A method and system for formation identification and adaptive parameter adjustment while drilling

Abstract

The application provides a drilling formation identification and adaptive parameter adjustment method and system, through constructing a formation response mode library, multi-source sensing data is converted into a structured formation feature representation, accurate identification of formation interface distribution is realized; by using parameter response area distribution analysis and missing information reconstruction technology, implicit formation features are deeply mined, and the internal law of formation change is revealed; by introducing the optimization selection of adjustment points and the reinforcement drilling path planning mechanism, an efficient drilling control network is constructed; by constructing the potential field distribution through the parameter change characteristics, the coupling correlation between the parameter characteristics and the spatial position is established; combined with the analysis of the harmonic force and the generation of the execution scheme, intelligent configuration and adaptive adjustment of the drilling parameters are realized; finally, a complete drilling formation identification and parameter control system is formed, which can provide accurate and efficient technical solutions for drilling operations in complex formation environment, and improve drilling efficiency and operation quality.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of drilling engineering, in particular to a drilling formation identification and adaptive parameter adjustment method and system. BACKGROUND

[0002] As a core technical means of geological exploration, resource development and engineering construction, drilling operation has always been a key challenge for the development of the industry in terms of accurate identification and adaptive control in complex formation environments. Modern drilling engineering often faces technical problems such as complex and variable formation structure, frequent lithology mutation, and difficulty in real-time optimization of drilling parameters, which directly affect drilling efficiency, cost control and operation safety.

[0003] Traditional drilling parameter adjustment methods mainly rely on the experience of operators and simple resistance feedback control, lacking depth understanding and systematic analysis of formation characteristics. Existing technologies cannot effectively process complex information from multi-source sensor data, making it difficult to identify implicit formation characteristics, let alone achieve intelligent and adaptive adjustment of drilling parameters. Especially in the case of complex formation interface distribution and dramatic impedance changes, traditional methods often result in low drilling efficiency, increased equipment wear and tear, and unstable drilling quality.

[0004] Therefore, there is an urgent need for a method to solve at least one of the above problems. SUMMARY

[0005] The present application provides a drilling formation identification and adaptive parameter adjustment method and system, aiming to achieve accurate identification of complex formation structure through complex processing of multi-source sensor data and construction of formation response mode library; reveal the internal change law of the formation through parameter response area distribution analysis and implicit formation characteristic reconstruction; establish the coupling correlation mechanism between drilling parameters and formation characteristics through eddy current parameter sequence transformation and potential field distribution construction; and finally realize intelligent generation of drilling execution scheme and adaptive adjustment of parameters.

[0006] The present application provides a drilling formation identification and adaptive parameter adjustment method and system, aiming to achieve accurate identification of complex formation structure through complex processing of multi-source sensor data and construction of formation response mode library; reveal the internal change law of the formation through parameter response area distribution analysis and implicit formation characteristic reconstruction; establish the coupling correlation mechanism between drilling parameters and formation characteristics through eddy current parameter sequence transformation and potential field distribution construction; and finally realize intelligent generation of drilling execution scheme and adaptive adjustment of parameters.

[0007] Collecting multi-source sensor data during drilling, the multi-source sensor data including vibration signals and resistance data, and performing complex processing on the multi-source sensor data to construct a formation response mode library;

[0008] Performing parameter response area distribution analysis based on the formation response mode library to generate a regional distribution atlas, extracting a parameter response blank area from the regional distribution atlas, reconstructing missing information for the parameter response blank area to obtain an implicit formation characteristic, and establishing a set of adjustment node candidates based on the regional distribution atlas and the implicit formation characteristic;

[0009] comprehensive evaluation on the alternative adjusting node set to obtain parameter action range and adjusting potential value of each alternative adjusting node, determine a drilling core adjusting node according to the parameter action range and the adjusting potential value, extract a high impedance core area and a low impedance transition area from the drilling core adjusting node;

[0010] plan a reinforced drilling path for the high impedance core area, formulate a gradual adjusting direction for the low impedance transition area, construct a conflict intersection point of the reinforced drilling path and the gradual adjusting direction, perform reconciliation force analysis on the conflict intersection point to establish a main adjusting path, and obtain an auxiliary adjusting flow based on the main adjusting path;

[0011] convert the auxiliary adjusting flow into a vortex parameter sequence, convert the main adjusting path into a drilling parameter adjusting sequence, and perform fusion processing based on the vortex parameter sequence and the drilling parameter adjusting sequence to generate parameter change characteristics;

[0012] construct a drilling potential field distribution based on the parameter change characteristics, collect dispersed stress from the drilling potential field distribution to generate concentrated stress data, and generate a final parameter combination by using the concentrated stress data.

[0013] The second aspect of the present application provides a drilling stratum identification and adaptive parameter adjusting system, comprising:

[0014] a data acquisition module configured to acquire multi-source sensing data in a drilling process, wherein the multi-source sensing data comprises vibration signals and resistance data, and perform composite processing on the multi-source sensing data to construct a stratum response mode library;

[0015] a region analysis module configured to perform parameter response region distribution analysis based on the stratum response mode library to generate a region distribution atlas, extract a parameter response blank region from the region distribution atlas, perform missing information reconstruction on the parameter response blank region to obtain implicit stratum characteristics, and establish an adjusting node alternative set based on the region distribution atlas and the implicit stratum characteristics;

[0016] a core identification module configured to perform comprehensive evaluation on the alternative adjusting node set to obtain parameter action range and adjusting potential value of each alternative adjusting node, determine a drilling core adjusting node according to the parameter action range and the adjusting potential value, and extract a high impedance core area and a low impedance transition area from the drilling core adjusting node;

[0017] a path planning module configured to plan a reinforced drilling path for the high impedance core area, formulate a gradual adjusting direction for the low impedance transition area, construct a conflict intersection point of the reinforced drilling path and the gradual adjusting direction, perform reconciliation force analysis on the conflict intersection point to establish a main adjusting path, and obtain an auxiliary adjusting flow based on the main adjusting path;

[0018] a parameter processing module, configured to convert the auxiliary regulating flow into an eddy current parameter sequence, convert the main regulating path into a drilling parameter regulating sequence, and generate a parameter variation feature by performing a fusion process based on the eddy current parameter sequence and the drilling parameter regulating sequence;

[0019] A parameter generation module is used to construct a drilling potential field distribution based on the parameter change characteristics, collect dispersed stress from the drilling potential field distribution to generate concentrated stress data, and use the concentrated stress data to generate a final parameter combination.

[0020] The beneficial effects of the present invention are reflected in the following aspects: First, a formation response pattern library is constructed through the composite processing of multi-source sensor data, integrating vibration signals and resistance data to form a complete formation feature identification system. This technology can accurately identify the distribution of formation interfaces and distinguish different types of formations, such as loose formations, dense formations, and hard rock formations, providing a reliable geological basis for drilling parameter adjustment. Furthermore, by analyzing the regional distribution of parameter responses and reconstructing implicit formation features, it can identify features of blank areas that are difficult to detect using traditional methods, thereby expanding the coverage and accuracy of formation identification. Second, a mechanism for demarcating high-impedance core areas and low-impedance transition areas is established, enabling the formulation of differentiated drilling strategies for different regions. By strengthening drilling path planning and formulating progressive adjustment directions, a systematic drilling control scheme is formed. Combined with the analysis of the harmonic force at conflicting intersections, conflicts between different adjustment strategies can be effectively resolved, ensuring the continuity and stability of the drilling process and improving the adaptability and control accuracy of drilling in complex formation environments. Finally, intelligent configuration and adaptive adjustment mechanisms for drilling parameters are implemented. By transforming eddy current parameter sequences and constructing potential field distributions, a correlation between parameter variation characteristics and spatial position is established. A drilling execution plan is generated based on the matching correlation matrix, automatically adjusting key parameters such as drilling power, feed rate, rotational torque, and axial pressure based on real-time formation changes. This adaptive control mechanism reduces the need for manual intervention, improves drilling efficiency, and reduces equipment wear and operational risks.

[0021] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The accompanying drawings herein illustrate specific examples of the technical solutions described in the present invention, and together with the specific implementation methods constitute a part of the specification, and are used to explain the technical solutions, principles and effects of the present invention.

[0023] Unless otherwise specified, the same reference numerals in different drawings represent the same or similar technical features, and the same or similar technical features may also be represented by different reference numerals.

[0024] Figure 1is a flowchart of a drilling formation identification and adaptive parameter adjustment method of the present application.

[0025] Figure 2 is a structural block diagram of a drilling formation identification and adaptive parameter adjustment system of the present application. DETAILED DESCRIPTION

[0026] In the following description, for purposes of explanation and not limitation, specific details are set forth such as particular architectures, techniques, etc. in order to provide a thorough understanding of the embodiments of the present application. However, it will be apparent to those skilled in the art that the present application can be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known methods, devices, circuits, and

[0027] It is to be understood that the terminology "including", "comprising", "consisting" and "consisting essentially of" used in the specification and the appended claims, indicates the presence of the stated features, integers, steps, operations, elements, and / or components, but does not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0028] Reference throughout this specification to "one embodiment" or "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the application. Thus, the appearances of the phrases "in one embodiment" or "in an embodiment" or "in other embodiments" or "in some embodiments" in various places throughout this specification are not necessarily referring to the same embodiment, unless otherwise specified. The terms "including", "comprising", "consisting essentially of" and "consisting of" are used interchangeably, unless otherwise specified.

[0029] The technical solutions of the embodiments of the present application are introduced as follows.

[0030] As shown in Figure 1 The present application provides a drilling formation identification and adaptive parameter adjustment method, which comprises the following steps S110-S160.

[0031] In step S110, multi-source sensing data in the drilling process is collected, the multi-source sensing data comprising vibration signals and resistance data, and the multi-source sensing data is subjected to composite processing to construct a formation response mode library.

