Drilling parameter intelligent optimization control method and system based on multi-field coupling

Abstract

The invention discloses a drilling parameter intelligent optimization control method and system based on multi-field coupling, and relates to the technical field of intelligent control, and the method comprises the steps: collecting the data of a thermal field, a force field, a sound field and a well wall stress field in a drilling process, building a linkage prediction model, recognizing the wear state of a drill bit and the type of a stratum working condition, and judging the co-evolution trend of the two, determining a control dominant factor, switching a corresponding dominant control mode, activating a target strategy optimization unit to generate an optimal drilling parameter combination to guide mining control, and adjusting the control mode and optimizing the target weight according to closed-loop feedback. The method solves the technical problems that the prior art depends on single physical field monitoring and static parameter control and is difficult to deal with rapid changes under complex geological conditions, parameter control response is lagged and adaptability is poor, and achieves dominant control mode recognition and optimization strategy self-adaptive switching based on multi-field coupling trend analysis. And the response speed and adaptability to complex working conditions are improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent control technology, specifically to a method and system for intelligent optimization control of drilling parameters based on multi-field coupling. Background Technology

[0002] In the drilling process for resources such as oil and natural gas, the rational control of drilling parameters is a crucial link in ensuring drilling efficiency, downhole equipment lifespan, and operational safety. Current technologies largely rely on empirical adjustments of single or a few physical quantities, such as drilling pressure, rotational speed, and mechanical drilling speed, lacking a comprehensive perception and analysis capability of the multi-physics coupling characteristics during drilling. Especially in complex geological environments, such as heterogeneous formations, fault zones, and high-temperature, high-pressure areas, drill bit wear and formation disturbances are often intertwined. Traditional parameter control methods struggle to identify the dominant risk sources in a timely manner, easily leading to control lag, parameter misadjustment, premature drill bit failure, or downhole accidents. Furthermore, existing control systems are mostly static or rule-driven, lacking joint modeling and dynamic response mechanisms for drill bit status and formation trend changes, making it difficult to achieve refined, adaptive parameter optimization control. Summary of the Invention

[0003] This application provides a method and system for intelligent optimization control of drilling parameters based on multi-field coupling, which is used to solve the technical problems of existing methods that rely on single physical field monitoring and static parameter control, making it difficult to cope with rapid changes under complex geological conditions, and resulting in lag in parameter control response and poor adaptability.

[0004] The first aspect of this application provides an intelligent optimization control method for drilling parameters based on multi-field coupling. The method includes: acquiring real-time multi-physics field data during the drilling process, including thermal field data, force field data, acoustic field data, and wellbore stress field data; constructing a linkage prediction model based on the real-time multi-physics field data, simultaneously identifying drill bit wear state and formation condition type, and determining their co-evolution trend to obtain wear trend prediction results and geological trend prediction results; determining the current dominant control factor based on the wear trend prediction results and geological trend prediction results; switching to the corresponding dominant control mode based on the dominant control factor, and activating the target strategy optimization unit to perform drilling parameter optimization according to the dominant control mode, outputting the optimal drilling parameter combination; guiding the automatic drilling control module to perform production control according to the optimal drilling parameter combination, and dynamically adjusting the dominant control mode and optimization target weights based on real-time closed-loop feedback.

[0005] The second aspect of this application provides an intelligent optimization control system for drilling parameters based on multi-field coupling. The system includes: a field data acquisition module for acquiring real-time multi-physics field data during the drilling process, including thermal field data, force field data, acoustic field data, and wellbore stress field data; a co-evolution trend prediction module for constructing a linked prediction model based on the real-time multi-physics field data, simultaneously identifying drill bit wear state and formation condition type, determining their co-evolution trend, and obtaining wear trend prediction results and geological trend prediction results; and a main control factor determination module. The system comprises the following modules: a master control factor determination module, used to determine the current dominant control factor based on the wear trend prediction results and geological trend prediction results; a drilling parameter optimization module, used to switch to the corresponding dominant control mode based on the dominant control factor, and to activate the target strategy optimization unit to perform drilling parameter optimization according to the dominant control mode, outputting the optimal drilling parameter combination; and a production control and feedback module, used to guide the automatic drilling control module to perform production control according to the optimal drilling parameter combination, and to dynamically adjust the dominant control mode and optimization target weights according to real-time closed-loop feedback.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages: The intelligent optimization control method and system for drilling parameters based on multi-field coupling provided in this application pertains to the field of intelligent control technology. By constructing a linkage prediction and control mechanism based on multi-physics field data, integrating drill bit wear status and formation condition trend identification, dynamically determining the dominant control factor, switching control modes as needed, and optimizing drilling parameters, it achieves adaptive decision-making and closed-loop adjustment of parameter control. This solves the technical problems of existing methods that rely on single-physics field monitoring and static parameter control, which are difficult to cope with rapid changes under complex geological conditions, and suffer from lag in parameter control response and poor adaptability. It realizes the identification of dominant control modes and adaptive switching of optimization strategies based on multi-field coupling trend analysis, thereby improving the response speed and adaptability to complex conditions. Attached Figure Description

[0007] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0008] Figure 1 A schematic diagram of the intelligent optimization control method for drilling parameters based on multi-field coupling provided in the embodiments of this application; Figure 2A schematic diagram of the structure of the intelligent optimization control system for drilling parameters based on multi-field coupling provided in the embodiments of this application.

[0009] Figure labeling: Field data acquisition module 11, Co-evolution trend prediction module 12, Main control factor determination module 13, Drilling parameter optimization module 14, Production control and feedback module 15. Detailed Implementation

[0010] This application provides a method and system for intelligent optimization control of drilling parameters based on multi-field coupling, which is used to solve the technical problems of existing methods that rely on single physical field monitoring and static parameter control, making it difficult to cope with rapid changes under complex geological conditions, and resulting in lag in parameter control response and poor adaptability.

[0011] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0012] It should be noted that the terms "first," "second," etc., in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules not explicitly listed or inherent to such processes, methods, products, or devices.

[0013] Example 1, as Figure 1 As shown, this application provides an intelligent optimization control method for drilling parameters based on multi-field coupling, the method comprising: P10: Collect real-time multi-physics field data during the drilling process, including thermal field data, force field data, acoustic field data, and wellbore stress field data.

[0014] Specifically, the first step is to acquire real-time multi-physics data during the drilling process to establish a multi-dimensional sensing system covering the drill bit status and formation response. Specifically, the multi-physics data includes, but is not limited to, thermal field data, force field data, acoustic field data, and wellbore stress field data. Each type of data is sensed and transmitted in real time through highly sensitive sensor devices deployed at key locations downhole.

