MWD (Measurement While Drilling) instrument centralizer test and fault diagnosis system

Abstract

The invention discloses an MWD (Measurement While Drilling) instrument centralizer test and fault diagnosis system, which relates to the field of instrument magnetic compatibility fault test and comprises an upper computer and an environment simulation cabin filled with simulation drilling fluid, the upper computer is used for executing the following operations: in a state that the centralizer is not installed and the centralizer is installed, the environment simulation equipment is controlled to regulate and control the internal environment of the environment simulation cabin according to underground environment simulation parameters, and a first magnetic signal or a second magnetic signal is collected based on each magnetometer; modeling a first magnetic field distribution model for the first MWD instrument according to each first magnetic signal, and modeling a second magnetic field distribution model for the second MWD instrument according to each second magnetic signal; and based on the first magnetic field distribution model and the second magnetic field distribution model, testing whether the centralizer has magnetic compatibility abnormity or not. Therefore, the magnetic interference distortion effect of the centralizer on the MWD instrument under the real working condition can be effectively captured, and high accuracy and reliability of magnetic compatibility detection of the centralizer are achieved.

Description

Technical Field

[0001] The present application relates to the field of instrument magnetic compatibility fault testing, and in particular to a MWD instrument centralizer testing and fault diagnosis system. Background Art

[0002] In oil and gas drilling operations, the MWD (Measurement While Drilling) instrument relies on geomagnetic field information to achieve real-time measurement of the borehole trajectory, and its working accuracy is directly related to the safety and efficiency of drilling. As an important auxiliary component of the MWD instrument, the centralizer can effectively support the positioning, stabilization and correction of the instrument, ensuring that the instrument is always in the ideal working position in the wellbore.

[0003] At present, the centralizers are mostly made of high-strength metals or alloys to meet the mechanical strength requirements under high temperature, high pressure, strong vibration and other working conditions underground, but this also inevitably introduces the distortion effect of magnetic materials on the geomagnetic field. Specifically, some metal materials of the centralizer or their surface treatment processes will produce residual magnetism, which may cause abnormal distribution of local magnetic fields under complex underground working conditions, causing deviations in MWD instruments that originally rely on the correction direction of the geomagnetic field. In addition, if the spatial layout design of the centralizer and the MWD instrument is unreasonable, it will also lead to mutual influence between metal components, easily forming magnetic shielding or local magnetic field enhancement areas, thereby aggravating the magnetic interference phenomenon.

[0004] The current detection method mainly evaluates the magnetic properties of the centralizer through a single magnetic field measurement, but lacks a comprehensive test of the magnetic compatibility between the centralizer and the measuring instrument. In particular, in the actual downhole environment, the magnetic interference is not only affected by the material and structure of the centralizer itself, but also by the combined effects of multiple factors such as downhole temperature, pressure, and drilling fluid. This makes it difficult for the traditional offline magnetic property test method for the centralizer to fully evaluate its instrument compatibility issues in actual work.

[0005] In response to the above problems, the industry has not yet proposed a better technical solution. Summary of the invention

[0006] The present application provides a MWD instrument centralizer testing and fault diagnosis system, which is used to at least solve the problem in the current related technology that a single magnetic field measurement method cannot evaluate the influence of the centralizer on the magnetic compatibility of the MWD instrument.

[0007] The embodiment of the present application provides a MWD instrument centralizer test and fault diagnosis system, comprising a host computer and an environmental simulation cabin filled with simulated drilling fluid; the environmental simulation cabin is also provided with a bracket and an environmental simulation device; a mounting part for fixing the MWD instrument is provided on the bracket, and a plurality of magnetometers are evenly arranged inside the environmental simulation cabin; The host computer is used to perform the following operations: In the case where the mounting member fixes the first MWD instrument or the second MWD instrument, a preset downhole environmental simulation parameter is sent to the environmental simulation device, so that the environmental simulation device adjusts the internal environment of the environmental simulation cabin according to the downhole environmental simulation parameter, and collects a first magnetic signal for the first MWD instrument or a second magnetic signal for the second MWD instrument based on each of the magnetometers; the first MWD instrument is an MWD instrument without a centralizer installed, and the second MWD instrument is an MWD instrument with the centralizer installed; Modeling a first magnetic field distribution model for the first MWD tool according to each of the first magnetic signals, and modeling a second magnetic field distribution model for the second MWD tool according to each of the second magnetic signals; Based on the first magnetic field distribution model and the second magnetic field distribution model, the centralizer is tested whether there is magnetic compatibility anomaly.

[0008] The MWD instrument centralizer testing and fault diagnosis system provided by the present application can produce at least the following technical effects: By constructing an environmental simulation cabin that can truly reproduce the downhole environment, based on simulated drilling fluid, and with the help of environmental simulation equipment to accurately control environmental conditions, a comprehensive test is conducted on the magnetic compatibility of the centralizer and the MWD instrument, and the actual downhole environmental parameters are accurately restored in the environmental simulation cabin. Furthermore, by arranging multiple magnetometers at different positions in the cabin, the host computer is used to establish magnetic field distribution models for the MWD instrument with and without the centralizer installed. Through the comparative analysis of the two types of distribution models, the magnetic field distortion area caused by the centralizer and its influence range can be intuitively and accurately reflected, thereby effectively capturing the magnetic interference distortion effect of the centralizer on the MWD instrument under real working conditions, and comprehensively and systematically evaluating the spatial distribution and severity of the magnetic interference of the centralizer, significantly improving the high accuracy and reliability of the magnetic compatibility detection results, thereby improving the wellbore trajectory measurement accuracy of the MWD instrument and the safety and efficiency of drilling operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0010] Figure 1 A structural block diagram of an example of a MWD instrument centralizer testing and fault diagnosis system according to an embodiment of the present application is shown; Figure 2 A schematic diagram of the structural connection of an example of a MWD instrument centralizer test and fault diagnosis system according to an embodiment of the present application is shown; Figure 3 is a flowchart of an example of a magnetic signal analysis and fault diagnosis method executed by a host computer according to an embodiment of the present application; Figure 4 An operation flow chart of an example of testing magnetic compatibility anomaly of a centralizer using a multi-scale analysis method based on a magnetic field difference spectrum according to an embodiment of the present application is shown; Figure 5 A schematic diagram showing a structural connection of an example of an abnormal type identification model according to an embodiment of the present application is shown; Figure 6 A schematic diagram of a comparative simulation effect of an example of the test method of an embodiment of the present application and a single-scale magnetic field comparison method in local magnetic anomaly identification is illustrated. DETAILED DESCRIPTION

[0011] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0012] It should be noted that in the current related technologies, the magnetic interference test method for the centralizer is single, which often uses a static magnetic field test based on a Helmholtz coil, which can only simulate the magnetic shielding effectiveness under a uniform magnetic field. However, it cannot reproduce the complex environment downhole, resulting in a ±20% deviation between the test results and the actual working conditions. In particular, the actual magnetic field downhole is affected by the actual environment downhole. For example, changes in the external environment such as temperature and pressure may also cause the centralizer metal to dynamically interfere with the magnetic field distribution, causing deviations in the measurement results of the MWD instrument.

[0013] It should be understood that the purpose of the above description of the current related art is only to facilitate the public to better understand the inventive spirit and motivation of the present application, and is not to be regarded as a limitation of the present application. In addition, the technical solutions described in the above-mentioned current related art are not prior art, and may also be undisclosed technical solutions, such as solutions under research or in the laboratory stage.