[0032] Specifically, multi-source sensing data is collected by deploying a multi-type sensor array at key positions of the drilling equipment. The vibration signal collection adopts a combination configuration of three-axis acceleration sensors and speed sensors. The acceleration sensors are arranged at three positions, i.e., the top of the drill pipe, the middle of the drill pipe, and the vicinity of the drill bit. The sampling frequency is set to 2 kHz, and the measurement range is ±50 g, which is used to capture high-frequency impact vibration in the drilling process. The speed sensor is installed at the connection between the main shaft of the drilling machine and the drill pipe, with a sampling frequency of 1 kHz, which measures the overall vibration speed change of the drill pipe. The resistance data collection is realized by a torque sensor and an axial force sensor. The torque sensor has a range of 0-5000 N·m and an accuracy of 0.1%, which monitors the rotational resistance change in the drilling process in real time. The axial force sensor has a range of 0-200 kN and a sampling frequency of 500 Hz, which measures the axial feed resistance of the drill bit. The data acquisition system uses a 16-bit A / D converter to ensure signal accuracy, and the data transmission uses a CAN bus protocol with strong anti-interference capability. The vibration signal data format is three-axis vibration components, corresponding to the vibration responses of X, Y, and Z axes, respectively. The resistance data includes torque sequences and axial force sequences, and the sampling interval is uniformly set to 0.5 ms. The original data is subjected to low-pass filtering to remove high-frequency noise, with a cutoff frequency of 500 Hz and an 8th-order Butterworth filter. The synchronization collection mechanism of multi-source sensing data is realized by a GPS clock reference, with a time synchronization accuracy of 1 μs.

[0033] In some embodiments, the composite processing of the multi-source sensing data constructs a formation response mode library, including: generating a formation response spectrum by exciting the vibration signal by frequency; identifying the formation interface distribution by frequency band disassembly of the formation response spectrum; constructing a formation response envelope line based on the formation interface distribution and the resistance data; and establishing a formation response mode library based on the formation response envelope line.

[0034] The formation response spectrum is generated by frequency excitation on the vibration signal. A linear sweep signal is used as the excitation source, with a sweep range of 10-1000 Hz and a sweep period of 2 s. The expression of the excitation signal is s(t) = A sin(2p(f0+kt)t), where A is the excitation amplitude, f0is the initial frequency of 10 Hz, and k is the sweep rate. The excitation signal is applied to the formation interface through the drill bit vibration exciter, while the drill pipe vibration response signal is collected. The formation response spectrum generation uses the transfer function analysis method to calculate the frequency domain transfer function H(ω) = Y(ω) / X(ω) of the excitation signal and the response signal, where X(ω) is the excitation signal spectrum and Y(ω) is the vibration response spectrum extracted from the characteristic parameter set. The transfer function calculation uses the Welch power spectrum estimation method, with a Hanning window, an overlap rate of 50%, and a frequency resolution of 0.5 Hz. The formation response spectrum contains two components: amplitude spectrum and phase spectrum. The amplitude spectrum reflects the attenuation characteristics of the formation to different frequency vibrations, and the phase spectrum reflects the propagation delay characteristics. The spectrum smoothing process uses a 1 / 3 octave smoothing algorithm to eliminate random fluctuations in the spectrum. The response spectrum data format is a frequency-amplitude correspondence, with 2000 frequency points covering the entire excitation frequency range.

[0035] The formation response spectrum is processed by frequency band decomposition to identify the distribution of different formation interfaces. The empirical mode decomposition (EMD) algorithm is used to decompose the complex response spectrum into multiple intrinsic mode function (IMF) components. The EMD decomposition process is achieved through iterative screening: first, identify the local maximum and minimum points in the spectrum, construct the upper and lower envelope lines, calculate the average envelope line, and subtract the average envelope line from the original spectrum to obtain the first IMF component. Repeat this process until the residual component becomes a monotonic function. The decomposition result is expressed as where IMF i is the i-th intrinsic mode function, r(ω) is the residual term, and m is the number of IMF components. The formation interface identification is based on the frequency characteristics of different IMF components: high-frequency IMF components (500-1000 Hz) correspond to shallow loose formation interfaces, medium-frequency components (100-500 Hz) correspond to medium-density formation interfaces, and low-frequency components (10-100 Hz) correspond to deep hard rock formation interfaces. The interface distribution characteristics are quantitatively represented by the energy distribution of each IMF component, and the energy proportion reflects the thickness and density characteristics of the corresponding formation type. The interface depth positioning is achieved through phase spectrum analysis, and the interface depth d = v·τ / 2 is calculated according to the propagation delay time of different frequency components, where v is the wave velocity in the formation and τ is the propagation delay time. The interface distribution result is represented as a depth-formation type correspondence, containing the main formation interface information.

[0036] Based on the formation interface distribution and resistance data, the formation response envelope is constructed. The resistance interpolation method constrained by the formation interface is adopted, and the torque sequence and axial force sequence collected in the first section are segmented according to the interface depth information identified in the fourth section. For each formation section, the torque time sequence and axial force time sequence in the corresponding depth range are extracted from the original resistance data for analysis and processing according to the depth interval division in the interface distribution. The resistance envelope is composed of the upper envelope and the lower envelope. The upper envelope connects the peak points of resistance in each depth section, and the lower envelope connects the valley points. The envelope fitting adopts the cubic spline interpolation algorithm to ensure the continuity and smoothness of the envelope. The expression of the formation response envelope is as follows: , the lower envelope , wherein T(d) is the torque value at depth d extracted from the original torque sequence, F(d) is the axial force value at depth d extracted from the original axial force sequence, and α and β are the weight coefficients of torque and axial force, which are determined according to the formation type identified in the interface distribution: α = 0.8, β = 1.2 for loose formation, α = 1.0, β = 1.0 for dense formation, and α = 1.2, β = 0.8 for hard rock formation. The envelope bandwidth W(d) = Rupper(d) - Rlower(d) reflects the degree of change in formation resistance, and a large bandwidth indicates strong formation heterogeneity. The envelope characteristic parameters include average resistance, resistance gradient, envelope area, and other key indicators.

[0037] Based on the constructed formation response envelope, a classified and stored formation response pattern library is established. A relational database architecture is adopted, with formation type and depth interval as the main index. The database includes four core tables: the formation type table stores the standard characteristic parameters of loose formation, dense formation, hard rock formation, etc.; the depth distribution table records the distribution of each formation type in different depth intervals; the response envelope table stores the upper and lower envelope values and characteristic parameters of each depth point; and the pattern matching table establishes the correspondence between envelope characteristics and formation types. The standard patterns in the pattern library are established through statistical analysis: the response envelope of the same type of formation is subjected to cluster analysis, and the typical envelope shape is extracted as the standard pattern. The clustering algorithm adopts the K-means method, and the clustering number k = 3 corresponds to loose formation, dense formation, and hard rock formation. The distance measurement adopts the dynamic time warping (DTW) algorithm, and the constraint bandwidth is set to 10% of the sequence length, which is suitable for similarity calculation of envelope sequences of different lengths. Each standard pattern contains pattern identification, formation type label, depth range, characteristic envelope sequence, statistical characteristic parameters, and other information. The pattern library contains multiple standard patterns, covering the response characteristics of common formation types.

[0038] At step S120, a parameter response area distribution analysis is performed based on the formation response mode library to generate an area distribution atlas, a parameter response blank area is extracted from the area distribution atlas, implicit formation characteristics are obtained by reconstructing missing information of the parameter response blank area, and a tuning node candidate set is established based on the area distribution atlas and the implicit formation characteristics.

[0039] Specifically, first, the drilling area is discretized in space according to a 100m*100m grid to form a two-dimensional analysis grid. For each grid node, a matching standard mode is retrieved from the formation response mode library, and the matching criteria are based on three dimensions of formation type, depth range, and response envelope similarity. The mode matching algorithm uses a minimum distance classifier to calculate the Euclidean distance between the measured response and each standard mode in the mode library, and selects the mode with the smallest distance as the matching result. Spatial interpolation processing is performed on grid nodes without direct drilling data, and the Kriging interpolation method is used, with the weight function being determined based on spatial distance and geological continuity assumption. The area distribution atlas is constructed using a hierarchical coloring method, with different formation types represented by different colors: loose formation is yellow, dense formation is green, and hard rock layer is red, and the color depth reflects the response intensity. The atlas resolution is set to 1m*1m, covering the entire drilling operation area. The parameter response intensity is quantified by comprehensive scoring, with a score range of 0-100, considering factors such as resistance envelope area, spectral energy distribution, and interface clarity. The atlas data structure is stored in raster format, with each pixel containing coordinate position, formation type, response intensity, matching confidence, and other attribute information. Through mode library matching and spatial analysis techniques, the area distribution atlas is obtained.

[0040] The parameter response blank area is extracted from the area distribution atlas. A threshold segmentation and connectivity analysis method is used, first setting the response intensity threshold to 20 (full score 100), and marking pixels below the threshold as potential blank areas. The connectivity analysis uses an 8-neighbor connectivity algorithm to merge adjacent low-response pixels into connected regions, and filter out fragment areas with an area less than 25 square meters to avoid noise interference. The geometric feature extraction of the blank area includes area, perimeter, shape factor, and direction angle, and the shape factor is defined as 4π*area / perimeter ² , which is used to quantify the regularity of the area. The blank area classification is based on shape and distribution characteristics: strip-shaped blank areas usually correspond to geological faults or weak interlayers, elliptical-shaped blank areas may indicate caves or soft soil areas, and irregular-shaped blank areas often represent complex geological structures. Each blank area is assigned a unique identifier, recording its boundary coordinate sequence, geometric parameters, surrounding formation background, and other information. The boundary smoothing of the blank area uses B-spline curve fitting to eliminate boundary sawtooth effects.