[0015] First, the acquisition of thermal field data mainly relies on high-temperature resistant temperature sensors (such as platinum resistance thermometers or fiber optic distributed temperature sensors) placed near the drill bit. The acquired data reflects the frictional heat generated by the contact between the drill bit cutting zone and the formation, as well as the cooling efficiency after the drilling fluid removes heat. Characteristic parameters such as the slope and peak value of the temperature change curve are of great significance for predicting the thermal fatigue failure of the drill bit.

[0016] Secondly, the force field data includes indicators such as weight on drill bit (WOB), torque, axial vibration, and lateral impact load. Acquisition devices include downhole force rings, torque sensors, and multi-directional accelerometers. These sensors are typically mounted on the upper stabilizer of the drill string or on the screw drill string near the drill bit, generating high-resolution force datasets. Parameters such as the frequency components and mean square fluctuation range of the force field changes are crucial for assessing formation hardness variations, drill bit jamming, and structural wear.

[0017] Furthermore, the acoustic field data acquisition module utilizes high-frequency piezoelectric acoustic emission sensors or micro-vibration pickups deployed in the drill string to receive weak high-frequency acoustic wave signals (Acoustic Emission, AE) generated during the drill bit's rock-breaking process in real time. By extracting features such as the frequency domain energy distribution, time domain envelope, and duration of the AE signal, highly sensitive identification of early failures such as structural cracks and gear detachment in the drill bit can be achieved.

[0018] Furthermore, wellbore stress field data are collected through drilling pressure measurement (LWD) devices or wellbore imaging sensors (such as resistivity imagers). The measurement range covers parameters such as wellbore pore pressure, radial stress, and rock interface anomalies, which can be used to identify wellbore stability boundary conditions and heterogeneous formation sections.

[0019] To achieve efficient and collaborative transmission and processing of the aforementioned multi-field data, this embodiment preferably uses a drilling information acquisition and wireless transmission system with a data synchronization and marking mechanism (such as a mud pulse or electromagnetic wave drilling measurement system), combined with a downhole local data cache and a wellhead receiving terminal, to construct a low-latency data channel. At the wellhead, a multi-physics field data fusion processing module is set up, including a timestamp correction algorithm for data parsing and synchronization of temperature / pressure / sound waves, an outlier removal mechanism, and a feature vector generation module, to ensure that various physical field data at each moment can form a stable state representation vector.

[0020] The above methods not only provide continuous input for the coordinated prediction of subsequent drill bit wear and formation change trends, but also lay a foundation for data support for control strategy switching and parameter adaptive optimization.

[0021] P20: Based on the real-time data of the multi-physics field, a linkage prediction model is constructed to simultaneously identify the drill bit wear state and formation condition type, and to determine the co-evolution trend of the two, thereby obtaining wear trend prediction results and geological trend prediction results.

[0022] Furthermore, based on the real-time multiphysics data, a linked prediction model is constructed to simultaneously identify the drill bit wear state and formation condition type. In this embodiment, step P20 further includes: P21: Segment the temperature distribution in the thermal field data, extract the temperature change rate of each region, and calculate the dynamic wear index by combining it with the fluctuation period of axial pressure in the force field data; P22: Divide the acoustic field data into multiple segments according to the time window, extract the frequency component features of each segment, and spatially match the frequency component features with the stress concentration area distribution in the wellbore stress field data to generate stress distribution features; P23: Compare the dynamic wear index with the preset wear stage threshold through a multi-layer feature fusion network to output the drill bit wear level; P24: Input the frequency component features and the stress distribution features into the formation feature matching library to output the formation classification result with confidence.

[0023] It should be understood that, based on real-time data from multiple physics fields, a linkage prediction model is constructed that integrates information from thermal field, force field, acoustic field, and wellbore stress field. This model is used to simultaneously identify the drill bit wear state and the type of formation conditions encountered during drilling, and further determine the co-evolution trend between the two over time. This results in wear trend prediction and geological trend prediction, providing a basis for the dynamic switching of subsequent control modes.

[0024] Specifically, the temperature distribution area in the thermal field data is first divided into multiple sensing sub-regions, and the temperature change rate of each region over time is extracted by performing derivative calculations. Then, combining the axial pressure (drill pressure) fluctuation signal over time in the force field data, autocorrelation analysis is used to calculate the energy concentration range within its main fluctuation period or frequency domain. Based on this, a dynamic wear index reflecting the local fatigue state of the drill bit under thermo-mechanical coupling is constructed. This index can quantify the wear intensity change trend of the drill bit under different operating conditions and is used for subsequent quantitative identification of wear levels.

[0025] Furthermore, the acoustic field data is divided into multiple segments according to time windows. The size of the time window can be determined based on the data sampling frequency and the characteristics of the acoustic signal during the actual drilling process. For each segment, its frequency component features are extracted. For example, the frequency distribution of the acoustic signal is analyzed using Fast Fourier Transform (FFT) to obtain the main frequency components and their corresponding amplitudes. These frequency component features can reflect the characteristics of the acoustic signal generated when the drill bit interacts with the rock, thus providing a basis for identifying the wear state of the drill bit. Simultaneously, these frequency component features are spatially matched with the stress concentration area distribution in the wellbore stress field data. The stress concentration area in the wellbore stress field data refers to the region with high stress values ​​in the wellbore rock. These regions are often where rock properties change or wellbore stability is poor. Through spatial matching, the correspondence between the frequency component features of the acoustic signal and the stress concentration area can be determined, thereby generating stress distribution features. This process can be achieved by establishing a spatial matching algorithm, such as coordinate system-based matching or image processing-based matching methods.

[0026] Next, a multi-layer feature fusion network is used to compare the dynamic wear index with a preset wear stage threshold to output the drill bit wear level. The multi-layer feature fusion network is a deep learning-based model that can automatically learn the feature representation of input data and fuse features from different levels. In this application, the dynamic wear index is used as one of the input features. By comparing it with a preset wear stage threshold, the network's classification ability is used to output the drill bit wear level. The preset wear stage threshold can be determined based on the relationship between the drill bit's service life, wear degree, and drilling efficiency; for example, the wear level can be divided into three stages: slight wear, moderate wear, and severe wear. The multi-layer feature fusion network can be trained using a large amount of historical data and labeled wear level samples to improve the model's accuracy and generalization ability.

