first and second half windows as the final calculation result, i.e. This, along with its threshold factor, forms the detection threshold; SO-CFAR selects the minimum value between the power estimate of the first half-window and the power estimate of the second half-window as the final calculation result, i.e. This, together with its threshold factor, forms the detection threshold. T .

[0090] For example, after obtaining the detection threshold, the target pulse eddy current array detection signal can be filtered based on the detection threshold to obtain the damage detection result signal of the downhole casing. Specifically, the signal power of the unit to be detected in the target pulse eddy current array detection signal is first compared with the detection threshold to obtain the comparison result of each receiving coil; then, the damage detection result signal of the downhole casing is obtained based on the comparison result of each receiving coil.

[0091] In some examples, the damage detection result signal of the downhole casing is obtained based on the comparison result of each receiving coil. According to the detection threshold, the power amplitude value of each detection unit is compared, and the existence of the damaged target is determined based on Equation (9). Specifically, if the comparison result of each receiving coil is that the signal power of the unit to be detected in the target pulse eddy current array detection signal is greater than the detection threshold, then the damage detection result signal of the downhole casing is obtained. If the comparison result of each array element is that the signal power of the unit to be detected in the target pulse eddy current array detection signal is less than or equal to the detection threshold, then it is determined that there is no damaged target in the downhole casing.

[0092] For example, after obtaining the damage detection result signal of the downhole casing by comparing the detection threshold with the power amplitude value of each detection unit, it is necessary to determine the quantity and location of the downhole casing damage. In some examples, the location of the downhole casing damage is determined based on the damage detection result signal. Specifically, the damage detection result signal of the downhole casing is analyzed to extract feature information characterizing the downhole casing damage. The feature information of the downhole casing damage includes at least the amplitude, phase shift, and frequency of the damage detection result signal; the amplitude, phase shift, and frequency of the damage detection result signal are compared with the set damage feature thresholds to obtain the damage assessment result of the downhole casing; and the location of the downhole casing damage is obtained based on the damage assessment result and the spatial information of the damage detection result signal of the downhole casing.

[0093] For the example above, the damage detection result signal of the downhole casing is preprocessed, including noise filtering, DC component removal, and normalization, to improve the accuracy of feature extraction. Amplitude, phase shift, and frequency features are extracted from the damage detection result signal. Amplitude features are used to measure the peak value or root mean square (RMS) of the signal to quantify its intensity. Phase shift features are used to analyze the phase change of the signal waveform and determine the amount of phase shift. Frequency features are used to determine the main frequency components of the signal using Fourier transform. Based on historical data, expert experience, and experimental results, a threshold is set for each damage feature to distinguish between normal and damaged states. The extracted amplitude, phase shift, and frequency features are compared with the set thresholds to determine the presence of a damage target. Based on the feature comparison results, a damage assessment is performed on each detection unit, generating a damage assessment result. The assessment result includes the presence, type, and extent of damage.

[0094] In some examples, to obtain the damage assessment result of the downhole casing, specifically, the amplitude, phase shift, and frequency of the damage detection result signal can be compared with a set damage characteristic threshold to obtain the damage assessment result of the downhole casing. More specifically, based on the transmit and receive distance of each receiving coil of the uniform multi-coil array, phase compensation is performed on the initial pulse eddy current array detection signal corresponding to the damage detection result signal to construct a standard detection curve characterizing the downhole casing as undamaged; the amplitude, phase shift, and frequency of the damage detection result signal are then compared with the standard detection curve to obtain the damage assessment result of the downhole casing.

[0095] For example, in the above example, a standard detection curve representing the absence of damage to the downhole casing is obtained after processing using the LCMV algorithm. At locations of damage to the downhole casing, such as where the casing wall thickness changes, the standard detection curve will change accordingly, showing varying degrees of amplitude increase or broadening. The amplitude, phase shift, and frequency of the damage detection result signal are compared with the standard detection curve to determine the assessment result of damage to the downhole casing.

[0096] The above steps enable precise detection and location of downhole casing damage, providing accurate guidance for the maintenance and repair of oil and gas wells. This improves detection accuracy and enhances the ability to manage downhole casing integrity.