[0041] In some embodiments, the parameter response blank area is subjected to missing information reconstruction to obtain implicit formation characteristics, including: obtaining boundary features of the parameter response blank area; inferring internal response rules of the parameter response blank area based on the boundary features; generating compensation response data through data interpolation processing based on the response rules; and reconstructing implicit formation characteristics based on the compensation response data.

[0042] The boundary features of the boundary coordinate sequence of each blank area are obtained in advance, and a method combining curvature analysis and gradient calculation is adopted. The boundary curvature calculation adopts a three-point method. For any point P i on the boundary, the local curvature κi=|det(P i-1 -P i+1 , P i-1 -P i )| / |P i+1 -P i |³ is calculated using adjacent points P i+1 and P i-1 . The curvature size reflects the bending degree of the boundary. The response gradient calculation is determined by the difference in response intensity inside and outside the boundary, and the gradient direction points to the direction in which the response intensity increases. The boundary feature vector contains four components of position coordinates, curvature value, gradient size and gradient direction, forming a complete description of the boundary. The boundary segmentation processing is based on the curvature change, and the continuous boundary is divided into three types of high-curvature segment, low-curvature segment and turning segment, different types of boundary segments correspond to different geological meanings. The high-curvature segment may correspond to sharp stratum change, the low-curvature segment represents relatively flat transition, and the turning segment often indicates a structure control point. The statistical analysis of the boundary features includes parameters such as average curvature, maximum gradient and main direction angle, which are used to represent the boundary complexity and anisotropy characteristics of the blank area. The boundary feature data is organized according to the boundary segment, and each boundary segment contains start and end positions, feature parameters, geological interpretation and other information.

[0043] The internal response rule inference adopts a field theory method under boundary constraints, and the blank area is regarded as a two-dimensional potential field, and the boundary condition is determined by the boundary features. The potential field equation adopts Laplace equation The description is given, where φ is a response potential function, the boundary condition is φ|boundary = f(boundary feature). The boundary response intensity is estimated according to gradient information: for the boundary point, the response intensity is equal to the response intensity of the adjacent non-blank area minus the product of the gradient size and the distance. The internal response distribution is assumed to follow the distance attenuation law, and the farther away from the boundary, the smaller the boundary influence. The attenuation function adopts an exponential form r(d) = r0·exp(-d / λ), where r0 is the boundary response intensity, d is the distance to the boundary, and λ is the attenuation constant. The attenuation constant λ is determined according to the geological type of the blank area: λ = 10 m for fault type, λ = 5 m for cave type, and λ = 15 m for soft soil type. The superposition of the influence of multiple boundaries adopts a linear weighting method, and the weight is inversely proportional to the distance. The anisotropy of the response distribution is realized by modulation of the main direction angle, and the attenuation speed is slower along the main direction and faster perpendicular to it. The internal response law is expressed in the form of a mathematical model, including three core elements of boundary constraint, distance attenuation and anisotropy.

[0044] The response law is used to generate compensated response data by data interpolation processing. A radial basis function method based on physical constraints is used to combine the inferred mathematical model with the measured boundary data for calculation. The boundary constraint value is obtained by extracting the average response intensity of the non-blank area within 3-5 meters outside the blank area boundary in the regional distribution map, ensuring the reliability of the boundary condition. The radial basis function selects a multi-quadratic function φ(r) = √(r²+c²), where r is the spatial distance and c is the shape parameter. The interpolation function expression is where wi is the weight coefficient, and P(x,y) is the polynomial term. The weight coefficient is determined by solving the linear equation set under the boundary constraint condition, and the boundary response intensity extracted from the regional distribution map is taken as the known value, combined with the inferred response distribution law as the internal constraint, to ensure that the interpolation function meets the measured response value on the boundary and follows the inferred distribution law in the interior. The interpolation accuracy control is realized by grid densification, and the initial interpolation grid is 5m x 5m, which is adaptively encrypted to 1m x 1m for areas with large gradients. The compensated response data format is consistent with the original regional distribution map, including position coordinates, response intensity, confidence, etc.

[0045] The implicit formation characteristics are reconstructed based on the compensated response data. Firstly, the feature extraction is performed on the compensated response data, including local response intensity, spatial gradient, second derivative, local variance and other dimensions, to form a feature vector of each grid point. The K-means algorithm is adopted for clustering analysis, and the number of clusters K is adaptively determined according to the complexity of the blank area, generally taking 3-7 categories. The clustering results correspond to different implicit formation units, each unit having similar response characteristics. The implicit formation characteristic parameters include average response intensity, response variability, spatial distribution pattern, and association with surrounding formations. The formation unit boundary identification is realized through spatial connectivity analysis of the clustering results to form the geometric outline of the implicit formation. The geological interpretation of the implicit characteristics is based on the comparative analysis of the response characteristics and known formation types: the area with low response intensity and small variability may correspond to homogeneous soft soil layer, and the area with high response intensity and large variability may exist hidden bedrock protrusion.

[0046] The candidate set of adjustment points is established based on the regional distribution atlas and the implicit formation characteristics. The preliminary screening of candidate points is performed in the regional distribution atlas, and the positions with large response intensity gradient are selected as potential adjustment points, and the gradient threshold is set to 10 units / m. The integration of implicit formation characteristics is realized by weighted superposition, and the weight of implicit characteristics is determined according to the reconstruction reliability, the weight of high reliability area is 0.8, the weight of medium reliability is 0.5, and the weight of low reliability is 0.2. The evaluation indexes of adjustment points include three dimensions of formation representativeness, spatial distribution uniformity and technical operability. The formation representativeness is quantified by the diversity of formation types around the adjustment points, and the diversity index is calculated by Shannon entropy H=-Σp i log(p i ), where p i is the area ratio of the i-th formation type. The spatial distribution uniformity is evaluated by nearest neighbor distance analysis to avoid excessive concentration or dispersion of adjustment points. The technical operability considers the engineering factors such as terrain conditions, traffic convenience and equipment accessibility. The optimization of candidate set adopts genetic algorithm, and the objective function is the maximization of comprehensive evaluation index, and the constraint conditions include the number limit of adjustment points and the minimum distance requirement. Through the comprehensive analysis of regional distribution atlas and implicit formation characteristics, the candidate set of adjustment points is successfully established.

[0047] In step S130, the comprehensive evaluation of the candidate set of adjustment points is performed to obtain the parameter action range and adjustment potential value of each candidate adjustment point, and the drilling core adjustment point is determined according to the parameter action range and adjustment potential value, and the high impedance core area and low impedance transition area are extracted from the drilling core adjustment point.

[0048] Specifically, the alternative adjustment nodes are systematically analyzed and calculated by using the adjustment node alternative set. The influence domain calculation method is used to obtain the parameter action range, and the influence radius is calculated according to the stratum propagation characteristics with each alternative adjustment node as the center. The influence radius calculation formula is R=k√(P / ρ), wherein k is the stratum coefficient, P is the stratum representative evaluation score of the adjustment node, and p is the surrounding stratum density. The stratum coefficient is determined according to the stratum background of the adjustment node: k=1.2 for loose stratum, k=1.0 for dense stratum, and k=0.8 for hard rock stratum. The influence domain overlap analysis is realized by calculating the intersection area of the influence circles of adjacent adjustment nodes, and the adjustment nodes with an overlap rate of more than 30% have mutual interference. The parameter action range quantization adopts the effective action area index, and the net influence area after deducting the overlapping part is taken as the actual action range of the adjustment node. The adjustment potential value is obtained by multi-factor fusion processing, which comprehensively considers factors such as stratum complexity, response change amplitude, construction difficulty, etc. The stratum complexity is quantified by the Shannon entropy of the stratum type around the adjustment node, the response change amplitude is represented by the standard deviation of the local response strength, and the construction difficulty is comprehensively calculated based on engineering parameters such as terrain slope, rock hardness, and equipment accessibility. The adjustment potential value calculation formula is V=w1xC+w2xR+w3x(1-D), wherein C is the stratum complexity, R is the response change amplitude, D is the construction difficulty, and the weight coefficients w1=0.4, w2=0.4, w3=0.2. The coordinate position, action range, potential value, and influence domain overlap of each alternative adjustment node are recorded.

[0049] The drilling core adjustment nodes are determined according to the parameter action range and the adjustment potential value. The core adjustment node selection follows the principle of maximum coverage and minimum redundancy, and the adjustment node combination with the highest potential value is selected on the premise of ensuring full coverage of the action range. The optimization problem is modeled as an integer programming form, the target is to maximize the total potential value of the adjustment nodes, and the constraint conditions include the regional coverage requirement and the adjustment node number limit. The genetic algorithm is used for solving algorithm, the population size is 100, the crossover probability is 0.8, the mutation probability is 0.1, and the iteration is 500 generations. The fitness function is designed as the weighted combination of the objective function value and the constraint violation degree to ensure the feasibility of the solution. The solution diversity is maintained by the niche technology to avoid premature convergence. The core adjustment node screening result usually contains 8-12 adjustment nodes, which ensures sufficient representativeness and facilitates engineering implementation. Each core adjustment node retains its original coordinate position, stratum background, evaluation score, and other attribute information, and adds core level, coverage responsibility area, and other identifiers.

[0050] In some embodiments, the extracting the high-resistivity core zone and the low-resistivity transition zone from the drilling core node comprises: performing impedance gradient operation on the drilling core node to obtain an impedance variation rate distribution; separating a high-gradient component and a low-gradient component from the impedance variation rate distribution; performing ratio analysis on the high-gradient component and the low-gradient component to obtain an impedance distribution feature; and dividing the high-resistivity core zone and the low-resistivity transition zone according to the impedance distribution feature.