[0027] Finally, the frequency component features and stress distribution features are input into the formation feature matching library, and the resulting formation classification results with confidence scores are output. The formation feature matching library is a database containing various formation types and their corresponding features. By matching the extracted frequency component features and stress distribution features with features in the library, the type of the current formation can be determined. During the matching process, similarity measurement methods, such as Euclidean distance and cosine similarity, can be used to calculate the similarity between the input features and features in the library. Based on the similarity score, a formation classification result with confidence scores is output. The confidence score reflects the reliability of the matching result and can be determined by setting a similarity threshold. For example, when the similarity score is higher than a certain threshold, the matching result is considered to have high confidence; conversely, it is considered to have low confidence. Through this process, real-time identification of formation condition types can be achieved, providing accurate formation information for subsequent drilling parameter optimization.

[0028] Through the aforementioned linkage prediction process based on the fusion of thermal, mechanical, acoustic, and stress data, the physical degradation state of the drill bit and the changing trends of the formation structure can be dynamically tracked during continuous drilling, and the relationship between the two can be collaboratively modeled. This mechanism enables the control system to accurately distinguish, in actual construction, whether the increased wear trend of the drill bit dominates the drilling risk, or whether abrupt changes in the formation structure cause abnormal drilling pressure and other changes in working conditions, thereby providing precise and feedforward state perception capabilities for the adaptive adjustment of subsequent parameter control modes.

[0029] Furthermore, to determine the co-evolution trend of the two and obtain the wear trend prediction results and geological trend prediction results, step P20 of this application embodiment also includes: P25: Construct a dual-channel trend analysis model, which includes a first channel and a second channel. The first channel generates a wear trend index by accumulating the product of the temperature change rate and the pressure fluctuation amplitude, and superimposing the proportion of high-frequency energy in the acoustic signal. The second channel generates a geological trend index by calculating the dynamic ratio of vertical stress to horizontal stress and combining it with the slope of formation pressure change. P26: Based on the dual-channel trend analysis model, a sliding time window is used to predict the wear trend growth rate and geological trend change rate within a set future time period.

[0030] Optionally, to effectively determine the co-evolution trend between drill bit wear state and formation condition type, a dual-channel trend analysis model is constructed based on the aforementioned wear and formation identification. This model is used to jointly model and predict the dynamic changes of both. The dual-channel trend analysis model includes a first channel and a second channel that are independent of each other but have collaborative analysis capabilities, respectively performing quantitative modeling and outputting results for wear trends and geological trends.

[0031] Specifically, the first channel is used to construct a wear trend index. This channel takes thermal, mechanical, and acoustic data as its main inputs. It obtains the historical cumulative amount of thermal-mechanical coupled load by performing time-series accumulation processing on the product of the temperature change rate of the drill bit surface area and the axial pressure fluctuation amplitude. Based on this, the proportion of high-frequency components in the acoustic field data is further introduced. This proportion typically reflects the frequency characteristics of high-intensity impacts or particle shedding between the drill bit and the formation, and is an early signal of rapid wear or structural damage to the drill bit. The information from these three dimensions is weighted and fused to generate a continuously updated wear trend index to characterize the performance degradation rate of the drill bit under current operating conditions.

[0032] Meanwhile, the second channel is used to construct geological trend indicators. This channel mainly utilizes wellbore stress field and formation pressure correlation data to model and analyze the trend of geological structure changes along the drilling path. The system calculates the ratio between vertical stress and horizontal stress in real time and monitors its dynamic changes over time. This ratio reflects the trend of formation stress state transitioning from stable to heterogeneous. Simultaneously, combined with the time slope information of formation pore pressure, it determines whether there is sudden formation pressurization, fluid anomalies, or lithological transitions. Through the joint modeling of the above force-pressure composite parameters, the system generates geological trend indicators reflecting the risk of geological abrupt changes and the degree of instability.

[0033] After constructing the dual-channel trend indicators, time series forecasting is performed on both indicators based on a sliding time window mechanism of a set length. The forecasting algorithm employs a dynamic modeling structure based on a recurrent neural network (such as LSTM) to learn and fit patterns from historical time series trends, and infers the wear trend growth rate and geological trend change rate within a drilling distance of T seconds or T meters from the current time point. The forecast results are output as two continuous variables, corresponding to the growth rate of drill bit wear trend and the rate of change of geological environment stability, respectively, with the option to add a prediction confidence range. These forecast results serve as key quantitative evidence for determining the current dominant control factor, directly driving the selection of subsequent control modes and the invocation of optimization strategies.

[0034] By introducing a dual-channel trend analysis model, we can not only achieve independent trend modeling of the drill bit and formation evolution, but also realize the state feedforward identification capability in the drilling process through trend prediction under a unified time axis, providing strong support for improving the response foresight and proactive parameter adjustment of the overall control system.

[0035] P30: Based on the wear trend prediction results and geological trend prediction results, determine the current dominant controlling factor.

[0036] Specifically, based on the wear trend prediction results and geological trend prediction results, the determination of the dominant control factors is further performed, thereby providing a decision-making basis for the selection of subsequent control strategies and the switching of parameter optimization modes. The dominant control factors refer to the factors that have a dominant impact on the safety, efficiency, and drill bit life of drilling parameters during the current drilling period, and are mainly dynamically judged between "drill bit wear risk" and "formation condition change risk".

[0037] Specifically, the system first receives and analyzes two trend indicators output by the dual-channel trend analysis model: the wear trend growth rate and the geological trend change rate. The wear trend growth rate reflects the rate at which the degree of drill bit wear changes over time or drilling depth; a higher value indicates that the drill bit is degrading at a faster rate. The geological trend change rate, on the other hand, characterizes the degree of increased heterogeneity, severe stress disturbance, or abrupt lithological changes in the formation environment. By quantitatively comparing the two indicators and combining their respective confidence ranges with historical reference benchmarks, the system determines which trend indicator is more sensitive to the stability and parameter misalignment of the drilling process under current operating conditions.

[0038] To improve the robustness of the judgment, multi-dimensional auxiliary quantities can be introduced as decision support information, including the current ROP (rate of penetration) fluctuation, abnormal torque amplitude, drill bit temperature rise rate, and drill string vibration intensity. If the system detects that the wear trend growth rate is significantly higher than the historical average, accompanied by an increase in torque and a faster rise in drill bit temperature, the current dominant factor is determined to be "wear-dominated"; conversely, if the geological trend change rate is abnormally high, accompanied by signs such as sudden changes in pore pressure, well diameter enlargement, or a surge in vibration amplitude, the dominant factor is determined to be "geologically dominated".

[0039] This determination process can employ a classification and discrimination model based on a weighted voting mechanism, or a trained supervised neural network classifier. Its output consists of the current control-dominant factor category label and its confidence probability value. Ultimately, this determination serves as the input trigger condition for control mode selection in subsequent steps, deciding whether to enter a wear control mode primarily focused on drill bit protection, or a formation control mode primarily focused on mitigating geological disturbances. This enables dynamic adaptive adjustment of the control strategy, enhancing the intelligent response capability and system stability throughout the drilling process.