[0097] To verify Figure 2 The effectiveness of the technical solution shown was verified through simulation experiments. The effectiveness of the uniform linear multi-coil array wellbore transient electromagnetic method for detecting downhole casing was analyzed and compared. The application effects of CR-CFAR, GO-CFAR and SO-CFAR algorithms were analyzed and compared.

[0098] First, prepare a customized outer diameter dimension of 5. 1 / 2 - An inch metal sleeve, 7.72 mm thick and 139.7 mm in outer diameter, with four different degrees of wall thickness increase at different longitudinal positions. For example... Figure 5 As shown, along the well axis, the casing wall thickness increases by 3 mm, 1 mm, 2 mm, and 4 mm, respectively, with corresponding longitudinal lengths of 15 cm, 10 cm, 10 cm, and 15 cm. Then, continuous movement (uniform speed) detection is performed using a uniform linear multi-coil array sensor composed of 8 evenly distributed receiving elements. Detailed sensor parameters and experimental environment are shown in Table 1. For example, the number of receiving coils M is 8, and the spacing between each receiving coil, i.e., the element spacing... The diameter is 20mm. Other parameters are detailed in Table 1.

[0099] Table 1 Casing Parameters

[0100]

[0101] For the above simulation experiments, see Figure 6 and Figure 7This document illustrates the original induced electromotive force waveforms of the early and late signals from a receiving coil. In non-destructive testing (NDT) techniques, when using a uniform linear multi-coil array for downhole casing inspection, the early and late signals typically refer to the initial pulsed eddy current array detection signals captured at different time points during the inspection process. Distinguishing between early and late signals helps analyze signal changes over time and the characteristics of the signal at different stages. The early signal is the signal received shortly after the transmitting coil transmits the signal; the late signal is the signal received a longer time after transmission. In this disclosure, the received signal within 30 ms after the transient excitation is turned off is considered the early signal, and the signal after 30 ms is considered the late signal. Figure 6 and Figure 7 As shown, all eight test curves reflect the four wall thickness variations of the actual metal casing, demonstrating the effectiveness of transient electromagnetic downhole casing non-destructive testing technology. However, due to the different longitudinal distances between each receiving coil, i.e., the receiving element, and the transmitting coil, the detection curves acquired by each receiving element show relative shifts.

[0102] To improve the accuracy of CFAR detection, phase compensation processing is required for the initial pulsed eddy current array detection signal. For example, the LCMV array weighting method is used to perform phase compensation processing on the above eight sets of test signals, resulting in the following... Figure 8 and Figure 9 The normalized received signal curves are shown below. A comparison reveals that the curve after phase compensation is smoother and has a better signal-to-noise ratio than... Figure 6 and Figure 7 The original test curve is shown; at the same time, the array weighting completes the phase compensation, so that the measured curve more accurately reflects the location and degree of change of the casing wall thickness.

[0103] based on Figure 8 The early signal obtained after phase compensation was used for damage target detection. Given the complex downhole environment and the significant impact of noise and clutter on the detection results, CFAR algorithms with different threshold factors were employed for damage target detection. Specifically, CR-CFAR, GO-CFAR, and SO-CFAR algorithms were used respectively for... Figure 8 The early signals shown are used for target detection, and the detection results are as follows: Figure 10 , Figure 11 as well as Figure 12 As shown. Based on Figure 9 The obtained phase-compensated late signals were processed using CR-CFAR, GO-CFAR, and SO-CFAR algorithms, respectively. Figure 9 The late-stage signal shown is used for target detection, and the detection results are as follows: Figure 13 , Figure 14 as well as Figure 15 As shown.

[0104] By comparing the detection results of the three algorithms for early and late signals, it can be seen that under a uniform background of Rayleigh or exponential clutter, the SO-CFAR algorithm selects a low-power reference window as the estimate of the background power level of the detection unit, thereby reducing the influence of interfering targets and showing better performance. However, when there are interfering targets on both sides of the sliding window, the detection performance of SO-CFAR will drop sharply, resulting in the largest error of the SO-CFAR algorithm. Its detection results not only fail to reflect the range of the casing change area, but also incorrectly report the number of casing wall thickness anomalies. Both the CA-CFAR and GO-CFAR algorithms can correctly detect four anomalous targets. The GO-CFAR algorithm selects high-power clutter as the estimate of the background power level, thus having better false alarm control capability. However, the range and location of the four anomalous areas detected by GO-CFAR deviate significantly from the actual situation. Therefore, the CA-CFAR algorithm, based on the maximum likelihood estimation of the background power level, shows better detection performance. Its detection results have the highest consistency with the actual casing wall thickness change location, proving that the CA-CFAR algorithm can achieve high-precision non-destructive testing of downhole metal casing.