[0014] In view of this, Figure 1 A structural schematic diagram of an example of a MWD instrument centralizer testing and fault diagnosis system according to an embodiment of the present application is shown.

[0015] like Figure 1As shown, the MWD instrument centralizer test and fault diagnosis system includes an environmental simulation cabin 110 filled with simulated drilling fluid and a host computer 120. A bracket 111 and an environmental simulation device 112 are also provided in the environmental simulation cabin 110. A mounting member 1111 for fixing an MWD instrument (not shown) is provided on the bracket 111, and a plurality of magnetometers 113 are evenly arranged inside the environmental simulation cabin 110.

[0016] In some embodiments, the environmental simulation chamber 110 is a closed simulated drilling wellbore environment, and a simulated drilling fluid with a preset viscosity and density is poured into the chamber to accurately reproduce the environmental characteristics of the drilling fluid in actual oil and gas well drilling operations. In addition, the wall of the environmental simulation chamber 110 is made of non-magnetic material (such as stainless steel or non-magnetic composite material) to prevent external magnetic interference from affecting the test accuracy.

[0017] The bracket 111 may be made of non-magnetic structural parts, such as special non-magnetic alloys or composite materials, to ensure that the bracket itself does not interfere with the measured magnetic field. In addition, the mounting member 1111 on the bracket 111 has a high-precision positioning structure, such as a guide groove, a buckle or a quick locking device, which is used to stably and reliably fix the first MWD instrument or the second MWD instrument to prevent vibration or drilling fluid flow from causing displacement of the instrument.

[0018] Multiple three-axis high-precision magnetometers 113 are evenly distributed in multiple directions and heights inside the environmental simulation cabin 110 to fully cover the magnetic field area around the first MWD instrument or the second MWD instrument. In addition, each magnetometer 113 can collect three-axis (X-axis, Y-axis, Z-axis) magnetic field intensity data in real time and record the change of magnetic field vector over time.

[0019] The environmental simulation device 112 can support the simulation of complex downhole environments. In some examples, the environmental simulation device 112 includes a temperature control module, a pressure control module, and a fluid circulation control module to achieve accurate simulation of working conditions such as temperature, pressure, and drilling fluid circulation. More specifically, downhole environmental parameters such as the depth of the actual drilling operation, the characteristics of the drilling area, and the type of rock formation are set to accurately simulate the actual drilling environment (conditions such as temperature, pressure, and fluid circulation). Preferably, the environmental simulation device 112 may also include an adjustable vibration simulation device to simulate the vibration spectrum of the drilling tool to truly reproduce the drilling vibration condition.

[0020] The host computer 120 can be various computing devices, such as industrial computers, etc. It needs to connect the magnetometer 113 and the environmental simulation device 112 in the environmental simulation cabin 110 to achieve data communication and real-time control through standard industrial communication protocols.

[0021] In some examples of the embodiments of the present application, the system further includes a motor for driving the mounting member to move on the bracket. In this way, the host computer 120 can generate a motor drive instruction according to the preset displacement control information, and then send the motor drive instruction to the motor to drive the mounting member to move on the bracket.

[0022] In some embodiments, the motor can be controlled in linkage according to multiple preset displacement points. At each preset displacement point, the motor automatically pauses after completing the movement, triggering the magnetometer to collect the magnetic field data of the current point. The host computer automatically associates the magnetic signal data of each point with the corresponding position coordinates, ensuring that a richer and clearer magnetic field distribution data set can be obtained.

[0023] Figure 2 A schematic diagram of the structural connection of an example of a MWD instrument centralizer testing and fault diagnosis system according to an embodiment of the present application is shown.

[0024] like Figure 2 As shown in the figure, the MWD instrument centralizer test and fault diagnosis system includes an industrial computer (or host computer), a main control PLC (programmable logic controller), a power supply control system, a temperature control module (electric heater, refrigeration unit, etc.), a vacuum module (such as a vacuum pump), a pressure control module (such as a pressure valve group, a regulating valve, etc.), and auxiliary power control equipment such as a frequency converter and a power regulator. Through the coordinated work of these functional modules, the system can simulate the complex environment such as high temperature, high pressure, vacuum and vibration that the MWD instrument may encounter underground, and realize online monitoring and fault diagnosis of the centralizer.

[0025] Here, the sealed cabin is the core environmental simulation space for testing the MWD instrument centralizer. The high-strength design ensures safety and sealing under extreme conditions such as high temperature, high pressure or vacuum. The MWD instrument to be tested and its centralizer are placed inside the cabin, and the temperature, pressure, vacuum degree, etc. in the cabin are changed through environmental simulation equipment. The power supply control system provides stable and controllable power to the sensors, actuators, and MWD instruments in the cabin.

[0026] Through the combination of temperature regulation module, pressure regulation module and vacuum module, the system can simulate a variety of extreme underground environments (high temperature and high pressure, low temperature and low pressure, vacuum exhaust, etc.) simultaneously or successively, and conduct a full range of tolerance and performance tests on the MWD instrument centralizer. In addition, the main control PLC monitors and controls the environmental parameters at the millisecond level, and the industrial computer can further perform data fusion and fault analysis. When it is detected that the temperature, pressure, etc. deviate from the set value, the PLC controls the actuator to compensate to ensure the stability of the sealed cabin test environment. Therefore, through the multi-physical quantity coupling test method, the underground environmental changes can be reproduced more realistically, and the reliability evaluation of the centralizer has a higher credibility.

[0027] Here, the industrial computer not only collects environmental parameters and MWD instrument output data in real time, but also runs a fault diagnosis process based on multi-sensor fusion and intelligent algorithms. Once the magnetic compatibility anomaly, structural fatigue or other signs of failure of the MWD instrument centralizer are found, the system can automatically trigger an early warning and save the abnormal data, providing a reference for subsequent improvements and maintenance.

[0028] Figure 3 It is a flowchart of an example of a magnetic signal analysis and fault diagnosis method executed by a host computer according to an embodiment of the present application.

[0029] like Figure 3 As shown, in step S310, when the mounting member 1111 fixes the first MWD instrument or the second MWD instrument, preset downhole environmental simulation parameters are sent to the environmental simulation device, so that the environmental simulation device adjusts the internal environment of the environmental simulation cabin according to the downhole environmental simulation parameters, and collects the first magnetic signal for the first MWD instrument or the second magnetic signal for the second MWD instrument based on each magnetometer.

[0030] Here, the first MWD tool is an MWD tool without a centralizer installed, and the second MWD tool is an MWD tool with a centralizer installed.

[0031] In some embodiments, a simulated downhole working condition parameter group is preset in the host computer 120. For example, the temperature range of the simulated downhole environment can be set to 50-180°C, the downhole environment pressure range can be adjusted to 0-150 MPa, and different concentrations can be set or specific materials can be added according to the needs of the actual well section to adjust the viscosity and density of the drilling fluid, and the fluid circulation control module can be controlled to adjust the fluid velocity of the drilling fluid according to the vibration response of the MWD tool during drilling.

[0032] Specifically, in the two states of not installing the centralizer (first MWD instrument) and installing the centralizer (second MWD instrument), the host computer 120 controls the environmental simulation equipment to complete the above environmental condition simulation. At the same time, the magnetic signals are collected synchronously in real time based on multiple magnetometers 113, and are transmitted back to the host computer 120 in the form of real-time data streams to ensure that continuous high-precision magnetic field measurement data is obtained. Therefore, by simulating the downhole environment with full parameters, the magnetic interference of the centralizer in actual working conditions is truly reproduced, effectively avoiding the problem of mismatch between offline test results and actual conditions.