[0051] The impedance gradient operation is performed on the drilling core node to obtain the impedance variation rate distribution. First, the response intensity value of each core node is extracted from the regional distribution map, and the response intensity is converted into an impedance value according to the linear conversion relationship between the response intensity and the formation impedance, wherein the conversion coefficient is determined according to the formation type, and the reference impedance value is set as the regional average value. Using the finite difference method, for the node i, the x-direction gradient is , and the y-direction gradient is , wherein Z i is the impedance value of the node i, and Δx and Δy are the spatial intervals. The gradient amplitude is obtained by vector composition method, and the gradient direction angle is calculated by arctangent function. In order to obtain a continuous gradient distribution field, the discrete node gradient value is spatially interpolated, and the inverse distance weighted method is used for interpolation. The resolution of the gradient distribution field is set to 10 m x 10 m, covering the influence area of all core nodes. The gradient field smoothing processing adopts the mean filter to eliminate high-frequency noise in the calculation process. The gradient statistical features include the maximum gradient value, the average gradient value, the gradient variance and other parameters. The spatial pattern of the impedance variation rate distribution is visualized by the gradient rose diagram, which shows the directional characteristics of the gradient.

[0052] From the obtained impedance variation rate distribution, the high-gradient component and the low-gradient component are separated by using the frequency domain analysis method. First, the two-dimensional fast Fourier transform (FFT) is performed on the gradient distribution to obtain the frequency domain representation. The spectrum analysis shows that the high-gradient component is mainly distributed in the high frequency band, corresponding to local sharp changes; the low-gradient component is concentrated in the low frequency band, reflecting regional slow changes. The separation threshold is determined by spectrum energy analysis, and the frequency at which the cumulative energy reaches 80% is taken as the high-low frequency demarcation point. The high-gradient component is extracted by using a high-pass filter, and the low-gradient component is obtained by using a low-pass filter. The filtered frequency domain signal is transformed back to the spatial domain by inverse FFT to obtain the separated high-gradient component and low-gradient component. The component separation effect is evaluated by reconstruction error, and the correlation coefficient between the reconstructed signal and the original gradient field should be greater than 0.95. The high-gradient component usually corresponds to geological structure boundaries such as faults and joints, and the low-gradient component reflects the overall change trend of different formation units. The separated components maintain the original spatial resolution and coordinate system.

[0053] The high gradient component and the low gradient component obtained by separation are analyzed by ratio analysis to obtain impedance distribution characteristics. For each position in space, the gradient ratio R(x, y) = G_high(x, y) / G_low(x, y) is calculated by point-by-point calculation. In the ratio calculation, the low gradient component is subjected to non-zero constraint. When the low gradient component is too small, a smooth replacement value is used to avoid division by zero error. Statistical analysis of the ratio field includes mean, standard deviation, skewness, kurtosis and other characteristic parameters, which are used to represent the overall characteristics of the impedance distribution. The ratio space distribution pattern is shown by contour lines. High ratio area represents local rapid change and background gentle change, and low ratio area represents overall change as the main part. The impedance distribution characteristics are extracted by pattern recognition method. The ratio field is divided into 5 levels according to the numerical range: very high ratio, high ratio, medium ratio, low ratio and very low ratio. Each level corresponds to different geological significance and impedance characteristics. The very high ratio area usually corresponds to the lithology mutation interface, and the very low ratio area represents relatively uniform lithology.

[0054] According to the impedance distribution characteristics, high impedance core area and low impedance transition area are divided. The area division adopts a multi-criteria decision method, and comprehensively considers the impedance absolute value, gradient ratio, spatial connectivity and other factors. The identification criterion of the high impedance core area is that the impedance value is greater than the overall mean plus one standard deviation, and the gradient ratio is at a medium level, indicating that the impedance is high and the interior is relatively uniform. The impedance peak center extraction in the core area adopts a method combining local extreme value search and clustering analysis: first, local extreme value detection is performed in the core area to identify the local maximum value point of the impedance value as the candidate peak point; then K-means clustering analysis is performed on the candidate points, and the number of clusters is adaptively determined according to the area of the core area, and 1 cluster center is set per 1000 square meters; the center point of the clustering result is the impedance peak center, and each peak center records its accurate coordinate position, peak impedance value, influence radius and other attribute parameters. The identification criterion of the low impedance transition area is that the impedance value is lower than the overall mean, and the gradient ratio is high, indicating that the impedance is low but changes rapidly, and plays a transition role in connecting different areas. Through the peak center extraction and area division algorithm, the peak center coordinates of each high impedance core area and the distribution of the low impedance transition area are obtained.

[0055] In step S140, the reinforced drilling path is planned for the high impedance core area, and the gradual adjustment direction is developed for the low impedance transition area, and the conflict intersection point of the reinforced drilling path and the gradual adjustment direction is constructed, and the main adjustment path is established by analyzing the reconciliation force of the conflict intersection point, and the auxiliary adjustment flow is obtained based on the main adjustment path.

[0056] Based on the high impedance core zones, the reinforced drilling path is planned by using the core zone connection optimization method. The impedance peak center of each high impedance core zone is taken as the control point that must be passed through, and the main control point and the secondary control point are determined according to the impedance value. The path optimization objective function considers the path length, average impedance value and path curvature, and the genetic algorithm is used to solve the optimal path. The constraint conditions include that the path must pass through all the main core zones, the turning angle is not more than 45 degrees, and the single segment length is controlled in the range of 50-200 meters. The reinforced drilling parameters are dynamically adjusted according to the impedance characteristics of the core zones along the path, and the power adjustment coefficient calculation formula is P_factor=(Z_local / Z_max)^0.8, where Z_local is the local impedance value of the path point, and Z_max is the maximum impedance value of all core zones. The drilling speed and rotary torque are adjusted in sections in proportion to the impedance value. The technical properties such as the start and end coordinates, segment length, average impedance, power adjustment coefficient, etc. are recorded for each path segment. The path geometric accuracy is ensured by B-spline curve fitting, which ensures to meet the guiding requirements of the drilling equipment.

[0057] The gradual adjustment direction is made for the low impedance transition zone. The adjustment direction is determined based on the main axis direction of the transition zone, and the main axis is calculated by the ellipse fitting method. The ellipse fitting uses the least squares method, and the boundary points of the transition zone are used as the fitting data. The long axis direction of the ellipse is the main gradual direction. The angle accuracy of the adjustment direction is controlled within ±5 degrees to ensure the direction stability in the drilling process. The gradual strategy adopts the step adjustment mode, and the transition zone is divided into multiple adjustment segments according to the impedance gradient. The length of each segment is determined according to the impedance change rate. The area with high impedance change rate is set with shorter adjustment segment (10-20 meters), and the area with low change rate can be appropriately extended (30-50 meters). The parameter adjustment amplitude is inversely proportional to the impedance gradient, and the area with large gradient adopts small amplitude and frequent adjustment, and the area with small gradient can be adjusted with large amplitude. The adjustment parameters include drilling speed, rotary torque, axial pressure, etc., and the adjustment formula is where ΔP is the parameter adjustment amount, α is the adjustment coefficient, is the local impedance gradient. The direction control adopts the real-time feedback mechanism, and whether the direction deviates from the predetermined direction is judged by monitoring the drilling resistance. When the deviation is more than 10 degrees, the direction correction is started. The properties information such as the start coordinate, adjustment direction, adjustment parameter, segment length, etc. are recorded for each adjustment segment.

[0058] The conflict intersection points of the reinforced drilling path and the gradually adjusted direction are constructed. A parameter equation solving method is adopted, the reinforced path segment is expressed as P(t) = P1 + t(P2 - P1), and the gradual path segment is expressed as Q(s) = Q0 + s x D(s), wherein Q0 is the starting point coordinate of the gradual path, and D(s) is a direction vector function. The intersection point calculation is achieved by solving the equation set P(t) = Q(s), and the existence of the solution is judged by using a discriminant method. The intersection point classification is based on the intersection angle: an acute intersection (θ < 60°) is defined as a strong conflict point, a right intersection (60° ≤ θ ≤ 120°) is a medium conflict point, and an obtuse intersection (θ > 120°) is a weak conflict point. Each conflict intersection point records the intersection point coordinate, the intersection angle, the involved path segment identifier, the conflict intensity level and other attributes. The conflict point density analysis is achieved by the nearest neighbor distance statistics, and the areas with high density need to be adjusted and optimized. The influence domain of the intersection point is defined by a circular area with the intersection point as the center, and the influence radius is determined according to the conflict intensity: the strong conflict point radius is 30 meters, the medium conflict point radius is 20 meters, and the weak conflict point radius is 10 meters. The influence domain overlap analysis identifies the complex conflict areas that need to be processed.

[0059] In some embodiments, the reconciliation force analysis on the conflict intersection points establishes the main adjustment path, including: detecting the path conflict intensity at the conflict intersection point; performing force balance analysis based on the path conflict intensity to obtain balance parameters; generating a reconciliation force vector according to the balance parameters; and establishing the main adjustment path by using the reconciliation force vector.

[0060] The path conflict intensity is detected at the conflict intersection point. The conflict intensity detection comprehensively considers the intersection angle, the path priority, the local formation condition and other factors to establish an evaluation model. The intersection angle factor is quantified by the degree of deviation from the right angle, and the calculation formula is A_factor = |sin(2θ)|, wherein θ is the intersection angle, and the factor reaches the maximum value 1.0 at 45 degrees and 135 degrees. The path priority factor is determined based on the technical importance difference between the reinforced drilling path and the gradually adjusted direction, the power adjustment coefficient average value of the reinforced path is used as the weight base of the reinforced path, the parameter adjustment amount average value of the gradual path is used as the weight base of the gradual path, and the priority factor P_factor is calculated by the ratio of the two path weight bases. The formation condition factor is evaluated by the impedance gradient and the formation complexity at the intersection point. The greater the gradient and the more complex the formation, the higher the conflict processing difficulty. The comprehensive conflict intensity calculation formula is I = A_factor x P_factor x G_factor, wherein G_factor is the formation condition factor. The conflict intensity classification standard is: I > 0.8 is a strong conflict, 0.5 < I ≤ 0.8 is a medium conflict, and I ≤ 0.5 is a weak conflict. The strong conflict point needs to be processed by a special reconciliation strategy, the medium conflict point is processed by a regular reconciliation, and the weak conflict point can be ignored.