[0105] In summary, although the limited downhole detection space restricts the application effect of the ULMA system in downhole non-destructive testing, by combining it with the CFAR algorithm based on phase compensation of the initial pulse eddy current array detection signal, it can achieve rapid and accurate target detection in complex environments, improving the performance, speed and accuracy of downhole casing non-destructive testing.

[0106] Based on the same inventive concept as the aforementioned technical solution, see [link to inventive concept]. Figure 16 This disclosure illustrates a non-destructive testing device 1600 for downhole casing based on a constant false alarm rate. The device 1600 may include: an acquisition section 1601, a screening section 1602, and a determination section 1603; wherein...

[0107] The acquisition section 1601 is configured to acquire the detection threshold based on the preset false alarm probability and the background noise power of the target pulse eddy current array detection signal. The false alarm probability refers to the probability that the detection system incorrectly judges the presence of a damaged target when there is no damaged target in the downhole casing.

[0108] The filtering section 1602 is configured to filter the target pulse eddy array detection signal based on the detection threshold to obtain the damage detection result signal of the downhole casing;

[0109] Part 1603 is configured to determine the location of downhole casing damage based on the damage detection result signal of the downhole casing.

[0110] In some examples, filter 1602 is configured as follows:

[0111] The signal power of the unit to be detected in the target pulse eddy current array detection signal is compared with the detection threshold to obtain the comparison result of each receiving coil;

[0112] The damage detection result signal of the downhole casing is obtained based on the comparison result of each receiving coil.

[0113] In some examples, filter 1602 is configured as follows:

[0114] If the comparison result of each receiving coil is that the signal power of the unit to be detected in the target pulse eddy current array detection signal is greater than the detection threshold, then the damage detection result signal of the downhole casing is obtained.

[0115] In some examples, part 1603 is configured as follows:

[0116] The damage detection results signal of the downhole casing is analyzed to extract characteristic information of the downhole casing damage. The characteristic information of the downhole casing damage includes at least the amplitude, phase shift and frequency of the signal.

[0117] The amplitude, phase shift, and frequency of the signal are compared with the set damage characteristic thresholds to obtain the damage assessment results of the downhole casing.

[0118] The location of the downhole casing damage is determined based on the spatial information of the damage assessment results and the damage detection results of the downhole casing.

[0119] In some examples, part 1603 is configured as follows:

[0120] Based on the transmit and receive distance of each receiving coil in the uniform multi-coil array, phase compensation is performed on the initial pulse eddy current array detection signal corresponding to the damage detection result signal to construct a standard detection curve characterizing the undamaged downhole casing.

[0121] The amplitude, phase shift, and frequency of the damage detection result signal are compared with the standard detection curve to obtain the damage assessment result of the downhole casing.

[0122] In some examples, part 1601 is retrieved and configured as follows:

[0123] The threshold factor is obtained based on the preset false alarm probability;

[0124] The background noise power is obtained by estimating the power spectral density of the target pulsed eddy current array detection signal.

[0125] The detection threshold is obtained by multiplying the threshold factor and the background noise power.

[0126] In some examples, part 1601 is retrieved and configured as follows:

[0127] Signal acquisition is performed using each receiving coil of a uniform multi-coil array to obtain the initial pulsed eddy current array detection signal for the downhole casing;

[0128] Phase compensation is performed on the initial pulse eddy current array detection signal based on the transmit and receive distances between each receiving coil and the transmitting coil in the uniform multi-coil array to obtain the target pulse eddy current array detection signal.

[0129] Understandably, the exemplary technical solution of the downhole casing non-destructive testing device 1600 based on constant false alarm rate described above belongs to the same concept as the technical solution of the downhole casing non-destructive testing method based on constant false alarm rate described above. Therefore, all details not described in detail in the technical solution of the downhole casing non-destructive testing device 1600 based on constant false alarm rate can be found in the description of the technical solution of the downhole casing non-destructive testing method based on constant false alarm rate described above. This disclosure will not elaborate further on these details.