[0033] In step S320, a first magnetic field distribution model for the first MWD tool is modeled according to each first magnetic signal, and a second magnetic field distribution model for the second MWD tool is modeled according to each second magnetic signal.

[0034] It should be noted that the modeling method of the magnetic field distribution model can be diverse, such as spatial interpolation and three-dimensional magnetic field modeling algorithms (such as three-dimensional spatial interpolation methods) to establish a magnetic field distribution model. Exemplarily, by adopting a spatial interpolation algorithm (such as Kriging method or radial basis function interpolation), the magnetic signals collected by each magnetometer 113 are interpolated to establish a spatial magnetic field continuous distribution model.

[0035] Furthermore, by analyzing the magnetic field gradient of the centralizer, the spatial position of the abnormal area and the numerical characteristics of the abnormal magnetic intensity, and generating the corresponding three-dimensional magnetic field distribution cloud map, the impact of the integrated centralizer on the magnetic field characteristic spatial model around the instrument can be effectively simulated, providing an intuitive and reliable basis for the detection and diagnosis of magnetic compatibility anomalies.

[0036] In some examples of the embodiments of the present application, an effective element analysis method can also be used to complete the modeling of the magnetic field distribution model, and more details will be discussed below in conjunction with other examples.

[0037] In step S330, based on the first magnetic field distribution model and the second magnetic field distribution model, it is tested whether the centralizer has magnetic compatibility anomaly.

[0038] In some embodiments, the host computer 120 uses the first magnetic field distribution model (without the centralizer installed) as the reference magnetic field model, and compares the difference between the second magnetic field distribution model and the first magnetic field distribution model, and adopts spatial magnetic field difference analysis. Specifically, the abnormal point positioning algorithm is used to compare the difference values ​​of the magnetic induction intensity between the two spatial magnetic field models, and mark the area where the difference value of the magnetic induction intensity exceeds the preset threshold as the magnetic compatibility anomaly area. In addition, the magnetic anomaly distribution heat map can be used for intuitive display to automatically locate the area and specific parts of severe magnetic interference. In this way, the position, intensity and source of the magnetic anomaly caused by the centralizer can be clearly revealed, helping engineers to quickly locate the cause of the magnetic compatibility problem and guide the next step of the centralizer design optimization or material selection.

[0039] Through the embodiments of the present application, a comprehensive magnetic compatibility detection and diagnosis test system with full parameters and multi-factor coupling is constructed, which can simulate a variety of real complex downhole environments, realize dynamic simulation testing and comprehensive analysis of the overall magnetic field compatibility between the centralizer and the MWD instrument, make up for the traditional single-parameter offline detection method's neglect of magnetic interference in complex downhole conditions, accurately identify the source of magnetic compatibility anomalies, effectively guide the optimization of the centralizer, and greatly improve the evaluation accuracy of the magnetic compatibility interference of the centralizer.

[0040] In some examples of the embodiments of the present application, the host computer 120 can also send environmental update parameters to the environmental simulation device 112, so that the environmental simulation device 112 adjusts the internal environment of the environmental simulation chamber from the downhole environmental simulation parameters to the environmental update parameters, so as to retest whether there is a magnetic compatibility anomaly in the stabilizer under the simulation environment corresponding to the environmental update parameters.

[0041] In some embodiments, multiple modes are pre-set in the system, and several typical environmental parameter combinations are defined, which can be called according to common downhole environmental changes. In addition, testers can also flexibly set parameter values ​​according to test requirements to meet personalized test needs. For example, in each test, the magnetic compatibility data of the stabilizer is recorded by testing using downhole environmental simulation parameters. Furthermore, after each environmental parameter update, the system retests the magnetic field distribution model of the MWD instrument to determine the magnetic compatibility changes of the stabilizer under different environmental conditions, and archives the magnetic signal data and abnormal analysis results corresponding to each set of environmental parameters. Therefore, by performing multiple rounds of magnetic compatibility tests on the MWD instrument under different environmental conditions, the potential magnetic anomalies of the MWD instrument in a specific environment can be effectively discovered, thereby improving the reliability of the magnetic compatibility test results.

[0042] Regarding the modeling details of the magnetic field distribution model in step S320, in some examples, it can be to use an effective element analysis method to discretize the stabilizer and its surrounding structures into representative effective units, simplify meshing, reduce computational complexity, and retain key geometric and material properties.

[0043] Specifically, the MWD instrument is discretized in spatial geometry and divided into Valid units, of which The volume of an effective unit cell is .

[0044] Since the magnetometers are evenly arranged inside the environmental simulation cabin, the discretely collected magnetic signals need to be classified into corresponding valid units according to their spatial positions. A valid unit, and a set of sensors near it is defined as , and the number of sensors is .

[0045] Extract the local average magnetic signal corresponding to each effective unit cell: , Formula (1) In the formula, Indicates The center position of the effective unit cell, Indicates classification into The local average magnetic field strength, Indicates The number of sensors participating in the statistics in a valid unit, Indicates coverage A magnetometer collection of active units, Represents the magnetometer in the collection The installation location of the magnetometer, Indicates the position measured by the magnetometer The original magnetic field strength at .

[0046] In formula (1), by The sensor data is averaged to eliminate local noise fluctuations and obtain a reference value reflecting the overall magnetic field state of the area. Therefore, the magnetic signals collected by multiple magnetometers are used to classify the discrete measurement point data distributed on the mounting parts into various "effective element" units, providing local representative magnetic signals for subsequent modeling.

[0047] In effective element analysis, each discrete unit is considered as an approximate magnetic dipole, whose magnetic properties are described by the effective magnetization (or magnetic moment density) and can be estimated using the local magnetic signal and geometric factors.

[0048] Each effective unit cell is regarded as an approximate magnetic dipole. For the magnetic dipole model, the magnetic field at the target point along the axis of the magnetic dipole is approximately expressed as: , Formula (2) In the formula, Indicates the location of the target point. express The magnetic field strength at represents the vacuum permeability, represents the magnetic moment of a magnetic dipole.

[0049] Here, the effective unit The distance between the nearby sensors and the center of the cell is defined as , estimate the average geometry factor: , Formula (3) In the formula, Indicates The average geometric factor of the effective element units is used to convert the local magnetic field signal into magnetic moment.

[0050] pass To comprehensively consider the influence of the distance from each sensor to the effective element center, it can effectively convert the magnetic signal into the magnetic dipole moment.

[0051] Based on the relationship between the local magnetic signal and the geometric factor, we get: , Formula (4) In the formula, Indicates The effective magnetization intensity of an effective unit cell reflects the magnetization degree and direction of the magnetic material.

[0052] So we can solve: , Formula (5) Using formula (5), The magnetic field measured locally at the effective unit cell It is converted into its effective magnetization intensity, and can be fine-tuned through further data fitting or iterative optimization. The magnetization intensity equivalent to the magnetic response of the region is inferred by using the locally measured magnetic field value and the average distance information between the sensor and the effective element center in the region. Therefore, each local area is regarded as a small magnetic dipole unit, and the collected magnetic signal is converted into the equivalent magnetization intensity of each effective element unit by using the local magnetic signal and geometric factors, reflecting the comprehensive magnetic characteristics of the magnetic material in the region.

[0053] Under the magnetic dipole approximation, each effective unit cell can produce a certain magnetic field contribution. Valid units at the target point The contribution of magnetic field intensity at is recorded as: , Formula (6) In the formula, Indicates The effective unit is The contribution of magnetic field strength generated at Indicates from arrive The Green's function is used to describe the attenuation and directionality of magnetic field propagation.