[0040] P40: Based on the aforementioned control dominant factor, switch to the corresponding dominant control mode, and according to the dominant control mode, activate the target strategy optimization unit to perform drilling parameter optimization and output the optimal drilling parameter combination. The dominant control modes include wear-dominated control mode, geological-dominated control mode, and composite control mode.

[0041] Furthermore, based on the aforementioned control dominance factor, switching to the corresponding dominant control mode, step P40 in this embodiment of the application further includes: P41: When the quantitative difference between the wear trend and the geological trend is less than the coordination threshold and the maximum trend growth rate reaches the critical value, the composite control mode is activated; P42: When the wear trend is higher than the geological trend and its growth rate exceeds the set proportional critical value, the wear-dominated control mode is activated; P43: When the geological trend is higher than the wear trend and its change rate exceeds the set proportional critical value, the geological-dominated control mode is activated.

[0042] It should be understood that, based on the results of the dominant control factor determined by the judgment, the system switches to the corresponding dominant control mode. Under this control mode, the target strategy optimization unit is invoked to perform targeted optimization of key parameters during the drilling process, thereby outputting the optimal combination of drilling parameters that meets the requirements of the current working conditions. The dominant control mode, as the upper-level structure of the control logic, is used to dynamically guide the optimization algorithm to run under different target orientations. Its classification includes wear-dominated control mode, geological-dominated control mode, and a composite control mode applicable when the trends of both are similar or jointly dominant.

[0043] Specifically, the switching of control modes is based on a quantitative comparison between the wear trend prediction results and the geological trend prediction results, as well as the relative positional relationship between their growth rate (or rate of change) and a preset threshold. The system pre-sets a set of discrimination thresholds for mode determination, including a synergy threshold (used to determine the degree of trend similarity) and a proportional threshold (used to determine whether a certain trend is significantly dominant).

[0044] In the control logic, when the difference between the wear trend and the geological trend (i.e., the absolute difference between their quantified values) is less than a set coordination threshold, it indicates that the two trends are similar in degree, and if the growth rate of either trend exceeds the maximum trend threshold (e.g., much higher than the historical average or physical warning standard), the system will activate the composite control mode. In this mode, the optimization algorithm incorporates both trends into the objective function construction logic simultaneously and assigns dynamic adjustment capabilities to the wear impact term and the geological disturbance term, thereby achieving parallel control of the two types of risks.

[0045] If the wear trend indicator is higher than the geological trend, and its growth rate relative to the geological trend exceeds a set critical value (e.g., more than 30%), the system will switch to a wear-dominated control mode. In this mode, the target optimization algorithm will focus on extending drill bit life and suppressing the wear rate, while allowing for the short-term sacrifice of some mechanical drilling rate or wellbore stability parameters to maximize the tool end's continuous operating capability. At this time, the strategy optimization unit may take measures such as reducing drilling pressure, slowing down drilling speed, and optimizing drilling fluid cooling circulation for control.

[0046] Conversely, when the geological trend is significantly higher than the wear trend, and the difference in the rate of change exceeds a set critical value, the system activates the geological-dominated control mode. This mode focuses on addressing issues such as lag in drilling pressure response and abnormal wellbore deformation caused by formation instability. Optimization objectives will emphasize maintaining a stable bottomhole pressure window, controlling wellbore fluctuations, and suppressing vibration peaks. Formation disturbances may be minimized by adjusting pump displacement, drilling rate, and density gradient.

[0047] The control mode switching process is driven in real time, exhibiting high responsiveness. It continuously evaluates control effectiveness and risk status changes through closed-loop feedback, and can switch between wear-dominated and geological-dominated modes or re-enter a composite control state when necessary. Finally, the target strategy optimization unit invoked by each control mode calculates the optimal drilling parameter combination that meets comprehensive performance requirements based on the objective function and constraints of the current mode using multi-objective optimization algorithms (such as reinforcement learning optimizers and improved particle swarm optimization). This combination is then transmitted to the automatic control execution module to complete command issuance and on-site adjustments, thereby achieving intelligent adaptive parameter control based on operating condition evolution.

[0048] Furthermore, according to the dominant control mode, the target strategy optimization unit is activated to perform drilling parameter optimization and output the optimal drilling parameter combination. Step P40 in this embodiment of the application further includes: P44: Construct a multi-objective optimization function that includes mechanical drilling rate loss rate, drilling pressure deviation, and rotational speed variation coefficient; P45: Set dynamic constraints, including a safe range for the slippage index and downhole pressure not exceeding formation fracture pressure; P46: Based on the multi-objective optimization function and the dynamic constraints, activate the target strategy optimization unit to perform drilling parameter optimization and output the optimal drilling parameter combination.

[0049] Optionally, in further processing, after determining the dominant control mode and completing the mode switch, to achieve precise adjustment of key parameters in the drilling process, a target strategy optimization unit matching the current control mode is activated, and parameter optimization tasks are executed. This process includes not only the selection and activation of optimization algorithms, but also the construction of optimization objectives, the setting of constraint boundaries, and the generation of output parameter combinations. To this end, the system introduces a multi-objective optimization function and combines it with multi-dimensional real-time operating condition constraints to ensure that parameter control satisfies both operational efficiency and operational safety and drill string life.

[0050] First, a multi-objective optimization function is constructed to guide the solution of drilling parameters. This function comprehensively considers multiple drilling performance indicators, covering dimensions such as drilling efficiency, operational stability, and control precision. The optimization objective function includes the following three core evaluation indicators: mechanical rate of penetration loss rate (MRP loss rate), which is the relative difference between the actual rate of penetration and the maximum rate of penetration supported by theoretical formation drillability, used to measure the impact of parameter settings on drilling efficiency; pressure on bit deviation rate (PBD deviation), used to quantify the degree of deviation between the current PBD value and the historical best PBD range, reflecting the accuracy of drill bit force control; and rotational speed variation coefficient, which is the ratio of the fluctuation range of drill bit rotational speed to the average value, used to characterize the stability of drill string rotation. The system normalizes the above objective function items and assigns different weights according to the dominant control mode to achieve optimized trade-offs among multiple objectives.

[0051] Secondly, to ensure that the optimization results operate within a safe and controllable range, dynamic constraints are set to match the actual downhole risk conditions. These constraints include, but are not limited to: a safe range for the sticking / slippage index, calculated based on real-time torque-speed ratio, drilling pressure fluctuations, and drill string vibration characteristics; this index is used to warn of potential sticking or slippage risks and must be maintained within the allowable safe range during optimization; and a downhole pressure control boundary, ensuring that the bottomhole pressure does not exceed the current formation fracture pressure or the critical leakage pressure during optimization to prevent wellbore damage or fluid runaway risks. Furthermore, constraints related to wellbore stability can be introduced, such as upper limits for wellbore change rate and vibration peak thresholds, to further enhance the coverage and adaptability of the constraint system.