[0130] Please refer to Figure 17 This illustration shows a schematic diagram of the hardware structure of a computing device provided in an exemplary embodiment of this disclosure. In some examples, the computing device can be at least one of devices such as a smartphone, smartwatch, desktop computer, laptop, virtual reality terminal, augmented reality terminal, wireless terminal, and laptop computer. The computing device has communication functions and can access wired or wireless networks. The computing device can refer to one of multiple terminals; those skilled in the art will understand that the number of such terminals can be more or less. In some examples, the computing device can receive data from a satellite fault diagnosis method based on wavelet convolution based on the accessed wired or wireless network. It is understood that the computing device undertakes the computation and processing work of the technical solution of this disclosure, and this disclosure does not limit it in this regard.

[0131] like Figure 17 As shown, the computing device in this disclosure may include one or more of the following components: processor 1710 and memory 1720.

[0132] Optionally, the processor 1710 connects various parts within the computing device using various interfaces and lines, and performs various functions and processes data by running or executing instructions, programs, code sets, or instruction sets stored in the memory 1720, and by calling data stored in the memory 1720. Optionally, the processor 1710 can be implemented using at least one hardware form of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). The processor 1710 can integrate one or a combination of several of the following: Central Processing Unit (CPU), Graphics Processing Unit (GPU), Neural-network Processing Unit (NPU), and baseband chip. The CPU primarily handles the operating system, user interface, and applications; the GPU is responsible for rendering and drawing the content required to be displayed on the touch screen; the NPU is used to implement Artificial Intelligence (AI) functions; and the baseband chip is used to handle wireless communication. It is understandable that the aforementioned baseband chip may not be integrated into the processor 1710, but may be implemented as a separate chip.

[0133] The memory 1720 may include random access memory (RAM) or read-only memory (ROM). Optionally, the memory 1720 may include a non-transitory computer-readable storage medium. The memory 1720 may be used to store instructions, programs, code, code sets, or instruction sets. The memory 1720 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the various method embodiments described above, etc.; the data storage area may store data created according to the use of the computing device, etc.

[0134] In addition, those skilled in the art will understand that the structure of the computing device shown in the above figures does not constitute a limitation on the computing device. The computing device may include more or fewer components than shown, or combine certain components, or have different component arrangements. For example, the computing device may also include a display screen, camera assembly, microphone, speaker, radio frequency circuit, input unit, sensors (such as accelerometer, angular velocity sensor, fiber optic sensor, etc.), audio circuit, WiFi module, power supply, Bluetooth module, etc., which will not be described in detail here.

[0135] This disclosure also provides a computer-readable storage medium storing at least one instruction, which is executed by a processor to implement the non-destructive testing method for downhole casing based on constant false alarm rate as described in the various embodiments above.

[0136] This disclosure also provides a computer program product including computer instructions stored in a computer-readable storage medium; a processor of a computing device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computing device to perform the downhole casing non-destructive testing method based on constant false alarm rate according to the various embodiments described above.

[0137] Those skilled in the art will recognize that the functions described in this disclosure in one or more of the foregoing examples can be implemented using hardware, software, firmware, or any combination thereof. When implemented in software, these functions can be stored in a computer-readable medium or transmitted as one or more instructions or code on a computer-readable medium. Computer-readable media include computer-readable storage media and communication media, wherein communication media include any medium that facilitates the transfer of a computer program from one place to another. A readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0138] The above are merely specific embodiments of this disclosure, but the scope of protection of this disclosure is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A constant false alarm rate based non-destructive testing method for downhole casing, characterized in that, The method comprises: signal acquisition by each receiving coil of a uniform multi-coil array to obtain an initial pulsed eddy current array detection signal of a downhole casing; phase compensation of the initial pulsed eddy current array detection signal according to the transmitting-receiving distance between each receiving coil and a transmitting coil in the uniform multi-coil array to obtain a target pulsed eddy current array detection signal; acquisition of a detection threshold according to a preset false alarm probability and background noise power of the target pulsed eddy current array detection signal, wherein the false alarm probability refers to the probability that the detection system erroneously judges that a damage target exists when a downhole casing has no damage target; screening of the target pulsed eddy current array detection signal based on the detection threshold to obtain a damage detection result signal of the downhole casing; determination of the position of the downhole casing damage according to the damage detection result signal of the downhole casing; wherein the determination of the position of the downhole casing damage according to the damage detection result signal of the downhole casing comprises: analysis of the damage detection result signal of the downhole casing to extract feature information representing the downhole casing damage, wherein the feature information of the downhole casing damage at least includes the amplitude, phase shift and frequency of the damage detection result signal; phase compensation of the initial pulsed eddy current array detection signal corresponding to the damage detection result signal based on the transmitting-receiving distance of each receiving coil of the uniform multi-coil array to construct a standard detection curve representing the downhole casing without damage; comparison of the amplitude, phase shift and frequency of the damage detection result signal with the standard detection curve respectively to obtain a damage assessment result of the downhole casing; acquisition of the position of the downhole casing damage according to the damage assessment result and spatial information of the damage detection result signal of the downhole casing.