[0054] In the magnetic dipole approximation, the Green's function is expressed as: , Formula (7) In the formula, Indicates The unit vector of the magnetization direction of an effective unit cell characterizes the main magnetization direction of the magnetic material.

[0055] Here, the Green's function describes the spatial propagation of the magnetic field from the source point (effective element) to the target point, taking into account the distance attenuation ( ) and directionality (dot products and vector components), providing a quantitative expression for the superposition of magnetic field contributions.

[0056] The magnetic field intensity contribution of each target point in all effective units is superimposed to obtain the magnetic field distribution model of the MWD instrument: , Formula (8) In the formula, The magnetic field distribution model at the target point The magnetic field strength is expressed.

[0057] Therefore, by calculating the target point By superimposing the contributions of , the magnetic field distribution expression of any spatial point in the entire structure can be obtained, and the magnetic field distribution model of the entire MWD instrument test area can be constructed.

[0058] Through the embodiment of the present application, multi-point magnetometers are deployed in the environmental simulation cabin to collect magnetic field signals at different positions. The sensor data near each "effective element" is locally averaged to obtain the average magnetic field condition in the unit area. Then, the local average magnetic field value is converted into an equivalent magnetization intensity by defining a geometric factor, so that each effective element unit can be approximated as a magnetic dipole in space, and local processing can be performed for the metal properties, stabilizer structure and downhole environment of different regions. Block-type effective element modeling is used to characterize the influence of the stabilizer on the refinement of the magnetic field in a fine-grained manner. In addition, the magnetic field contribution of each effective element unit is superimposed to the target point through the Green's function, thereby constructing a magnetic field distribution model for the entire space. In this way, compared with direct global modeling, this method greatly reduces the computational complexity while taking into account physical accuracy.

[0059] It should be noted that the first magnetic field distribution model represents the reference state of the magnetic field when the MWD instrument is not installed with a centralizer; the second magnetic field distribution model represents the magnetic field distribution state after the centralizer is installed, and the difference between the two reflects the influence of the centralizer itself on the magnetic compatibility of the instrument. In some embodiments, the difference can be simply compared by point-by-point subtraction, but such a processing method often ignores the local feature differences in space and makes it difficult to distinguish the type and severity of magnetic field anomalies.

[0060] In view of this, regarding the implementation details of step S330, in some examples, the first magnetic field distribution model and the second magnetic field distribution model are compared based on the multi-scale analysis method of the magnetic field difference spectrum to test whether the centralizer has magnetic compatibility anomaly. Specifically, by performing a spatial domain-frequency domain joint analysis on the two magnetic field models, the magnetic compatibility anomaly of the centralizer is accurately identified, and an intuitive and interpretable diagnostic result is provided.

[0061] Figure 4 An operational flow chart of an example of testing magnetic compatibility anomalies of a centralizer using a multi-scale analysis method based on a magnetic field difference spectrum according to an embodiment of the present application is shown.

[0062] like Figure 4 As shown, in step S410, a magnetic field difference map in three-dimensional space is constructed according to the magnetic field difference of each target point.

[0063] , Formula (9) In the formula, Indicates that the first magnetic field distribution model is at the target point The magnetic field strength, and Indicates that the second magnetic field distribution model is at the target point The magnetic field strength; Indicates the target point The magnetic field intensity difference at each target point is calculated, and then a magnetic field difference map in three-dimensional space is constructed according to the magnetic field intensity difference of all target points in the space.

[0064] Here, two sets of magnetic field distribution models are first constructed without the centralizer installed (benchmark) and after the centralizer is installed. Then, the two models are subtracted point by point to obtain the difference value of the magnetic field in space. In this way, the actual distribution characteristics of the impact of the installation of the centralizer on the magnetic field can be clearly obtained, forming a clear magnetic field difference map. In this way, the magnetic field anomaly introduced by the centralizer itself can be clearly isolated, avoiding interference of other factors on the magnetic field analysis.

[0065] In step S420, the magnetic field difference map is decomposed using a three-dimensional discrete wavelet transform to obtain multi-scale wavelet coefficients.

[0066] , Formula (10) In the formula, the target point r The position is determined by the three-dimensional discrete space grid coordinates To express, Represents the magnetic field difference map against the position In the The scale of The wavelet coefficients in the directions, Indicates the target point r The three-dimensional coordinates of the neighborhood space position; is the total number of neighborhood points analyzed in the direction of the corresponding coordinate axis, which represents the spatial discretization accuracy used for magnetic field analysis; Indicates The scale of The mother wavelet function in each direction.

[0067] It should be noted that magnetic field anomalies of different scales may exist simultaneously in the magnetic field difference map. For example, small-scale local anomalies may be caused by local material defects of the stabilizer, while larger-scale overall anomalies may come from overall structural design problems of the stabilizer. Therefore, the 3D-DWT (Discrete Wavelet Transform) method is used to transform the magnetic field difference map data through the three-dimensional mother wavelet function. The spatial convolution is decomposed scale by scale and direction by direction. Through such operation, the detailed characteristic information of magnetic field anomaly at different spatial scales and directions can be obtained, namely, the wavelet coefficients .

[0068] Therefore, wavelet transform can extract characteristic information of magnetic field differences at different scales and directions, which is conducive to identifying magnetic compatibility anomalies at different spatial scales. By capturing the characteristics of magnetic compatibility anomalies caused by the centralizer in a multi-scale and refined manner, it can ensure that both subtle anomalies (such as local material defects) and overall anomalies (such as overall design defects) can be effectively identified.

[0069] In step S430, the local magnetic field anomaly intensity index and the global magnetic field anomaly energy index are calculated based on the multi-scale wavelet coefficients to construct a multi-scale magnetic anomaly feature matrix.

[0070] Specifically, the local magnetic field anomaly intensity index is calculated by the following formula: , Formula (11) In the formula, Indicates Position at scale The local magnetic field anomaly intensity index, Indicates Position at scale The vector of wavelet coefficients in all directions.

[0071] About Location In scale The wavelet coefficients under the energy calculation (squared) are obtained to obtain the magnetic anomaly intensity index of each spatial position. The meaning of LMASI is that the more obvious the anomaly at a specific location, the larger the value of the wavelet coefficient. The squared LMASI index significantly amplifies the difference characteristics, which is more conducive to clearly reflecting the spatial location and severity of the magnetic compatibility anomaly. In this way, the accurate quantification of local magnetic anomalies is achieved, and it is also convenient to locate the spatial position of the anomaly.

[0072] The global magnetic field anomaly energy index is calculated by the following formula: , Formula (12) In the formula, Indicates The global magnetic field anomaly energy index of the scale, It represents the sum of the positions of all target points in the space. Indicates the sum of all analysis directions.

[0073] In formula (12), based on the local anomaly index, the energy of the wavelet coefficients at all target point positions and all analysis directions in the space is accumulated to characterize the degree of magnetic interference of the centralizer at different scales as a whole. If the overall magnetic anomaly energy index exceeds the statistical threshold of the normal range, it means that the centralizer has caused a global magnetic compatibility problem for the entire MWD instrument, and the severity of the overall magnetic compatibility anomaly can be quickly determined.

[0074] The multi-scale magnetic anomaly characteristic matrix is ​​calculated by the following formula: , Formula (13) In the formula, represents the multi-scale magnetic anomaly feature matrix, Indicates the total number of scales.