[0061] The force balance analysis is performed based on the path conflict intensity to obtain the balance parameters. The value of the power adjustment factor P_factor recorded at each segment of the reinforced drilling path is extracted as the basis for the driving force calculation at the corresponding intersection position. The value of the parameter adjustment amount ΔP recorded at each adjustment segment of the progressive adjustment direction is extracted as the basis for the adjustment force calculation at the corresponding position. The drilling path is regarded as a force system, and the driving force generated by the reinforced drilling path and the adjustment force generated by the progressive adjustment direction form the force action at the intersection point. The driving force is proportional to the power adjustment factor of the reinforced path at the intersection point, and the driving force calculation formula is F_drive=k1×P_factor, wherein k1 is a force conversion coefficient. The adjustment force is related to the parameter adjustment amount of the progressive path at the intersection point, and the adjustment force formula is F_adjust=k2×ΔP, wherein k2 is an adjustment force coefficient. The force balance condition requires the resultant force to be minimized, and the balance equation is F_resultant=F_drive+F_adjust=0, wherein α is the included angle between the two forces. The balance parameters include the force amplitude ratio F_resultant / F_drive, the force direction angle β, and the resultant force size F_resultant. The optimal balance state corresponds to the minimum resultant force, and the constrained optimization problem is solved by the Lagrange multiplier method. The stability of the balance parameters is evaluated by sensitivity analysis, and the influence of parameter changes on the resultant force is quantified as the sensitivity coefficient. The balance parameters include the force amplitude ratio F_resultant / F_drive, the force direction angle β, and the resultant force size F_resultant. The optimal balance state corresponds to the minimum resultant force, and the constrained optimization problem is solved by the Lagrange multiplier method. The stability of the balance parameters is evaluated by sensitivity analysis, and the influence of parameter changes on the resultant force is quantified as the sensitivity coefficient.

[0062] According to the obtained balance parameter values, the vector synthesis method is used to generate the reconciliation force vector of the conflict intersection point. The force vector is represented in polar form, and the driving force vector is F_d(r1, θ1), and the adjustment force vector is F_a(r2, θ2), wherein r is the size of the force, and θ is the direction angle of the force. The vector synthesis is realized by component synthesis, and the size and direction angle of the reconciliation force vector are obtained by geometric calculation. The normalization processing of the reconciliation force vector ensures the comparability between different intersection points, and the normalization coefficient takes the maximum force value of each intersection point. The reconciliation effect evaluation is measured by the deviation angle between the force vector and the original conflict direction, and the smaller the deviation angle, the better the reconciliation effect. Special case processing includes: when the two forces are collinear, the algebraic sum is used, when the two forces are perpendicular, the Pythagorean theorem is used, and when the two forces are opposite, the dominant force direction needs to be judged. The stability of the reconciliation force vector is verified by time series analysis to ensure the stability of the direction in the dynamic drilling process.

[0063] ​The generated harmonic force vector is used to establish the main adjustment path of the drilling operation. The path adjustment strategy is hierarchical based on the size of the harmonic force: large harmonic force (|F|>0.8F_max) corresponds to large path offset, moderate harmonic force (0.4F_max<|F|≤0.8F_max) is adjusted moderately, and small harmonic force (|F|≤0.4F_max) adopts a fine-tuning strategy. The path offset calculation uses the vector projection method, and the offset distance is proportional to the size of the harmonic force, and the offset direction is the direction of the harmonic force vector. The main adjustment path is formed by connecting the adjusted path control points, and the path smoothing uses cubic spline interpolation to ensure continuity. The path quality evaluation indicators include path length, curvature change, deviation from the original path, etc. The path coordination of multiple intersections is realized through global optimization, and the genetic algorithm is used to solve the multi-objective optimization problem, and the objective function includes the minimization of the total path length and the minimization of the conflict degree. The path segmentation identifier retains the original reinforcement segment and gradual segment attributes, and a new harmonic segment identifier is added. The geometric accuracy of the main adjustment path is controlled by the control point density and the interpolation order to ensure that the guidance accuracy requirements of the drilling equipment are met.

[0064] Based on the main adjustment path, the auxiliary adjustment flow is obtained. Using potential flow theory, the main adjustment path is regarded as a streamline, and an auxiliary flow field is established around it. The flow field is established based on the Laplace equation where φ is the velocity potential function, and the boundary condition is determined by the main adjustment path. The streamline calculation uses the orthogonal relationship between the stream function ψ and the potential function, and the streamline equation is dψ=0. The strength of the auxiliary flow is proportional to the importance of the main path, and the auxiliary flow strength coefficient of the reinforcement segment is 1.0, the gradual segment is 0.7, and the harmonic segment is 0.5. The flow field grid division uses structured grid, and the grid density is densified near the main path and sparse in the area away from the path. The direction of the auxiliary flow is determined by the tangent of the streamline, and the size of the flow velocity is calculated by the gradient of the potential function. The convergence of the flow field is verified by residual analysis to ensure the stability of the numerical calculation. The range of the auxiliary flow is determined by the streamline tracking, and the tracking termination condition is that the flow velocity drops below the threshold or reaches the calculation domain boundary. Each auxiliary streamline records the starting point coordinates, flow direction, flow velocity distribution, and influence range, etc. According to the geometric characteristics of the main adjustment path and the potential flow theory analysis, the auxiliary adjustment flow distribution is finally obtained.

[0065] In step S150, the auxiliary adjustment flow is converted into a vortex parameter sequence, the main adjustment path is converted into a drilling parameter adjustment sequence, and the parameter change characteristics are generated based on the fusion processing of the vortex parameter sequence and the drilling parameter adjustment sequence.

[0066] In some embodiments, the converting the auxiliary adjustment flow into a vortex parameter sequence comprises: identifying flow pulsation characteristics using the auxiliary adjustment flow; distinguishing a dominant evolution path from an auxiliary evolution path through the flow pulsation characteristics; establishing a vortex generation benchmark according to the dominant evolution path, performing stage division on the auxiliary evolution path using the vortex generation benchmark, and determining a vortex formation stage; and constructing a vortex parameter sequence using the vortex formation stage.

[0067] The flow pulsation characteristics are identified using the auxiliary adjustment flow. This is achieved by setting time sampling points on the drilling path, extracting the flow velocity vector at the corresponding position from the auxiliary adjustment flow vector field every fixed time interval (set to 10 seconds) to form a time series data of the flow velocity. The flow pulsation characteristics are identified based on the time series of streamlines and velocity field analysis, by calculating the curvature change of the streamlines and the time evolution of the velocity gradient to identify the pulsation mode. The pulsation characteristic parameters include time domain characteristics such as pulsation frequency, amplitude variation, phase relationship, etc. The pulsation frequency is obtained by power spectrum analysis of the flow velocity time series at each sampling point, and the main pulsation frequency corresponds to the position of the spectral density peak. The amplitude variation is obtained by extracting the envelope of the flow velocity time series, and the envelope is calculated using the Hilbert transform method. The phase relationship analysis is used to analyze the synchronicity of the flow velocity pulsation at different spatial positions, and the region with a phase difference less than π / 4 is considered to have synchronous pulsation characteristics. The pulsation intensity is quantified by the ratio of the pulsation energy to the average flow energy, and the larger the ratio, the more intense the pulsation. The spatial pulsation mode is visualized by the isopulsation line diagram to identify the spatial distribution law of the pulsation. The pulsation data is organized in the form of a space-time sequence, containing information such as time label, spatial coordinates, flow velocity vector, pulsation parameters, etc.

[0068] The dominant evolution path is distinguished from the auxiliary evolution path through the flow pulsation characteristics. The path distinction is based on the pulsation intensity and spatial continuity criteria, and the streamlines with pulsation intensity greater than a threshold value and spatial continuity are defined as the dominant evolution path. The dominant path recognition uses a connectivity analysis algorithm to spatially cluster high-pulsation-intensity regions, and the largest connected region corresponds to the dominant path. The auxiliary evolution path is other streamlines excluding the dominant path, which usually has a weak pulsation intensity or a discontinuous spatial distribution. The path importance evaluation is calculated by the flow weight, and the flow carried by the dominant path accounts for more than 60% of the total flow. The path stability analysis evaluates the time persistence of different paths, and the stable path has a presence time of more than 80% of the total observation time. The interaction between the dominant path and the auxiliary path is identified through stream line intersection analysis, and flow regime conversion may occur at the intersection point. The path classification results are stored in hierarchical vector data, and the dominant path is marked as level-1, and the auxiliary path is marked as level-2, level-3, etc. according to the importance. Each path records its streamline coordinates, pulsation parameters, importance level, stability index, etc.