[0052] After establishing the objective function and constraint boundaries, the corresponding optimization algorithm structure is selected based on the current dominant control mode (wear-dominated, geological-dominated, or combined mode), and the objective strategy optimization unit is activated to perform drilling parameter optimization calculations. The objective strategy optimization unit can employ multi-objective evolutionary algorithms (such as Improved Particle Swarm Optimization (MOPSO), reinforcement learning methods (such as Depth Deterministic Policy Gradient (DDPG)), or experience-replay hybrid strategy optimizers to calculate the optimal parameter combination under the current drilling state based on the input objective and constraints. The output drilling parameter combination includes, but is not limited to: drilling pressure setpoint, rotational speed setpoint, pump displacement, drilling fluid density, and drill bit contact angle, and this parameter combination is fed back to the automatic control module for on-site execution and dynamic adjustment.

[0053] Furthermore, if the dominant control mode is a wear-dominated control mode, prioritizing the optimization of mechanical drilling rate loss rate and temperature change rate, step P46 in this embodiment further includes: P46-1a: Retrieve successful case parameters with similar wear characteristics from the historical parameter database; P46-2a: Using the successful case parameters as the starting point, conduct step-by-step parameter trials within a preset safety range, continuously adjust, and select the drilling parameter combination that maximizes the temperature gradient reduction and minimizes the mechanical drilling rate loss rate. At the same time, monitor the constraints in real time, forcibly limit the drilling pressure to not exceed the safety upper limit, and the rotational speed to not be lower than the anti-sticking critical value, to obtain the optimal drilling parameter combination.

[0054] In one possible embodiment of this application, when the dominant control factor is determined to be wear-dominant and the system enters the wear-dominant control mode, the optimization logic will prioritize extending drill bit life and slowing down the degradation rate, focusing on the comprehensive control of the two key performance indicators: mechanical drilling rate loss rate and temperature change rate. In this mode, the target strategy optimization unit will adjust the weight structure of the objective function of the optimization strategy, making reducing wear load and maintaining reasonable drilling efficiency the core optimization directions.

[0055] Specifically, to improve the targeting and convergence efficiency of the optimization, a case matching search is first performed in a pre-set historical parameter database. This database contains multiple representative historical successful drilling case samples. Each sample record includes the corresponding wear trend curve, formation information, drilling pressure, rotation speed, pump displacement, and other drilling parameters, as well as the corresponding wear control results and key performance indicators. In this step, the system performs similarity matching between the current wear trend indicators (such as wear level, growth rate, thermo-mechanical synergy intensity, etc.) and historical samples. Algorithms such as Euclidean distance, Mahalanobis distance, or multidimensional cosine similarity can be used to retrieve the successful case parameters with the highest similarity to the current working condition, which serve as the initial reference solution for subsequent optimization processes.

[0056] After obtaining the initial parameter combination, a step-by-step parameter trial search is performed within a preset safety boundary range, using this combination as a benchmark. Specifically, drilling pressure and rotational speed are the main adjustable parameters. Small perturbation increments and decrements are set around the initial values, and trial calculations are performed incrementally. In each iteration, the corresponding temperature gradient change and mechanical drilling rate loss rate are calculated. The temperature gradient reduction is used to measure the mitigation effect of thermal load on the drill bit surface, and the mechanical drilling rate loss rate is used to measure the degree to which drilling efficiency is maintained. The optimal combination is selected from the candidate solution set using a dual-objective ranking rule (such as a Pareto optimal front or a weighted scoring function).

[0057] During parameter probing, the system monitors all dynamic constraints in real time to ensure that the optimization process does not exceed physical and safety boundaries. In particular, for wear-dominated modes, the drilling pressure is forcibly constrained to not exceed the set safety upper limit to avoid accelerated wear caused by sudden increases in local load; at the same time, the rotational speed is limited to not be lower than the preset anti-jamming critical value, which is calculated based on the drill string inertial load, formation friction model, and rotational impact coefficient, and is used to prevent viscous slippage, stuck drill, or a surge in vibration frequency caused by excessively low rotational speed.

[0058] Finally, the optimal combination of drilling parameters that simultaneously satisfies the dual objectives of temperature gradient control and mechanical drilling rate maintenance is selected from the feasible solution space. This combination is then used as the optimal control solution under the current wear-dominated mode and pushed to the control execution module to issue the construction operation command, thereby improving the life of the drill bit and the continuity of drilling operations.

[0059] Furthermore, if the dominant control mode is a geologically dominant control mode, prioritizing the optimization of drill pressure deviation and mechanical drilling rate loss rate, step P46 in this embodiment further includes: P46-1b: Load the corresponding parameter experience table based on the current formation classification results; P46-2b: Using the experience parameters in the parameter experience table as the median, expand and generate a parameter candidate set according to a set ratio; P46-3b: Based on the parameter candidate set, select the drilling parameter combination that minimizes the drilling pressure deviation and the mechanical drilling rate loss rate through real-time simulation calculation; P46-4b: At the same time, adjust the dynamic constraint conditions to limit the downhole pressure to not exceed the safety factor value of the formation fracture pressure, and keep the rotation speed within the recommended rotation speed range for the current lithology.

[0060] Optionally, under the geological-dominated control mode, when the formation trend prediction results indicate that there are obvious lithological abrupt changes, increased wellbore instability, or changes in formation stress structure in the current drilling environment, priority should be given to optimizing the two target indicators of drill pressure deviation and mechanical drilling rate loss rate to achieve a balance between drill pressure stability and drilling efficiency, thereby improving operational stability and propulsion efficiency in complex geological environments.

[0061] In the specific implementation process, the first step is to call the parameter experience table corresponding to the formation type based on the formation classification results output in step P24. This experience table is constructed from a large amount of historical drilling data and a geological-parameter mapping model. It includes recommended typical drilling pressure ranges, rotation speed ranges, pump displacement configurations, and drilling fluid performance parameters under the target lithology conditions. It reflects historical experience and has structural adaptability to the current working conditions, providing a reliable reference for the subsequent generation of parameter candidates.