[0075] In the multi-scale magnetic anomaly feature matrix, local anomalies and global anomalies at each scale are unified and integrated to form a complete anomaly feature description system. Information at each scale, from local anomalies to global anomalies, is clearly presented, facilitating subsequent rapid and accurate magnetic compatibility anomaly detection and diagnosis.

[0076] In step S440, based on the multi-scale magnetic anomaly feature matrix, a corresponding magnetic compatibility anomaly threshold is adaptively determined.

[0077] , Formula (14) , Formula (15) In the formula, and Corresponding to The spatial statistical mean and standard deviation of the local magnetic field anomaly intensity index corresponding to all target points within a scale, is the local abnormal sensitivity coefficient, is the global abnormal sensitivity coefficient, For the The local scale magnetic compatibility anomaly threshold of scales, For the The global scale magnetic compatibility anomaly threshold of scales, and Corresponding to The spatial statistical mean and standard deviation of the global magnetic field anomaly intensity index corresponding to all target points within a scale; is the adaptive fine-tuning coefficient, and They represent the first The local base threshold and the global base threshold of each scale are calculated.

[0078] Here, the robust threshold (i.e., basic threshold) determined by historical data is combined with the dynamic threshold of the real-time test environment to set the adaptive coefficient , a balance is achieved between historical data and current data, so that the threshold is both robust and realizes an adaptive anomaly threshold calculation method based on the combination of historical data and current real-time measurement data. Specifically, the obtained LMASI and GMAEI indexes are not simply compared with a fixed threshold, but a dynamic threshold is adopted, that is, the initial robust threshold is established according to historical data, and it is adaptively fine-tuned according to the statistical characteristics of the current actual test data, which can more effectively adapt to the differences in different downhole environments (such as temperature and pressure changes), greatly reducing the misjudgment rate and missed judgment rate of magnetic compatibility anomaly detection.

[0079] In step S450, the magnetic compatibility anomaly threshold and the multi-scale magnetic anomaly feature matrix are compared to determine whether a magnetic compatibility anomaly exists.

[0080] If detected in the multi-scale magnetic anomaly feature matrix , then the judgment scale Lower position There is a local magnetic compatibility anomaly; if it is detected in the multi-scale magnetic anomaly feature matrix , then the judgment scale There is an overall magnetic compatibility anomaly. In this way, by comparing the anomaly index with the adaptive threshold, the specific location, severity and impact range of the magnetic compatibility anomaly can be automatically determined.

[0081] Through the embodiments of the present application, the corresponding magnetic field difference map is constructed by comparing the magnetic field distribution models, and the magnetic field interference introduced by the MWD instrument stabilizer is accurately isolated. Then, because the magnetic compatibility problem of the stabilizer has multi-scale characteristics, it is specifically manifested as follows: on the one hand, the stabilizer material or manufacturing process may cause local microstructural inhomogeneity, which usually manifests as local magnetic field anomalies, which are small in scale but may cause the local accuracy of the measuring instrument to decrease; on the other hand, the overall material selection, mechanical structure design or assembly method of the stabilizer may bring about overall magnetic field anomalies on a macro scale. Such anomalies usually affect the overall measurement benchmark of the instrument and even seriously reduce the overall measurement accuracy. Therefore, by adopting the three-dimensional discrete wavelet transform method, the magnetic field difference is spatially decomposed in multiple scales and directions.

[0082] In addition, the Local Magnetic Anomaly Intensity Index (LMASI) and the Global Magnetic Anomaly Energy Index (GMAEI) were further defined based on the spatial multi-scale analysis. Specifically, LMASI can accurately capture local micro-magnetic anomalies, especially magnetic compatibility anomalies caused by local material defects, assembly deviations, or local environmental factors (such as local temperature differences or pressure changes) in the stabilizer. This helps to quickly and accurately discover and locate the location and severity of these local defects, thereby providing targeted guidance for local correction or material optimization design of the stabilizer.

[0083] GMAEI can effectively identify systematic magnetic compatibility problems that may exist in the structure or material selection of the stabilizer at the overall scale. Specifically, GMAEI evaluates the magnetic field interference intensity of the stabilizer from the overall scale by accumulating the energy of the wavelet coefficients at all positions and directions in space at a specific scale. If the GMAEI value exceeds the adaptive threshold at the corresponding scale, it clearly indicates that there are systematic magnetic compatibility problems in the overall design, structure or material selection of the stabilizer, which usually cannot be solved by simple local corrections, and may require engineers to make fundamental improvements and optimizations to the overall design, materials or assembly process of the stabilizer.

[0084] It should be noted that the magnetic field difference map obtained by comparing LMASI and GMAEI can distinguish the local and overall magnetic compatibility anomalies caused by the stabilizer to a certain extent, and can intuitively reflect the intensity and range of the abnormal area. However, as statistical indicators, although LMASI and GMAEI can quantify the intensity of anomalies, they may not be able to capture the subtle structure, spatial distribution and pattern complexity of the anomaly characteristics. For example, the same numerical change may come from different types of faults (such as local enhancement and local attenuation), and a single indicator is difficult to fully distinguish these situations, making it possible to rely solely on fixed statistical indicators for classification, which may not be robust enough when facing complex and dynamically changing underground environments.

[0085] In view of this, in some examples of the embodiments of the present application, the deep learning model can automatically learn abnormal patterns from a large amount of data, comprehensively consider multi-scale, multi-directional and spatial information, and improve the accuracy and stability of classification. In addition, the deep learning model can also support the classification of magnetic compatibility anomalies into more fine-grained classifications, which can provide clearer guidance for subsequent maintenance and design improvements.

[0086] More specifically, after comparing the first magnetic field distribution model and the second magnetic field distribution model to test whether the stabilizer has a magnetic compatibility anomaly based on a multi-scale analysis method of the magnetic field difference map, the upper computer 120 is also used to construct an abnormal input feature vector based on the abnormal local magnetic field anomaly intensity index corresponding to the local magnetic compatibility anomaly, the global magnetic field anomaly energy index corresponding to the overall magnetic compatibility anomaly, and the spatial anomaly feature code corresponding to the local magnetic compatibility anomaly position when it is determined that the stabilizer has a magnetic compatibility anomaly.

[0087] , Formula (16) , Formula (17) In the formula, represents the abnormal input feature vector, Indicates the location of local magnetic compatibility anomaly The corresponding spatial anomaly feature code, express The three-dimensional coordinates of and The scale Lower position The mean and standard deviation of the local magnetic field anomaly intensity index corresponding to each neighborhood spatial location.

[0088] Here, when identifying the type of magnetic compatibility anomaly of the centralizer, different types of magnetic anomalies (local anomalies or global anomalies) show completely different spatial structural characteristics. To this end, the input feature vector is constructed by combining the local magnetic anomaly intensity index (LMASI) and the global magnetic anomaly energy index (GMAEI) with spatial feature coding, which clearly reflects the multi-scale structural characteristics and spatial characteristics of the magnetic compatibility anomaly, and effectively avoids the misjudgment when relying solely on the magnetic intensity value for judgment.

[0089] The abnormal input feature vector is input into the abnormal type recognition model to determine the corresponding magnetic compatibility fault type.

[0090] Figure 5 A schematic diagram of the structural connection of an example of an abnormal type identification model according to an embodiment of the present application is shown.

[0091] like Figure 5 As shown, the abnormal type recognition model 500 adopts a multi-scale adaptive attention convolutional network and includes a cascaded input layer 510, a multi-scale convolutional feature extraction layer 520, an adaptive attention fusion layer 530, a feature enhancement and dimensionality reduction layer 540 and a classification output layer 550.