[0069] The vortex generation criterion is established and the vortex formation stage is determined according to the dominant evolution path. The streamline coordinates of the dominant evolution path are extracted from the level-1 marked path, and the node coordinate sequence and connection relationship constituting the dominant path are obtained. The rhythm parameters are extracted from the time-space sequence data, including the pulsation frequency, amplitude change and rhythm intensity of the dominant path corresponding position. The vortex generation criterion is determined by the vorticity distribution on the dominant path, and the vorticity value is calculated by the flow field velocity data. The vorticity is defined as wherein is the gradient operator, v is the velocity vector, and ω is the vorticity vector. The vorticity values of each point on the dominant path are calculated by the finite difference method. In the two-dimensional case, the vorticity calculation formula is wherein vx and vy are the x and y components of the flow velocity, respectively, denotes the partial derivative operator. The threshold value of the vortex generation is taken as the threshold value of the reference vorticity value in the strongest 25% region of all calculation results. The reference criterion includes two dimensions of spatial reference and intensity reference. The spatial reference defines the effective area of vortex generation, which is determined by analyzing the high curvature section of the streamline coordinates of the dominant path. The intensity reference specifies the minimum vorticity requirement for vortex formation. The stage division of the auxiliary evolution path is based on the relative relationship between the identified auxiliary path and the dominant path. Each auxiliary path is divided into three stages according to the interaction intensity with the dominant path. The first stage is the approach stage, in which the auxiliary path gradually approaches the dominant path, and the interaction gradually strengthens. The second stage is the interaction stage, in which the two paths are closest to each other, and the vorticity exchange is most active. The third stage is the separation stage, in which the auxiliary path moves away from the dominant path, and the interaction weakens. The stage boundaries are determined by the change points of the distance and vorticity gradient between the paths. The distance extreme point and the gradient zero point are used as the stage boundary line. The vortex formation stage refers to the time period in the second stage when the vorticity exceeds the threshold value. In this stage, the auxiliary path has the condition for vortex generation. Each stage records the characteristic parameters such as start and end time, average vorticity and interaction intensity.

[0070] The vortex parameter sequence is constructed by using the vortex formation stage. The vorticity calculation formula is applied to the obtained flow velocity time sequence to obtain The two-dimensional vorticity is calculated by the finite difference method: wherein vx and vy are the x and y components of the flow velocity, , denotes partial derivative of spatial coordinates. Vortex parameter sequence is constructed with vortex formation stage as time window, and vortex feature parameters in this time period are extracted. Sequence parameters include four core elements of vortex core position trajectory, vorticity intensity variation, vortex scale evolution, and energy transfer characteristics. Vortex core trajectory is obtained by tracking the spatial movement path of vorticity extreme points in the calculated vorticity field, and the trajectory data is represented by coordinate time series. Vorticity intensity variation directly uses the calculated vorticity modulus |ω| to record the intensity evolution law of vortex from generation to dissipation, and the variation mode usually presents the characteristics of first enhancement and then weakening. Vortex scale evolution is described by the time variation of the area surrounded by the isovorticity line, and the isovorticity line is drawn based on the calculated vorticity distribution to calculate the surrounding area, reflecting the dynamic change of the influence range of vortex. Energy transfer characteristics are calculated by the time derivative of vortex kinetic energy, and the kinetic energy is calculated based on the flow velocity data. Positive values represent energy input, and negative values represent energy dissipation. The data structure of vortex parameter sequence adopts multi-dimensional array form, which supports efficient time series analysis.

[0071] The main adjustment path is converted into a drilling parameter adjustment sequence. According to the formation type and impedance characteristics of each segment of the main adjustment path, the corresponding drilling power, feed speed, rotary torque, and axial pressure parameters are configured according to the drilling process standards: low power and high rotation speed configuration for loose formation, medium power and medium rotation speed configuration for dense formation, and high power and low rotation speed configuration for hard rock formation. Drilling parameters include four core control quantities of drilling power, feed speed, rotary torque, and axial pressure, and parameter values are directly extracted from the node attributes of the main adjustment path. Sequential processing arranges parameters in the order of path travel to form adjustment instructions in time sequence. Parameter interpolation processing is used to control the transition between points to ensure the smoothness of parameter changes. The sampling interval of the adjustment sequence is set to 1 meter, corresponding to the parameter update frequency in the drilling process. Parameter standardization processing maps parameters of different dimensions to a unified numerical range [0, 1], facilitating subsequent fusion analysis. Sequence segmentation identification retains the properties of the reinforcement segment, gradual segment, and harmonic segment of the original path, and different parameter adjustment strategies are used for each segment. The data format of the drilling parameter adjustment sequence adopts the form of time-parameter pairs, with time labels corresponding to drilling mileage and parameter vectors containing set values of the four control quantities.

[0072] In some embodiments, the fusion processing based on the vortex parameter sequence and the drilling parameter adjustment sequence generates parameter variation characteristics, including: identifying the main frequency component through the vortex parameter sequence; implementing cross-modulation processing using the main frequency component and the drilling parameter adjustment sequence to form a modulation frequency spectrum; determining parameter fusion key points according to the modulation frequency spectrum; and forming parameter variation characteristics through the parameter fusion key points.

[0073] The main frequency components are identified from the vortex parameter sequence. The frequency spectrum of the vortex core position trajectory reflects the periodic characteristics of the vortex movement, and the main frequency corresponds to the main oscillation mode of the vortex. The frequency spectrum of the vortex intensity variation reveals the time scale characteristics of the vortex strengthening and decay, and the low frequency component corresponds to the slow energy accumulation process, and the high frequency component reflects the rapid vortex fluctuation. The frequency spectrum of the vortex scale evolution describes the variation law of the vortex geometric size, and the main frequency is usually related to the life cycle of the vortex. The frequency spectrum of the energy transfer characteristics characterizes the energy exchange mode between the vortex and the surrounding flow field, and the peak frequency corresponds to the most active energy transfer process. The main frequency components are identified by the power spectrum density peak value, and the first three frequencies with the largest amplitude in the power spectrum are selected as the main components. The stability of the frequency components is evaluated by the consistency of the frequency spectrum in different time windows, and the frequency with a consistency coefficient greater than 0.8 is considered as a stable main component.

[0074] The main frequency components are identified from the vortex parameter sequence. The frequency spectrum of the vortex core position trajectory reflects the periodic characteristics of the vortex movement, and the main frequency corresponds to the main oscillation mode of the vortex. The frequency spectrum of the vortex intensity variation reveals the time scale characteristics of the vortex strengthening and decay, and the low frequency component corresponds to the slow energy accumulation process, and the high frequency component reflects the rapid vortex fluctuation. The frequency spectrum of the vortex scale evolution describes the variation law of the vortex geometric size, and the main frequency is usually related to the life cycle of the vortex. The frequency spectrum of the energy transfer characteristics characterizes the energy exchange mode between the vortex and the surrounding flow field, and the peak frequency corresponds to the most active energy transfer process. The main frequency components are identified by the power spectrum density peak value, and the first three frequencies with the largest amplitude in the power spectrum are selected as the main components. The stability of the frequency components is evaluated by the consistency of the frequency spectrum in different time windows, and the frequency with a consistency coefficient greater than 0.8 is considered as a stable main component.

[0075] According to the formed modulation spectrum, the key control points of parameter fusion are determined by spectral analysis method. Through spectral power density evaluation, the frequency points with power density greater than twice the standard deviation of the average value are defined as energy concentration points. The frequency response sensitivity is calculated by the derivative of the modulation depth with respect to the frequency change, and the frequency points with large absolute value of the derivative represent the response sensitivity. The key points are screened by a double threshold criterion, and the frequency points that meet the energy and sensitivity requirements are selected as the key points. The spatial distribution of the key points is optimized by cluster analysis to avoid excessive aggregation or dispersion of the key points. Each key point records its frequency position, power density, sensitivity coefficient, influence range and other attribute parameters. The stability of the key points is verified by time-varying analysis, and the stable key points maintain similar attribute characteristics in different time windows. The number of key points is determined by the maximum information entropy criterion, which ensures sufficient information and avoids redundancy.

[0076] The parameter change characteristics are generated by fusing the vortex parameter sequence and the drilling parameter adjustment sequence. Time-frequency analysis is performed on the key points to generate the parameter change characteristics, and dynamic characteristics such as amplitude change, phase change, and frequency shift at the key frequency points are extracted. The amplitude change characteristic reflects the modulation effect of the drilling parameter on the vortex intensity, and the amplitude increase indicates that the parameter adjustment enhances the vortex activity, and the amplitude decrease indicates the inhibition effect. The phase change characteristic describes the time relationship between the drilling parameter and the vortex dynamics, and the phase advance indicates that the parameter adjustment has predictability, and the phase lag indicates that the response is delayed. The frequency shift characteristic reflects the nonlinear response of the system, and the frequency shift upward usually corresponds to the system excitation enhancement, and the frequency shift downward corresponds to the excitation weakening. The normalization processing of the characteristic parameters ensures the comparability of different types of characteristics, and the normalization range is set to [-1, 1]. The time resolution of the parameter change characteristics matches the drilling control period, ensuring the real-time updating of the characteristics. The feature vector construction combines the characteristic parameters of multiple key points into a high-dimensional vector, and the vector dimension is equal to the product of the key point number and the characteristic type number. Through the time-frequency feature extraction and analysis of the key points based on parameter fusion, the parameter change characteristics are finally generated.

[0077] In step S160, a drilling potential field distribution is constructed based on the parameter change characteristics, and concentrated stress data is collected from the drilling potential field distribution. The final parameter combination is generated using the concentrated stress data.

[0078] Specifically, the drilling potential field distribution is constructed based on the parameter change characteristics. The parameter change characteristics are mapped into a spatial potential field distribution through potential energy conversion, and dynamic characteristics such as amplitude change, phase change, and frequency shift of the feature vector are used as potential energy items to construct the potential field intensity distribution. The potential field intensity calculation adopts a weighted superposition method, and the intensity function expression is Φ(x, y) = Σ iwi ×F i (x, y), where Φ is the potential field intensity, wi is the characteristic weight, and F i is the spatial distribution function of the i-th characteristic component. The weight allocation is determined based on the variance contribution of each characteristic component, and the characteristic component with larger variance obtains higher weight. The spatial resolution of the potential field is set to 2m×2m grid, covering the entire drilling operation area. The potential field gradient calculation adopts a central difference format, representing the spatial change direction and rate of the potential field intensity. The potential field contour is drawn to visualize the potential field distribution pattern, and the contour dense area corresponds to the area with larger potential field gradient. The potential field data is stored in a grid format, and each grid cell contains potential field intensity value, gradient size, gradient direction, and other attribute information.