[0062] After loading the empirical parameters, the main drilling parameters are expanded within a range using the empirical parameter values ​​as the center, according to a set scaling factor (e.g., ±10%, ±15%), generating a parameter perturbation space and constructing multiple parameter candidate sets with differentiated combinations. For example, if the empirical parameter is a drill pressure of 10 MPa, and the expansion ratio is set to ±10%, the drill pressure range in the generated parameter candidate set will be 9 MPa to 11 MPa. This candidate set covers different control strategy values ​​from conservative to aggressive, which is beneficial for subsequent thorough searching and dynamic adaptation in formation environments with high uncertainty.

[0063] Subsequently, based on the generated parameter candidate set, real-time simulation calculations are performed in conjunction with current drilling state variables (such as bottom hole temperature, mud density, well inclination, and well depth). The simulation process employs a simplified force-fluid-mechanical integrated model, or replaces traditional physical modeling methods with a pre-trained neural network prediction model, to perform rapid evaluation on each set of parameter candidate solutions. The simulation output will generate the corresponding drill pressure deviation and mechanical drilling rate loss rate for that combination. Drill pressure deviation refers to the normalized deviation between the currently set drill pressure and the historical recommended drill pressure range for this formation, and is a key indicator for measuring whether the drill pressure "fits the formation"; the mechanical drilling rate loss rate reflects whether drilling efficiency is effectively maintained under the current combination. Through a dual-objective ranking or weighted evaluation mechanism, the optimal parameter combination that satisfies the minimization requirements of both objectives is selected from the candidate set.

[0064] Simultaneously, under the geologically driven control mode, the dynamic constraints during the optimization process are adaptively adjusted to ensure formation safety and drilling stability. Specifically, the bottom hole annulus pressure is limited to a safety factor value set for the formation fracture pressure. This safety factor can be adjusted according to the lithology type; for example, a stricter safety factor (e.g., 0.85~0.9 times the fracture pressure) is used in fracture zones. Furthermore, the drill bit rotation speed is forcibly constrained within the recommended rotation speed range for the current lithology to prevent excessive rotation speed from causing formation disturbance or wellbore spalling. The upper and lower limits of the rotation speed are determined by the safe operating window provided in the parameter experience table.

[0065] Finally, the optimal drilling parameter combination output from the above optimization process is used for control execution to ensure that drilling operations still have good control stability, footage efficiency and risk suppression capabilities in heterogeneous and complex formation environments, thereby achieving the intelligent adaptive control objective of this scheme under the geological-dominated control mode.

[0066] P50: Based on the optimal combination of drilling parameters, the automatic drilling control module is guided to perform mining control, and the dominant control mode and target weights are dynamically adjusted and optimized based on real-time closed-loop feedback.

[0067] Furthermore, step P50 in this embodiment of the application also includes: P51: Update the multiphysics data sampling window at fixed time intervals; P52: When the trend prediction error continuously exceeds the allowable range, trigger the online update mechanism of the dual-channel trend analysis model, automatically adjust the weight coefficient of the optimization target according to the latest operating condition characteristics, and the weight adjustment range is proportional to the degree of deviation of the current operating condition from the expected value.

[0068] Specifically, based on the optimal drilling parameter combination output from the aforementioned optimization process, commands are sent to the automatic drilling control module to execute real-time adjustments to key drilling parameters such as drilling pressure, rotational speed, and pump displacement, thereby achieving dynamic control and continuous optimization of on-site drilling operations. The automatic drilling control module can be a closed-loop control execution unit built on a PLC or industrial controller, possessing real-time communication capabilities with surface equipment and downhole sensor systems, and capable of precisely applying and adjusting parameters according to optimization commands.

[0069] In this control process, a closed-loop feedback mechanism is further introduced. This involves comparing the real-time physical field data (including thermal, force, acoustic, and wellbore stress fields) collected during drilling with the prediction results of the previous optimization model, and continuously monitoring the operational deviations of key performance indicators (such as actual mechanical drilling rate, drill bit temperature, and wellbore response). When a significant difference is detected between the feedback signal and the model prediction, the adaptability of the current control mode and the optimized weight settings is assessed. If necessary, the dominant control mode is switched, or the weight structure of the objective function is dynamically adjusted to enhance system adaptability and improve control accuracy.

[0070] For example, to ensure the real-time performance and effectiveness of the dual-channel trend prediction model, firstly, the sampling window of the multiphysics data is updated at fixed time intervals, such as every 60 seconds, every 10 meters of advance, or other user-defined cycles. The latest thermo-mechanical-acoustic-stress data package is then re-integrated, and the input feature set is updated to ensure that the data relied upon by the trend analysis and control optimization modules is representative and timely. Simultaneously, a sliding window mechanism is used to save the differences between predicted and actual values ​​from past periods for use by the error trend analysis module.

[0071] Subsequently, when the system continuously detects deviations between the trend prediction results and actual observations exceeding the set error tolerance (e.g., the difference between the predicted and actual wear levels exceeds one level, or the stratigraphic classification confidence drops below the threshold) for multiple sampling cycles, it automatically triggers the online update mechanism of the dual-channel trend analysis model. The online update process can be based on an incremental learning strategy to fine-tune the internal parameters of the model and reconstruct the trend mapping function under the current operating conditions.

[0072] While updating the model, the system automatically adjusts the weight coefficients in the multi-objective optimization function based on the deviation between the current operating conditions and the expected performance. For example, when the actual wear rate exceeds the expected upper limit, the system will increase the weight of the wear control term; when the wellbore stress response changes abruptly or the stability index deteriorates, the system will correspondingly increase the priority of formation control-related objectives. The magnitude of the weight adjustment is proportional to the distance between the current operating condition and the expected target value, and a linear or nonlinear mapping function is used to achieve smooth adjustment, ensuring that the optimization direction is sufficiently sensitive to abnormal states without causing system oscillations.

[0073] By introducing the aforementioned closed-loop feedback execution and online adaptive adjustment mechanism, not only can continuous optimization and dynamic control of drilling parameters be achieved, but the control strategy and target focus can also be adjusted based on actual operational performance, thereby enhancing adaptability and intelligence in nonlinear, unsteady, and complex downhole environments, and effectively ensuring the stability, safety, and efficiency of drilling operations.

[0074] In summary, the embodiments of this application have at least the following technical effects: This application constructs a joint identification and trend prediction model for drill bit wear status and formation condition type through comprehensive perception and fusion analysis of multi-physics field information, thereby improving the completeness and feedforward nature of downhole condition perception. By introducing a dynamic judgment mechanism for the dominant control factor, it enables on-demand switching between wear-dominated and geological-dominated control modes, and improves the responsiveness and adaptability of parameter control by combining adaptive weight adjustment of optimization objectives. By constructing a multi-objective optimization function and setting dynamic safety constraints, it ensures a comprehensive balance between drill bit life, drilling efficiency, and downhole operation stability. At the same time, relying on a closed-loop feedback mechanism and an online trend model update mechanism, it achieves continuous correction of deviations from actual working conditions, enhancing the system's intelligent control capability and robustness in complex and unsteady drilling environments.