[0092] The input layer 510 is used to receive an input feature vector.

[0093] The multi-scale convolution feature extraction layer 520 is used to utilize multiple parallel three-dimensional convolution layers to respectively extract multi-scale local features of the magnetic field difference feature map with convolution kernels of different scales.

[0094] , Formula (18) In the formula, Indicates the scale of extraction The abnormal spectrum characteristics of represents the convolution operation, and Respectively represent The convolution kernel weights and bias parameters of each scale, Represents the RELU nonlinear activation function.

[0095] Here, multi-scale spatial features in the input features are extracted simultaneously through three-dimensional convolution kernels of different sizes. The parallel use of convolution kernels of different scales enables the model to fully explore the characteristic expressions of magnetic field anomalies at different spatial scales (small-scale local anomalies and large-scale overall anomalies), and can fully capture the subtle differences of the magnetic anomalies of the stabilizer at multiple spatial scales.

[0096] Through this embodiment, a multi-scale magnetic field difference map is constructed and combined with spatial anomaly position encoding, which can effectively capture the difference characteristics of different magnetic compatibility fault types and significantly improve the recognition accuracy of magnetic anomaly types. Through a multi-scale three-dimensional convolution structure, it is possible to sensitively capture the magnetic anomaly characteristics of different spatial ranges (local or overall), avoid the omission or misjudgment of magnetic anomaly feature information, and ensure the accuracy of magnetic compatibility anomaly type recognition.

[0097] The adaptive attention fusion layer 530 is used to effectively fuse multi-scale local feature maps through an adaptive scale attention mechanism, and automatically capture the most discriminative magnetic field anomaly scale features.

[0098] , Formula (19) , formula (20) In the formula, It represents the magnetic anomaly scale fusion feature obtained by fusing feature maps of different scales. represents the number of convolution kernels, Indicates Adaptive attention weights at each scale, represents the softmax function, represents the hyperbolic tangent function, is the trainable parameter of the attention mechanism, express The transposed vector of .

[0099] Here, based on the features extracted by multi-scale convolution, the adaptive attention mechanism is used to automatically learn and assign weights. The model automatically and adaptively increases the weights of key scales according to the importance of the magnetic field anomaly pattern to highlight the most discriminating scale features of the magnetic field anomaly. Therefore, by introducing the adaptive attention fusion layer, the model can autonomously learn and dynamically emphasize the most representative and discriminating scale features, making the model more sensitive to different magnetic anomaly patterns.

[0100] The feature enhancement and dimension reduction layer 540 is used to further enhance and reduce the fused features through a fully connected network to extract key high-discriminative features.

[0101] , Formula (21) In the formula, represents the dimension reduction enhanced feature vector, and They represent the weight matrix and bias parameters of the fully connected network respectively.

[0102] It should be noted that after multi-scale feature fusion, the feature dimension is high and may contain redundant information. Therefore, a fully connected layer is used to further nonlinearly enhance and reduce the feature dimension, significantly compress the feature dimension, remove redundancy, emphasize the essential feature expression of magnetic compatibility anomaly, and reduce the computational complexity of the model.

[0103] The classification output layer 550 is used to output the magnetic compatibility anomaly type through the softmax layer: , Formula (22) In the formula, represents the probability vector of magnetic compatibility anomaly type classification, each element of which represents the predicted probability of the corresponding anomaly type; and They represent the weight matrix and bias parameters of the classification output layer respectively.

[0104] Here, the softmax function is used to map the enhanced dimensionality-reduced feature vector into a probability distribution of the anomaly type, providing a specific and intuitive classification probability for each magnetic anomaly type. This ensures that the model output results have probabilistic interpretability, helps to clarify the uncertainty and confidence of the anomaly type prediction, and then assists decision makers to quickly and effectively determine the reliability of the diagnosis of magnetic compatibility anomaly types.

[0105] Through the embodiments of the present application, spatial feature encoding is used to fuse spatial position and statistical feature information, so that the model can not only distinguish the pattern types of different anomalies, but also accurately locate the location where the magnetic anomaly occurs, and realize an anomaly identification method based on spatial sensitivity, which provides clear and intuitive guidance for subsequent centralizer design optimization, structural improvement and precise fault location.

[0106] Exemplarily, the magnetic compatibility fault type includes at least one of the following: a local magnetic field abnormal enhancement fault, a local magnetic field abnormal attenuation fault, a magnetic field gradient abnormal fault, and a magnetic field offset and direction distortion fault.

[0107] Regarding the explanation of the local magnetic field abnormal enhancement fault or the local magnetic field abnormal attenuation fault, it may be caused by the use of materials with strong residual magnetism or obvious magnetic saturation effect in the stabilizer, or the uneven local metal stacking in the structural design leads to local magnetic field abnormality. Accordingly, it can be recommended to use surface treatment technology to reduce the residual magnetism of the material, and to add a magnetic field compensation device (such as a magnetic shielding sheet or a reverse magnetic source) to the stabilizer structure to make the local magnetic field distribution more uniform.

[0108] Regarding the explanation of the abnormal magnetic field gradient failure, on the one hand, it may be caused by the uneven structure of the stabilizer, which leads to a sudden change in the magnetic flux density gradient. On the other hand, it may be caused by environmental factors (such as local temperature and pressure changes) that cause changes in the magnetic properties of the material, thereby affecting the gradient distribution. Accordingly, it is recommended to increase the sampling points of magnetometers in key areas, capture gradient changes through high-density data, and use data fusion algorithms to obtain a smoother magnetic field gradient distribution. Alternatively, in combination with real-time data, a magnetic gradient anomaly detection module is established. When a gradient mutation is found, the magnetic sensor parameters of the MWD instrument are adjusted through a compensation algorithm to reduce the impact of the abnormal gradient.

[0109] Regarding the explanation of magnetic field offset and direction distortion faults, the stabilizer interferes with the original geomagnetic field, causing the overall offset or local direction of the magnetic field to change, or the sensor installation angle or calibration error may also cause direction distortion. The suggestion for this type of fault is to complete a comprehensive calibration in a known standard magnetic field environment during the system debugging phase to establish an initial correction model; in actual testing, use an adaptive algorithm to correct the magnetic field offset in real time. In addition, the magnetic field measurement results can be cross-checked in combination with attitude sensor data such as accelerometers and gyroscopes to correct the offset and direction distortion caused by installation angle errors. In addition, a real-time magnetic field compensation module can be developed to automatically adjust the correction coefficient inside the MWD instrument when the overall magnetic field offset is detected to restore it to the actual geomagnetic field state.

[0110] In this way, for various types of magnetic field distortion caused by the stabilizer, it is possible to reduce source interference from aspects such as materials, structural design and hardware shielding, and to achieve real-time detection and online compensation through advanced signal processing, intelligent algorithms and adaptive calibration, thereby improving the accuracy and robustness of the MWD instrument in wellbore trajectory measurement.

[0111] The relevant details of the experimental part of the embodiments of the present application will be expanded below.

[0112] refer to Figure 2 The environmental simulation system shown in the figure consists of a sealed cabin, a temperature control module, a pressure control module, a vacuum module, a main control PLC, an industrial computer, etc. An MWD instrument with a centralizer is placed in the cabin, and the magnetic field data and vibration data of the centralizer are collected under different temperature (30℃~150℃), pressure (1atm~50MPa) and vibration (5~50Hz) conditions.