[0079] In some embodiments, collecting the concentrated stress data from the drilling potential field distribution includes: identifying a stress intensive area in the drilling potential field distribution; extracting the stress value of the stress intensive area to obtain the dispersed stress value; and performing spatial aggregation processing on the dispersed stress value to form the concentrated stress data.

[0080] In the constructed drilling potential field distribution grid data, based on the corresponding relationship between potential field and stress field, the potential field intensity gradient is converted into equivalent stress gradient for stress concentration area identification. According to the spatial distribution characteristics of the potential field intensity gradient, the area with gradient size exceeding the threshold value is defined as the potential stress concentration area. The threshold value is set to the average value of the potential field gradient plus two times the standard deviation, which ensures that the significant stress concentration position is identified. Spatial clustering analysis combines adjacent high gradient grids into continuous dense areas, and the region growing method is used for clustering algorithm, and the seed point is selected as the local maximum gradient point. The region boundary is determined by the equal gradient line, and the boundary gradient value is set to 70% of the threshold value to ensure the integrity of the region. The geometric characteristics of the stress concentration area include area, perimeter, shape factor, principal axis direction and other parameters. The region classification is based on the average gradient value: the average gradient of the first-class dense area is greater than 3 times the threshold value, the average gradient of the second-class dense area is between 1.5-3 times the threshold value, and the average gradient of the third-class dense area is between 1-1.5 times the threshold value. Each dense area is assigned a unique identifier, and its geometric boundary, gradient statistics, classification level and other attribute information are recorded. The area centroid coordinates are calculated by area weighted average, which is the representative position of the area.

[0081] Detailed stress field analysis is performed on the identified stress concentration area, and the dispersed stress distribution values in the area are extracted. Stress extraction is based on the calculation of potential field second-order derivative, and a local coordinate system is established in each dense area for detailed analysis. The original 2m×2m grid is subdivided into 0.5m×0.5m high-resolution grid through grid encryption processing in the area, which improves the stress calculation accuracy. The central difference method is used to calculate the stress tensor components, and the normal stress and shear stress components are obtained by the spatial second-order derivative of the potential field value. The principal stress is calculated by solving the eigenvalue of the stress tensor, and the principal stress size is . The stress direction is determined by the eigenvector, and the principal stress direction angle is calculated by the inverse tangent function. The dispersed stress value is stored in the form of tensor, and each grid point records three stress components and two principal stress values. Stress statistical analysis includes the maximum, minimum, average, standard deviation and other statistics of the stress in the area. The stress distribution pattern is visualized by stress contour and equal stress line.

[0082] The extracted dispersed stress values are processed by spatial aggregation, and the concentrated stress data representing the characteristics of the entire dense area are formed by stress integration and weight allocation method. Spatial aggregation uses area weighted integration method, and the concentrated stress value calculation formula is where σ(x, y) is the stress distribution function in the region, and the integral domain is the boundary of the dense region. Discrete calculation converts the surface integral into a grid summation form, considering the stress value and area weight of each grid. The weight distribution considers the influence of stress gradient, and the stress value in the high gradient area obtains greater weight. The convergence of the principal stress direction uses the vector average method, considering the stress size in each direction as a weight factor. The convergence results include the numerical size of the concentrated stress, the principal stress direction, the stress concentration coefficient, and other comprehensive characteristics. The stress centroid position is determined by stress moment calculation, considering the weighted average of stress size and position. The effectiveness of the concentrated stress data is verified by stress conservation, and the total stress integral before and after convergence should remain consistent. The data format uses point element storage, and each concentrated stress point contains position coordinates, stress size, direction angle, concentration coefficient, and other attributes.

[0083] In some embodiments, the generating a final parameter combination using the concentrated stress data includes: transferring the concentrated stress data to the drilling core adjustment node to form a stress concentration area; deducing a drilling force transmission path using the stress concentration area; generating a drilling execution scheme according to the coupling matching of the drilling force transmission path and the parameter change characteristics; and generating a final parameter combination based on the drilling execution scheme. The final parameter combination is generated based on the concentrated stress data.

[0084] The concentrated stress data is transferred to the drilling core adjustment node to form a stress concentration area. The coordinates of the drilling core adjustment node determined in the foregoing are used as the stress transfer target, and material parameters such as the elastic modulus and Poisson's ratio of the stratum medium are considered. The transfer path adopts the shortest distance principle, the straight-line distance from the centroid of each stress concentration body to the nearest core adjustment node is calculated, and the adjustment node with the shortest distance is taken as the stress transfer target. The stress attenuation model considers the influence of the transmission distance, and the attenuation function adopts an exponential form σ(r)=σ0×exp(-r / λ), where σ0 is the initial stress value, r is the transmission distance, and λ is the attenuation length constant. The attenuation length is determined according to the stratum type: λ=50m for loose stratum, λ=80m for dense stratum, and λ=120m for hard rock stratum. The stress superposition processing considers the case where multiple stress concentration bodies are transferred to the same adjustment node, and the vector superposition principle is used to calculate the combined stress. The stress concentration area is formed in a circular area centered on the core adjustment node and with an effective stress influence radius as the boundary. The influence radius R=-λ×ln(0.1)≈2.3λ is determined by the distance at which the stress decays to 10% of the initial value. Each stress concentration area records the center coordinates, the concentrated stress value, the influence radius, the stress direction, and other geometric and mechanical attributes.

[0085] The stress concentration area is used to deduce the drilling force transmission path. Based on the stress stream line tracking theory, the stress concentration area is regarded as the force source, and the force propagation path in the stratum is tracked. The stress stream line equation adopts Lagrange description, and the stream line differential equation is dx / σx=dy / σy=ds, where σx and σy are stress components, and ds is the stream line arc length infinitesimal. The path tracking adopts the fourth-order Runge-Kutta numerical integration method, and the integration step length is set to 1 meter, so as to ensure the accuracy of the path tracking. The change of the force transmission strength along the path is calculated through the stress attenuation model, and the strength gradually attenuates along with the transmission distance. The path branch identification process is used to deal with the convergence and dispersion phenomena of the stress stream line. The convergence point corresponds to the stress concentration position, and the dispersion point corresponds to the stress release area. The main transmission path is defined as the path with the bearing stress exceeding 50% of the average value, and the secondary path is the path with the bearing stress lower than the average value. The path connectivity analysis ensures that all stress concentration areas have effective force transmission channels. The transmission path data is stored in the polyline element, each path record has the attributes such as the starting and ending point coordinates, path length, transmission strength, path level and the like. The path network topological structure is described through the graph theory method, the node represents the stress concentration area, and the edge represents the transmission path.

[0086] According to the coupling matching of drilling force transmission path and parameter variation characteristics, a drilling execution scheme is generated. First, the spatial position information is identified according to the drilling force transmission path. The node coordinates, path intersection points, force convergence points and other key positions are extracted from the transmission path data. The node coordinates are directly obtained from the start and end points of the path elements. The path intersection points are identified by a geometric intersection detection algorithm to determine the spatial intersection positions between different transmission paths. The force convergence points are determined by calculating the force line density at each spatial position. The importance of the positions is evaluated based on the connectivity and force transmission strength. The influence range of the spatial positions is determined by Voronoi diagram division. Then, the K-means improved algorithm is used to determine the core node positions based on the spatial position information. The number of clusters K is determined to be 3-8 by the elbow rule. The weighted Euclidean distance is used for distance measurement. The weight is determined by the position importance score. The K most important positions are selected as the cluster centers. In the iteration process, the cluster centers move towards the importance weighted centroid xc=Σi(wi×xi) / Σiwi. The convergence criterion is that the center position changes less than 1 meter and the condition is met for 3 consecutive iterations. Then, the matching correlation matrix is established according to the core node positions and parameter variation characteristics. The parameter variation characteristics are mapped to the operation area spatial grid through spatial interpolation. An m×n dimensional correlation matrix (m is the number of core nodes, n is the number of feature components) is constructed. The matrix element M(i,j)=α×R_spatial(i,j)+β×R_feature(i,j). The spatial correlation R_spatial=exp(-d² / 2σ²). The feature similarity R_feature is calculated by the correlation coefficient of the feature vector. The correlation strength threshold is set to 0.5. Finally, the drilling execution scheme is formed by the matching correlation matrix. The feature response mode of each core node is identified by analyzing the matrix row vector. The spatial influence mode of the feature component is identified by the column vector. Based on the high correlation strength (>0.5) of the node-feature pair, four strategy types of power regulation, speed control, torque regulation and pressure control are developed. The strategy parameters are determined by quantitative mapping of the correlation strength. The execution timing is arranged based on the spatial order of the core nodes and the time sequence of the features. Finally, a complete drilling execution scheme including control strategy type, adjustment parameter and execution timing is formed.

[0087] The final parameter combination is generated based on the drilling execution plan. The control strategy in the execution plan is converted into specific device operation parameters, and the core parameters include drilling power, feed speed, rotary torque, and axial pressure, which are the four main control quantities. The parameter values are extracted from the strategy module of the execution plan and are numerically processed. The parameter boundaries are determined based on the technical specifications of the device: the drilling power range is 50-500 kW, the feed speed range is 0.1-5.0 m / min, the rotary torque range is 1000-8000 N·m, and the axial pressure range is 10-200 kN. The formation identification function is realized through the drilling process driven by the parameter combination. Real-time signals such as drilling resistance feedback, vibration response, and feed resistance are used to identify the formation type and characteristics. The identification algorithm uses a pattern matching method to compare the measured signals with the formation response pattern library, and the highest matching degree corresponds to the current formation type. The adaptive parameter adjustment mechanism is based on the feedback control strategy of the execution plan and the formation identification result. When the identified formation type does not match the expected one, the corresponding parameter adjustment strategy is automatically called. The adjustment strategy includes gradual adjustment and step adjustment: gradual adjustment is suitable for slow changes in the formation, and the adjustment amplitude is controlled within ±10% of the current value; step adjustment is suitable for sudden changes in the formation, and the adjustment amplitude can reach ±30%. The real-time performance of parameter adjustment is guaranteed by the fast response mechanism, and the time from signal detection to parameter adjustment completion is controlled within 2 seconds. Finally, the drilling formation identification and adaptive parameter adjustment are completed.