[0075] It achieves the technical effect of identifying and adaptively switching the dominant control mode based on multi-field coupling trend analysis, thereby improving the response speed and adaptability to complex working conditions.

[0076] Example 2 is based on the same inventive concept as the intelligent optimization control method for drilling parameters based on multi-field coupling in the previous examples, such as... Figure 2 As shown, this application provides an intelligent optimization control system for drilling parameters based on multi-field coupling. The system and method embodiments in this application are based on the same inventive concept. The system includes: The field data acquisition module 11 is used to acquire real-time multi-physics field data during the drilling process. The real-time multi-physics field data includes thermal field data, force field data, acoustic field data, and wellbore stress field data.

[0077] The co-evolution trend prediction module 12 is used to construct a linkage prediction model based on the real-time data of the multi-physics field, and simultaneously identify the drill bit wear state and formation working condition type, and determine the co-evolution trend of the two to obtain wear trend prediction results and geological trend prediction results.

[0078] The main control factor determination module 13 is used to determine the current dominant control factor based on the wear trend prediction results and the geological trend prediction results.

[0079] The drilling parameter optimization module 14 is used to switch to the corresponding dominant control mode based on the dominant control factor, and activate the target strategy optimization unit to perform drilling parameter optimization according to the dominant control mode, outputting the optimal combination of drilling parameters. The dominant control modes include wear-dominated control mode, geological-dominated control mode, and composite control mode.

[0080] The mining control and feedback module 15 is used to guide the automatic drilling control module to perform mining control according to the optimal drilling parameter combination, and dynamically adjust the dominant control mode and optimize the target weight according to real-time closed-loop feedback.

[0081] Furthermore, the co-evolutionary trend prediction module 12 is also used to perform the following steps: The temperature distribution in the thermal field data is segmented into regions, and the temperature change rate of each region is extracted. Combined with the fluctuation period of axial pressure in the force field data, a dynamic wear index is calculated. The acoustic field data is divided into multiple segments according to time windows, and the frequency component features of each segment are extracted. The frequency component features are then spatially matched with the stress concentration area distribution in the wellbore stress field data to generate stress distribution features. Through a multi-layer feature fusion network, the dynamic wear index is compared with a preset wear stage threshold to output the drill bit wear level. The frequency component features and the stress distribution features are input into a formation feature matching library to output a formation classification result with confidence.

[0082] Furthermore, the co-evolutionary trend prediction module 12 is also used to perform the following steps: A dual-channel trend analysis model is constructed, comprising a first channel and a second channel. The first channel generates a wear trend index by accumulating the product of the temperature change rate and the pressure fluctuation amplitude, and superimposing the proportion of high-frequency energy in the acoustic signal. The second channel generates a geological trend index by calculating the dynamic ratio of vertical stress to horizontal stress and combining it with the slope of formation pressure change. Based on the dual-channel trend analysis model, a sliding time window is used to predict the wear trend growth rate and geological trend change rate within a set future time period.

[0083] Furthermore, the drilling parameter optimization module 14 is also used to perform the following steps: When the quantitative difference between the wear trend and the geological trend is less than the coordination threshold and the maximum trend growth rate reaches the critical value, the composite control mode is activated; when the wear trend is higher than the geological trend and its growth rate exceeds the set proportional critical value, the wear-dominated control mode is activated; when the geological trend is higher than the wear trend and its rate of change exceeds the set proportional critical value, the geological-dominated control mode is activated.

[0084] Furthermore, the drilling parameter optimization module 14 is also used to perform the following steps: A multi-objective optimization function is constructed, which includes mechanical drilling rate loss rate, drilling pressure deviation, and rotational speed variation coefficient; dynamic constraints are set, including the safe range of the slippage index and the downhole pressure not exceeding the formation fracture pressure; based on the multi-objective optimization function and the dynamic constraints, the target strategy optimization unit is activated to perform drilling parameter optimization and output the optimal drilling parameter combination.

[0085] Furthermore, the drilling parameter optimization module 14 is also used to perform the following steps: If the dominant control mode is the wear-dominated control mode, priority is given to optimizing the mechanical drilling rate loss rate and temperature change rate, including: retrieving successful case parameters with similar wear characteristics from the historical parameter database; using the successful case parameters as the initial point, conducting step-by-step parameter trials within a preset safety range, continuously adjusting, and selecting the drilling parameter combination that maximizes the temperature gradient reduction and minimizes the mechanical drilling rate loss rate; at the same time, monitoring the constraints in real time, forcibly limiting the drilling pressure to not exceed the safety upper limit, and the rotational speed to not be lower than the anti-sticking critical value, to obtain the optimal drilling parameter combination.

[0086] Furthermore, the drilling parameter optimization module 14 is also used to perform the following steps: If the dominant control mode is the geological-dominated control mode, priority is given to optimizing the drilling pressure deviation and mechanical drilling speed loss rate, including: loading the corresponding parameter experience table according to the current formation classification results; using the experience parameters in the parameter experience table as the median, expanding and generating a parameter candidate set according to a set ratio; based on the parameter candidate set, selecting the drilling parameter combination that minimizes the drilling pressure deviation and the mechanical drilling speed loss rate through real-time simulation calculation; at the same time, adjusting the dynamic constraint conditions to limit the downhole pressure to not exceed the safety factor value of the formation fracture pressure, and keeping the rotation speed within the recommended rotation speed range for the current lithology.

[0087] Furthermore, the mining control and feedback module 15 is also used to perform the following steps: The multiphysics data sampling window is updated at fixed time intervals. When the trend prediction error exceeds the allowable range continuously, the online update mechanism of the dual-channel trend analysis model is triggered. The weight coefficients of the optimization target are automatically adjusted according to the latest operating conditions. The weight adjustment range is proportional to the degree to which the current operating conditions deviate from the expected value.

[0088] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0089] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0090] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and variations fall within the scope of this application and its equivalents, this application intends to include such modifications and variations.

Claims

1. A method for intelligent optimization control of drilling parameters based on multi-field coupling, characterized in that, The method comprises: Collecting multi-physical field real-time data in the drilling process, wherein the multi-physical field real-time data comprises thermal field data, force field data, acoustic field data and well wall stress field data; Based on the multi-physical field real-time data, a linkage prediction model is constructed, the drilling bit wear state and the formation working condition type are identified, and the collaborative evolution trend of the two is judged to obtain wear trend prediction results and geological trend prediction results; According to the wear trend prediction results and the geological trend prediction results, the current control dominant factor is judged; Based on the control dominant factor, the corresponding dominant control mode is switched to, and according to the dominant control mode, the target strategy optimization unit is activated to execute drilling parameter optimization to output the optimal drilling parameter combination; According to the optimal drilling parameter combination, the automatic drilling control module is guided to carry out mining control, and the dominant control mode and the optimization target weight are dynamically adjusted according to real-time closed-loop feedback.