[0113] Under different combinations of environmental parameters (such as high temperature and high pressure, high temperature and low pressure, low temperature and high pressure, etc.), the magnetic field distribution signal around the stabilizer is obtained through the magnetometer array, and sensor data such as vibration, temperature, and pressure are recorded in real time.

[0114] In order to verify the system more comprehensively, fault simulations such as local defects (such as local material damage and uneven magnetization) and overall structural abnormalities (such as unbalanced assembly) were applied to the stabilizer to produce typical magnetic compatibility anomalies.

[0115] Through the above deployment operations, the environmental simulation chamber test conditions are completed.

[0116] On the other hand, a multi-scale three-dimensional discrete wavelet transform (3D-DWT) and adaptive threshold algorithm module were built on the Simulink platform to read the magnetic field difference data collected by the experimental cabin. Furthermore, this technical solution was compared with the traditional single threshold judgment method, single-scale magnetic field comparison method, and a simplified model lacking multi-physical quantity coupling to observe the differences in local / overall magnetic compatibility anomaly recognition accuracy, false alarm rate, and detection time.

[0117] Through the above deployment operations, the comparative simulation analysis conditions are completed.

[0118] In terms of evaluation indicators, “detection accuracy”, “false alarm rate”, “fault location accuracy” and “average detection time” were selected as indicators.

[0119] The detection accuracy rate expresses the ratio of correctly identifying magnetic compatibility anomalies (including local and overall). The false alarm rate reflects the ratio of being misjudged as abnormal under normal circumstances. The fault location accuracy reflects the positioning deviation of the local abnormal position. The average detection time can reflect the average time from data collection to the completion of abnormality judgment. Therefore, the detection accuracy rate and false alarm rate reflect the robustness and reliability of the algorithm in identifying the type of magnetic compatibility anomalies; the fault location accuracy focuses on the spatial positioning ability of local anomalies; the average detection time measures the availability of the algorithm in online or near real-time scenarios.

[0120] Referring to Table 1, the comparison results of the test method of the embodiment of the present application (denoted as "Proposed") and three existing methods (single threshold method "Single-Threshold", single scale magnetic field comparison method "Single-Scale" and "Basic Fusion" lacking multi-physical quantity coupling) under typical test scenarios (high temperature and high pressure + vibration) are given.

[0121] Table 1 Comparative experimental results of various methods under high temperature, high pressure and vibration conditions As can be seen from Table 1, this technical solution is significantly superior to the other three methods in terms of detection accuracy and false alarm rate, and the fault location accuracy is also greatly improved. This is mainly due to the refined analysis of multi-scale 3D-DWT in local and overall anomaly recognition, as well as the effective distinction of different magnetic field distribution patterns by adaptive attention fusion and threshold adjustment.

[0122] In addition, although the average detection time is slightly higher than other methods, the difference is not significant (about 0.3~0.5s), and it can still meet the near real-time requirements in most MWD instrument testing and fault diagnosis scenarios.

[0123] Figure 6 The figure shows a comparative simulation effect of the test method of the embodiment of the present application and the current single-scale magnetic field comparison method in the local magnetic anomaly identification.

[0124] Figure 6 The horizontal axis of the coordinate system is the spatial position (dm), which indicates the distribution position of the magnetic field signal in space and represents the spatial position inside the actual MWD instrument. The vertical axis of the coordinate system is the magnetic anomaly intensity (normalized), which ranges from [0, 1]. The larger the value, the more obvious the anomaly intensity. The gray area in the coordinate system indicates the real magnetic field anomaly area ("Ground Truth"), which is the distribution range of the actual magnetic anomaly signal, and serves as the ideal output benchmark of the detection algorithm.

[0125] Depend on Figure 6 As can be seen from part (a) of the figure, the detection result of the Single-Scale method is presented as a fluctuating curve. However, the abnormal signal as a whole exhibits smooth fluctuations, and the performance is weak near the edge of the abnormal area. It fails to accurately lock the specific boundary of the abnormal area, which is prone to missed detection or inaccurate range estimation.

[0126] Depend on Figure 6From part (b), we can see that the detected abnormal signal of the proposed method is more consistent with the distribution range of the real abnormality in the gray abnormal area, and the abnormal intensity curve maintains a more obvious distinction. This proves that the proposed method combines multi-scale analysis with an adaptive threshold adjustment mechanism, which is more accurate in abnormal location and abnormal intensity identification, especially in abnormal area boundary identification. As a result, the spatial resolution is improved and false positives and false negatives are reduced.

[0127] It should be noted that although the multi-scale algorithm used in the proposed method slightly increases the amount of calculation, the experimental results show that its impact on real-time performance is relatively controllable and still meets the real-time requirements of most MWD instrument tests. In subsequent work, hardware parallel computing or edge computing modules can be combined to further reduce detection delays.

[0128] It should be noted that, for the aforementioned system embodiments, for the sake of simplicity of description, they are all expressed as a series of combined actions, but those skilled in the art should be aware that the present application is not limited by the order of actions described, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application. In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0129] The system embodiment described above is merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0130] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a general hardware platform, and of course, by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the relevant technology can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the system described in each embodiment or some parts of the embodiment.

[0131] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A MWD instrument centralizer testing and fault diagnosis system, characterized in that: It includes a host computer and an environmental simulation cabin filled with simulated drilling fluid; the environmental simulation cabin is also provided with a bracket and an environmental simulation device; a mounting part for fixing an MWD instrument is provided on the bracket, and a plurality of magnetometers are evenly arranged inside the environmental simulation cabin; The host computer is used to perform the following operations: In the case where the mounting member fixes the first MWD instrument or the second MWD instrument, a preset downhole environmental simulation parameter is sent to the environmental simulation device, so that the environmental simulation device adjusts the internal environment of the environmental simulation cabin according to the downhole environmental simulation parameter, and collects a first magnetic signal for the first MWD instrument or a second magnetic signal for the second MWD instrument based on each magnetometer; the first MWD instrument is an MWD instrument without a centralizer installed, and the second MWD instrument is an MWD instrument with the centralizer installed; Modeling a first magnetic field distribution model for the first MWD tool according to each of the first magnetic signals, and modeling a second magnetic field distribution model for the second MWD tool according to each of the second magnetic signals; Based on the first magnetic field distribution model and the second magnetic field distribution model, the centralizer is tested whether there is magnetic compatibility anomaly.

2. The system according to claim 1, characterized in that The host computer is also used to perform the following operations: Environmental update parameters are sent to the environmental simulation device so that the environmental simulation device adjusts the internal environment of the environmental simulation chamber from the downhole environmental simulation parameters to the environmental update parameters, so as to retest whether there is a magnetic compatibility anomaly in the centralizer under the simulation environment corresponding to the environmental update parameters.

3. The system according to claim 1, characterized in that The system also includes a motor for driving the mounting member to move on the bracket; the host computer is also used to perform the following operations: Generate motor drive instructions according to preset displacement control information; The motor driving instruction is sent to the motor to drive the mounting member to move on the bracket.