2. The method of claim 1, wherein, The screening of the target pulsed eddy current array detection signal based on the detection threshold to obtain the damage detection result signal of the downhole casing comprises: comparison of the signal power of a to-be-detected unit in the target pulsed eddy current array detection signal with the detection threshold to obtain a comparison result of each receiving coil; acquisition of the damage detection result signal of the downhole casing according to the comparison result of each receiving coil.

3. The method of claim 2, wherein, The acquisition of the damage detection result signal of the downhole casing according to the comparison result of each receiving coil comprises: if the comparison result of each receiving coil is that the signal power of the to-be-detected unit in the target pulsed eddy current array detection signal is greater than the detection threshold, the damage detection result signal of the downhole casing is obtained.

4. The method of claim 1, wherein, The acquisition of the detection threshold according to the preset false alarm probability and the background noise power of the target pulsed eddy current array detection signal comprises: acquisition of a threshold factor according to the preset false alarm probability; power spectrum density estimation of the target pulsed eddy current array detection signal to obtain the background noise power; acquisition of the detection threshold according to the product of the threshold factor and the background noise power.

5. A constant false alarm rate based downhole casing non-destructive testing device, characterized in that, The device comprises an acquisition part, a screening part and a determination part, wherein: the acquisition part is configured to perform signal acquisition by each receiving coil of a uniform multi-coil array to obtain an initial pulsed eddy current array detection signal of a downhole casing; Phase compensation is performed on the initial pulsed eddy current array probe signal according to the transmit-receive distance between each receiving coil and the transmitting coil in the uniform multi-coil array to obtain a target pulsed eddy current array probe signal; and A detection threshold is obtained according to a preset false alarm probability and a background noise power of the target pulsed eddy current array probe signal, wherein the false alarm probability refers to a probability that the detection system incorrectly judges that a damaged target exists when a non-damaged target exists in the downhole casing. The screening part is configured to screen the target pulsed eddy current array probe signal based on the detection threshold to obtain a damaged detection result signal of the downhole casing. The determining part is configured to determine the position of the downhole casing damage according to the damaged detection result signal of the downhole casing. The determining part is configured to analyze the damaged detection result signal of the downhole casing to extract feature information representing the downhole casing damage, wherein the feature information of the downhole casing damage at least includes the amplitude, phase shift and frequency of the damaged detection result signal. Phase compensation is performed on the initial pulsed eddy current array probe signal corresponding to the damaged detection result signal based on the transmit-receive distance of each receiving coil in the uniform multi-coil array to construct a standard probe curve representing the non-damaged downhole casing. The amplitude, phase shift and frequency of the damaged detection result signal are compared with the standard probe curve respectively to obtain a damage evaluation result of the downhole casing. The position of the downhole casing damage is obtained according to the damage evaluation result and spatial information of the damaged detection result signal of the downhole casing.

6. A computing device, comprising: The computing device comprises a processor and a memory; wherein The memory is used to store a computer program capable of running on the processor; The processor is used to execute the constant false alarm rate based non-destructive detection method of the downhole casing as claimed in any one of claims 1 to 4 when the computer program is run.

7. A computer storage medium, characterized in that The computer storage medium stores at least one instruction for being executed by the processor to implement the constant false alarm rate based non-destructive detection method of the downhole casing as claimed in any one of claims 1 to 4.

Images