2. The method for intelligent optimization control of drilling parameters based on multi-field coupling of claim 1, wherein, The dominant control mode comprises a wear dominant control mode, a geological dominant control mode and a composite control mode.

3. The method for intelligent optimization control of drilling parameters based on multi-field coupling of claim 1, wherein, Based on the multi-physical field real-time data, a linkage prediction model is constructed, the drilling bit wear state and the formation working condition type are identified, and the collaborative evolution trend of the two is judged to obtain wear trend prediction results and geological trend prediction results, comprising: The temperature distribution in the thermal field data is regionally segmented, the temperature change rate of each region is extracted, the dynamic wear index is calculated in combination with the fluctuation period of the axial pressure in the force field data; The acoustic field data is divided into multiple segments according to the time window, the frequency component features of each segment are extracted, and the frequency component features are spatially matched with the stress concentration area distribution in the well wall stress field data to generate stress distribution features; Through a multi-layer feature fusion network, the dynamic wear index is compared with a preset wear stage threshold to output a drilling bit wear grade; The frequency component features and the stress distribution features are input into a formation feature matching library to output a formation classification result with a confidence degree.

4. The method for intelligent optimization control of drilling parameters based on multi-field coupling of claim 3, wherein, The judgment of the collaborative evolution trend of the two, the wear trend prediction results and the geological trend prediction results, comprise: A double-channel trend analysis model is constructed, the double-channel trend analysis model comprises a first channel and a second channel, wherein: The first channel generates a wear trend index by accumulating the product value of the temperature change rate and the pressure fluctuation amplitude, and superimposing the high-frequency energy proportion in the acoustic wave signal; The second channel generates a geological trend index by calculating the dynamic ratio of the vertical stress to the horizontal stress in combination with the formation pressure change slope; Based on the double-channel trend analysis model, a sliding time window is used to predict the wear trend growth rate and the geological trend change rate in a future set time period.

5. The method for intelligent optimization control of drilling parameters based on multi-field coupling of claim 4, wherein, Based on the control dominant factor, the corresponding dominant control mode is switched to, comprising: When the quantitative difference between the wear trend and the geological trend is less than a collaborative threshold, and the maximum trend growth rate reaches a critical value, a composite control mode is started; When the wear trend is higher than the geological trend and the growth rate thereof exceeds a set proportion critical value, a wear dominant control mode is activated; When the geological trend is higher than the wear trend and the change rate thereof exceeds a set proportion critical value, a geological dominant control mode is activated.

6. The method for intelligent optimization control of drilling parameters based on multi-field coupling of claim 2, wherein, According to the dominant control mode, the target strategy optimization unit is activated to execute drilling parameter optimization to output the optimal drilling parameter combination, comprising: A multi-objective optimization function is constructed, including a rate of loss of rate of penetration, a bit deviation degree, and a coefficient of variation of rotary speed; Dynamic constraints are set, including a safety range of a sticking probability and a downhole pressure not exceeding a formation breakdown pressure; Based on the multi-objective optimization function and the dynamic constraints, a target strategy optimization unit is activated to perform drilling parameter optimization, and an optimal drilling parameter combination is output.

7. The method for multi-field coupling based intelligent optimization control of drilling parameters as claimed in claim 6, wherein, If the dominant control mode is a wear dominant control mode, the rate of loss of rate of penetration and the temperature change rate are preferentially optimized, including: Successful case parameters with similar wear characteristics are retrieved from a historical parameter library; The successful case parameters are used as initial points, and a step-by-step parameter trial is performed within a preset safety range, and the drilling parameter combination that maximizes the temperature gradient reduction and minimizes the rate of loss of rate of penetration is selected, while the constraints are monitored in real time, and the drilling pressure is forced to be limited to a safety upper limit and the rotary speed is forced to be limited to a critical value to prevent sticking, to obtain the optimal drilling parameter combination.

8. The method for intelligent optimization control of drilling parameters based on multi-field coupling of claim 7, wherein, If the dominant control mode is a geology dominant control mode, the bit deviation degree and the rate of loss of rate of penetration are preferentially optimized, including: According to the current formation classification result, a corresponding parameter experience table is loaded; The experience parameters in the parameter experience table are used as a median value, and a parameter candidate set is generated by expanding at a set proportion; Based on the parameter candidate set, a drilling parameter combination that minimizes the bit deviation degree and the rate of loss of rate of penetration is selected through real-time simulation calculation; At the same time, the dynamic constraints are adjusted to limit the downhole pressure to a safety factor value not exceeding the formation breakdown pressure, and the rotary speed is limited to a current lithology recommended rotary speed interval.

9. The method for intelligent optimization control of drilling parameters based on multi-field coupling of claim 4, wherein, The dominant control mode and the optimization target weight are dynamically adjusted according to real-time closed-loop feedback, including: The multi-physical field data sampling window is updated at a fixed time interval; When the trend prediction error continuously exceeds the allowed range, a double-channel trend analysis model online updating mechanism is triggered, and the weight coefficient of the optimization target is automatically adjusted according to the latest working condition characteristics, and the weight adjustment amplitude is proportional to the degree of deviation of the current working condition from the expected value.

10. A drilling parameter intelligent optimization control system based on multi-field coupling, characterized in that, The system includes: A field data acquisition module, which is configured to acquire multi-physical field real-time data in a drilling process, including thermal field data, force field data, acoustic field data, and wellbore stress field data; A collaborative evolution trend prediction module, which is configured to construct a linkage prediction model based on the multi-physical field real-time data, simultaneously identify a drill bit wear state and a formation working condition type, and judge a collaborative evolution trend of the two, to obtain a wear trend prediction result and a geology trend prediction result; A main control factor determination module, which is configured to determine a current control dominant factor according to the wear trend prediction result and the geology trend prediction result; A drilling parameter optimization module, which is configured to switch to a corresponding dominant control mode based on the control dominant factor, and activate a target strategy optimization unit to perform drilling parameter optimization according to the dominant control mode, to output an optimal drilling parameter combination. a production control and feedback module for guiding automatic drilling control module for production control according to the optimal drilling parameter combination, and dynamically adjusting the main control mode and optimization target weight according to real-time closed-loop feedback.

Images