4. The system according to claim 1, characterized in that Modeling of magnetic field distribution models, including: The MWD instrument is discretized in spatial geometry and divided into Valid units, of which The volume of an effective unit cell is ; Extract the local average magnetic signal corresponding to each effective unit cell: ， In the formula, Indicates The center position of the effective unit cell, Indicates classification into The local average magnetic field strength, Indicates The number of sensors participating in the statistics in a valid unit, Indicates coverage A magnetometer collection of active units, Indicates the magnetometer set The installation location of the magnetometer, Indicates the position measured by the magnetometer The original magnetic field strength at ; Each effective unit cell is regarded as an approximate magnetic dipole. For the magnetic dipole model, the magnetic field at the target point along the axis of the magnetic dipole is approximately expressed as: ， In the formula, Indicates the location of the target point. express The magnetic field strength at represents the vacuum permeability, represents the magnetic moment of the magnetic dipole; Estimate the mean geometry factor: ， In the formula, Indicates The average geometric factor of the effective element units is used to convert the local magnetic field signal into magnetic moment; Based on the relationship between the local magnetic signal and the geometric factor, we get: ， In the formula, Indicates The effective magnetization intensity of each effective unit cell; So we can solve: ， Under the magnetic dipole approximation, Valid units at the target point The contribution of magnetic field intensity at is recorded as: ， In the formula, Indicates The effective unit is The contribution of magnetic field strength generated at Indicates from arrive Green's function, used to describe the attenuation and directionality of magnetic field propagation; In the magnetic dipole approximation, the Green's function is expressed as: ， In the formula, Indicates The unit vector of the magnetization direction of an effective unit cell characterizes the main magnetization direction of the magnetic material; The magnetic field intensity contribution of each target point in all effective units is superimposed to obtain the magnetic field distribution model of the MWD instrument: ， In the formula, The magnetic field distribution model at the target point The magnetic field strength is expressed.

5. The system according to claim 4, characterized in that The testing whether the centralizer has magnetic compatibility anomaly based on the first magnetic field distribution model and the second magnetic field distribution model includes: The first magnetic field distribution model and the second magnetic field distribution model are compared based on a multi-scale analysis method of a magnetic field difference spectrum to test whether the centralizer has a magnetic compatibility anomaly, specifically including: According to the magnetic field differences of each target point, a magnetic field difference map in three-dimensional space is constructed: ， In the formula, Indicates that the first magnetic field distribution model is at the target point The magnetic field strength, and Indicates that the second magnetic field distribution model is at the target point The magnetic field strength; Indicates the target point The magnetic field intensity difference at each target point is calculated, and then a magnetic field difference map in three-dimensional space is constructed according to the magnetic field intensity difference of all target points in the space. The three-dimensional discrete wavelet transform is used to decompose the magnetic field difference map to obtain multi-scale wavelet coefficients: ， In the formula, the target point r The position is determined by the three-dimensional discrete space grid coordinates To express, Represents the magnetic field difference map against the position In the The scale of The wavelet coefficients in the directions, Indicates the target point r The three-dimensional coordinates of the neighborhood space position; is the total number of neighborhood points analyzed in the direction of the corresponding coordinate axis, which represents the spatial discretization accuracy used for magnetic field analysis; Indicates The scale of The mother wavelet function in each direction; Based on the multi-scale wavelet coefficients, the local magnetic field anomaly intensity index and the global magnetic field anomaly energy index are calculated to construct a multi-scale magnetic anomaly feature matrix: The local magnetic field anomaly intensity index is calculated by the following formula: ， In the formula, Indicates Position at scale The local magnetic field anomaly intensity index, Indicates Position at scale The wavelet coefficient vector in all directions; The global magnetic field anomaly energy index is calculated by the following formula: ， In the formula, Indicates The global magnetic field anomaly energy index of the scale, It represents the sum of the positions of all target points in the space. It means summing all analysis directions; The multi-scale magnetic anomaly characteristic matrix is ​​calculated by the following formula: ， In the formula, represents the multi-scale magnetic anomaly feature matrix, Indicates the total number of scales; Based on the multi-scale magnetic anomaly feature matrix, the corresponding magnetic compatibility anomaly threshold is adaptively determined: ， ， In the formula, and Corresponding to The spatial statistical mean and standard deviation of the local magnetic field anomaly intensity index corresponding to all target points within a scale, is the local abnormal sensitivity coefficient, is the global abnormal sensitivity coefficient, For the The local scale magnetic compatibility anomaly threshold of scales, For the The global scale magnetic compatibility anomaly threshold of scales, and Corresponding to The spatial statistical mean and standard deviation of the global magnetic field anomaly intensity index corresponding to all target points within a scale; is the adaptive fine-tuning coefficient, and They represent the first The local base threshold and global base threshold of each scale; Compare the magnetic compatibility anomaly threshold and the multi-scale magnetic anomaly feature matrix to determine whether there is a magnetic compatibility anomaly: If detected in the multi-scale magnetic anomaly feature matrix , then the judgment scale Lower position There is a local magnetic compatibility anomaly; if it is detected in the multi-scale magnetic anomaly feature matrix , then the judgment scale There is an overall magnetic compatibility anomaly.

6. The system according to claim 5, characterized in that After comparing the first magnetic field distribution model and the second magnetic field distribution model based on the multi-scale analysis method of the magnetic field difference map to test whether the centralizer has magnetic compatibility anomaly, the host computer is further used to perform the following operations: When it is determined that the centralizer has a magnetic compatibility anomaly, an abnormal input feature vector is constructed according to the abnormal local magnetic field anomaly intensity index corresponding to the local magnetic compatibility anomaly, the global magnetic field anomaly energy index corresponding to the overall magnetic compatibility anomaly, and the spatial anomaly feature code corresponding to the local magnetic compatibility anomaly position: ， ， In the formula, represents the abnormal input feature vector, Indicates the location of local magnetic compatibility anomaly The corresponding spatial anomaly feature code, express The three-dimensional coordinates of and The scale Lower position The mean and standard deviation of the local magnetic field anomaly intensity index corresponding to each neighborhood spatial position; The abnormal input feature vector is input into an abnormal type recognition model to determine the corresponding magnetic compatibility fault type; the abnormal type recognition model adopts a multi-scale adaptive attention convolutional network and includes a cascaded input layer, a multi-scale convolutional feature extraction layer, an adaptive attention fusion layer, a feature enhancement and dimensionality reduction layer, and a classification output layer: The input layer is used to receive the input feature vector; The multi-scale convolution feature extraction layer is used to use multiple parallel three-dimensional convolution layers to extract multi-scale local features of the magnetic field difference feature map with convolution kernels of different scales: ， In the formula, Indicates the scale of extraction The abnormal spectrum characteristics of represents the convolution operation, and Respectively represent The convolution kernel weights and bias parameters of each scale, Represents the RELU nonlinear activation function; The adaptive attention fusion layer is used to effectively fuse multi-scale local feature maps through an adaptive scale attention mechanism to automatically capture the most discriminative magnetic field anomaly scale features: ， ， In the formula, It represents the magnetic anomaly scale fusion feature obtained by fusing feature maps of different scales. represents the number of convolution kernels, Indicates Adaptive attention weights at each scale, represents the softmax function, represents the hyperbolic tangent function, is the trainable parameter of the attention mechanism, express The transposed vector of ; The feature enhancement and dimensionality reduction layer is used to further enhance and reduce the fused features through a fully connected network to extract key high-distinguishing features: ， In the formula, represents the dimension reduction enhanced feature vector, and Represent the weight matrix and bias parameters of the fully connected network respectively; The classification output layer is used to output the magnetic compatibility anomaly type through the softmax layer: ， In the formula, represents the probability vector of magnetic compatibility anomaly type classification, each element of which represents the predicted probability of the corresponding anomaly type; and They represent the weight matrix and bias parameters of the classification output layer respectively.

7. The system according to claim 6, characterized in that The magnetic compatibility fault type includes at least one of the following: a local magnetic field abnormal enhancement fault, a local magnetic field abnormal attenuation fault, a magnetic field gradient abnormal fault, and a magnetic field offset and direction distortion fault.

Images