[0088] To perform the drilling formation identification and adaptive parameter adjustment method corresponding to the above-mentioned method embodiment, to realize the corresponding functions and technical effects. Referring to Figure 2 , Figure 2 A structure block diagram of a drilling formation identification and adaptive parameter adjustment system 200 provided by an embodiment of the present application is shown. For ease of illustration, only the parts related to the present embodiment are shown. The drilling formation identification and adaptive parameter adjustment system 200 provided by the present embodiment includes:

[0089] The data acquisition module 201 is configured to acquire multi-source sensing data in the drilling process, wherein the multi-source sensing data includes vibration signals and resistance data, and the multi-source sensing data is subjected to composite processing to construct a formation response pattern library.

[0090] The regional analysis module 202 is configured to analyze the parameter response region distribution based on the formation response pattern library to generate a regional distribution atlas, extract a parameter response blank region from the regional distribution atlas, reconstruct the missing information of the parameter response blank region to obtain an implicit formation feature, and establish a set of adjustment node candidates based on the regional distribution atlas and the implicit formation feature.

[0091] The core identification module 203 is configured to comprehensively evaluate the candidate adjustment points to obtain a parameter action range and an adjustment potential value of each candidate adjustment point, determine a drilling core adjustment point according to the parameter action range and the adjustment potential value, and extract a high-impedance core area and a low-impedance transition area from the drilling core adjustment point.

[0092] The path planning module 204 is configured to plan a reinforced drilling path for the high-impedance core area, formulate a gradual adjustment direction for the low-impedance transition area, construct a conflict intersection point of the reinforced drilling path and the gradual adjustment direction, perform reconciliation force analysis on the conflict intersection point to determine a main adjustment path, and obtain an auxiliary adjustment flow based on the main adjustment path.

[0093] The parameter processing module 205 is configured to convert the auxiliary adjustment flow into an eddy current parameter sequence, convert the main adjustment path into a drilling parameter adjustment sequence, and perform fusion processing based on the eddy current parameter sequence and the drilling parameter adjustment sequence to generate a parameter change feature.

[0094] The parameter generation module 206 is configured to construct a drilling potential field distribution based on the parameter change feature, collect dispersed stress from the drilling potential field distribution to generate concentrated stress data, and generate a final parameter combination using the concentrated stress data.

[0095] The drilling formation identification and adaptive parameter adjustment system 200 described above can implement the drilling formation identification and adaptive parameter adjustment method of the method embodiment described above. The optional items in the method embodiment described above are also applicable to this embodiment, and will not be described in detail here. The remaining content of the embodiment of the present application can refer to the content of the method embodiment described above, and will not be described in detail in this embodiment.

[0096] The purpose of the above embodiments is to exemplarily reproduce and deduce the technical solutions of the present application, and to completely describe the technical solutions, purposes and effects of the present application. The purpose is to make the public more thoroughly and comprehensively understand the disclosure of the present application, and does not limit the protection scope of the present application.

[0097] The above embodiments are not based on an exhaustive enumeration of the present application, and there can be many other unlisted embodiments. Any substitution and improvement made without violating the concept of the present application is within the protection scope of the present application.

Claims

1. A drilling formation identification and adaptive parameter adjustment method, characterized in that: include: Collecting multi-source sensor data during the drilling process, the multi-source sensor data including vibration signals and resistance data, and performing composite processing on the multi-source sensor data to construct a formation response pattern library; performing parameter response regional distribution analysis based on the formation response pattern library to generate a regional distribution map, extracting parameter response blank areas from the regional distribution map, reconstructing missing information in the parameter response blank areas to obtain implicit formation characteristics, and establishing a candidate set of adjustment points based on the regional distribution map and the implicit formation characteristics; Comprehensively evaluating the candidate set of adjustment points to obtain a parameter range and an adjustment potential value of each candidate adjustment point, determining a core drilling adjustment point based on the parameter range and the adjustment potential value, and extracting a high-impedance core area and a low-impedance transition area from the core drilling adjustment point; Planning an enhanced drilling path for the high-impedance core area, formulating a progressive adjustment direction for the low-impedance transition area, constructing a conflicting intersection between the enhanced drilling path and the progressive adjustment direction, performing a harmonic force analysis on the conflicting intersection to determine a primary adjustment path, and obtaining an auxiliary adjustment flow based on the primary adjustment path; Converting the auxiliary regulating flow into an eddy current parameter sequence, converting the main regulating path into a drilling parameter regulating sequence, and generating a parameter variation feature based on a fusion process of the eddy current parameter sequence and the drilling parameter regulating sequence; A drilling potential field distribution is constructed based on the parameter variation characteristics, dispersed stress is collected from the drilling potential field distribution to generate concentrated stress data, and the concentrated stress data is used to generate a final parameter combination.

2. The method according to claim 1, characterized in that The composite processing of the multi-source sensing data to construct a formation response pattern library includes: generating a formation response spectrum by applying frequency excitation to the vibration signal; performing frequency band decomposition on the formation response spectrum to identify formation interface distribution; constructing a formation response envelope based on the formation interface distribution and the resistance data; A formation response pattern library is established based on the formation response envelope.

3. The method according to claim 1, characterized in that The step of reconstructing missing information in the parameter response blank area to obtain implicit formation characteristics includes: Obtaining boundary features of the parameter response blank area; Inferring the internal response law of the parameter response blank area based on the boundary characteristics; Performing data interpolation processing according to the response rule to generate compensation response data; Implicit formation characteristics are reconstructed based on the compensated response data.

4. The method according to claim 1, wherein The extracting of the high-impedance core area and the low-impedance transition area from the drilling core adjustment point comprises: Performing impedance gradient calculation on the drilling core adjustment point to obtain impedance change rate distribution; separating a high gradient component and a low gradient component from the impedance change rate distribution; Performing ratio analysis on the high gradient component and the low gradient component to obtain impedance distribution characteristics; A high-impedance core region and a low-impedance transition region are divided according to the impedance distribution characteristics.

5. The method according to claim 1, wherein The said analysis of the reconciliation force of the conflict intersection to establish the main adjustment path includes: detecting path conflict intensity at the conflict intersection; Performing force balance analysis based on the path conflict strength to obtain balance parameters; generating a harmonic force vector according to the balance parameter; The main adjustment path is established using the harmonic force vector.

6. The method according to claim 1, characterized in that The step of converting the auxiliary regulating flow into a vortex parameter sequence comprises: Using the auxiliary regulating flow to identify flow rhythm characteristics; Distinguishing the main evolution path from the auxiliary evolution path by the flow rhythm characteristics; Establishing a vortex generation benchmark based on the main evolution path, dividing the auxiliary evolution path into stages using the vortex generation benchmark, and determining the vortex formation stage; The vortex formation stage is used to construct a vortex parameter sequence.

7. The method according to claim 1, characterized in that The generating of parameter change characteristics by performing fusion processing based on the eddy current parameter sequence and the drilling parameter adjustment sequence includes: identifying main frequency components through the eddy current parameter sequence; cross-modulation processing is performed using the main frequency component and the drilling parameter adjustment sequence to form a modulation spectrum; Determining a parameter fusion key point according to the modulation spectrum; The parameter variation features are formed by fusing the key points with the parameters.

8. The method according to claim 1, characterized in that The step of collecting dispersed stress from the drilling potential field distribution to generate concentrated stress data includes: identifying stress-intensive regions in the drilling potential field distribution; Performing stress extraction on the stress-intensive area to obtain a dispersed stress value; The dispersed stress values ​​are spatially aggregated to form concentrated stress data.

9. The method according to claim 1, characterized in that The generating a final parameter combination by using the concentrated stress data includes: transferring the concentrated stress data to the drilling core adjustment point to form a stress concentration area; Deducing a drilling force transmission path using the stress concentration area; Generate a drilling execution plan based on coupling matching between the drilling force transmission path and the parameter change characteristics; A final parameter combination is generated based on the drilling execution plan.

10. A drilling formation identification and adaptive parameter adjustment system, characterized in that: include: A data acquisition module is used to collect multi-source sensor data during the drilling process, wherein the multi-source sensor data includes vibration signals and resistance data, and compositely process the multi-source sensor data to construct a formation response pattern library; a regional analysis module configured to analyze the regional distribution of parameter responses based on the formation response pattern library to generate a regional distribution map, extract parameter response blank areas from the regional distribution map, reconstruct missing information in the parameter response blank areas to obtain implicit formation characteristics, and establish a candidate set of adjustment points based on the regional distribution map and the implicit formation characteristics; a core identification module configured to comprehensively evaluate the candidate set of adjustment points to obtain a parameter range and an adjustment potential value of each candidate adjustment point, determine a drilling core adjustment point based on the parameter range and the adjustment potential value, and extract a high-impedance core area and a low-impedance transition area from the drilling core adjustment point; a path planning module for planning an enhanced drilling path for the high-impedance core area, formulating a progressive adjustment direction for the low-impedance transition area, constructing a conflicting intersection between the enhanced drilling path and the progressive adjustment direction, performing a harmonic force analysis on the conflicting intersection to determine a primary adjustment path, and obtaining an auxiliary adjustment flow based on the primary adjustment path; a parameter processing module, configured to convert the auxiliary regulating flow into an eddy current parameter sequence, convert the main regulating path into a drilling parameter regulating sequence, and generate a parameter variation feature by performing a fusion process based on the eddy current parameter sequence and the drilling parameter regulating sequence; A parameter generation module is used to construct a drilling potential field distribution based on the parameter change characteristics, collect dispersed stress from the drilling potential field distribution to generate concentrated stress data, and use the concentrated stress data to generate a final parameter combination.

Images