Adaptive drilling control method and system based on multi-mode power switching

Abstract

This invention discloses an adaptive drilling control method and system based on multi-mode power switching, belonging to the field of drilling control technology. The method includes the following steps: collecting sensor data during the drilling process; determining potential switching risk intervals and candidate power mode switching trigger conditions through feature extraction and joint pattern recognition; predicting the impact of different power mode switching transients on drill string vibration in real time based on the trigger conditions; generating different power mode switching schemes using a multi-objective weighted optimization algorithm based on the prediction results, and selecting the optimal switching scheme; executing the power mode switching based on the optimal switching scheme, and using the process data to update the prediction model and optimization strategy. This invention effectively suppresses drill string vibration by predicting the power switching process and dynamically optimizing the switching strategy, thereby solving the problems of unpredictable power switching transients and difficulty in coordinating vibration control and the switching process in existing technologies.

Description

Technical Field

[0001] This invention relates to the field of drilling control technology, and more specifically, to an adaptive drilling control method and system based on multi-mode power switching. Background Technology

[0002] In oil and gas drilling, geological exploration, and construction in complex formations, drill string systems typically need to frequently switch power modes between different operating conditions to adapt to formation changes, load fluctuations, and drilling efficiency requirements. Existing drilling equipment generally achieves power mode switching by adjusting rotational speed, torque, or power output to improve drilling stability or increase operational efficiency.

[0003] However, in practical engineering applications, drill string systems exhibit significant nonlinear, strongly coupled, and time-varying characteristics. Their dynamic response is influenced by a variety of factors, including formation characteristics, drill string structure, changes in operating conditions, and control strategies. In particular, during power mode switching, sudden changes in power input can easily trigger transient accumulation of drill string vibration energy, which can induce strong axial, torsional, or coupled vibrations, leading to decreased drilling efficiency and even fatigue damage or failure of the drill string.

[0004] In existing technologies, power mode switching is typically controlled based on empirical thresholds or single operating parameters, lacking an effective prediction mechanism for the transient dynamic behavior during switching. This makes it difficult to assess the impact of different switching modes on drill string vibration in advance, resulting in significant uncertainty in the switching process. Furthermore, existing methods often treat power switching and vibration control separately, lacking unified prediction and collaborative optimization methods, making it difficult to achieve stable and efficient power switching control under nonlinear and time-varying operating conditions.

[0005] The above-disclosed technical solutions have at least the following technical problems: the transient impact of power switching is difficult to predict, which makes it difficult to coordinate vibration control and power mode switching. Existing technologies cannot achieve predictive drive switching optimization in nonlinear, time-varying drill string power environments. Summary of the Invention

[0006] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide an adaptive drilling control method and system based on multi-mode power switching. By predicting the power switching process and dynamically optimizing the switching strategy, the method effectively suppresses drill string vibration, thereby solving the problems of difficulty in predicting transient power switching and difficulty in coordinating vibration control and the switching process in the prior art.

[0007] To achieve the above objectives, the present invention provides the following technical solution: On the one hand, the adaptive drilling control method based on multi-mode power switching includes the following steps: collecting sensor data during the drilling process, determining the potential switching risk range during the drilling process and identifying candidate power mode switching trigger conditions through feature extraction and joint pattern recognition; predicting the impact of different power mode switching transients on drill string vibration in real time based on the trigger conditions, and generating prediction results; generating different power mode switching schemes based on the prediction results through a multi-objective weighted optimization algorithm, and selecting the comprehensive optimal switching scheme; executing the power mode switching based on the optimal switching scheme, continuously monitoring drill string vibration and working condition changes, and using process data to update the prediction model and optimization strategy to improve the performance of the next switching.

[0008] In a preferred embodiment, the step of determining the potential switching risk range during the drilling process through feature extraction and joint pattern recognition includes: constructing a time-series feature vector reflecting the relationship between the changes of each parameter over time based on the multi-source operating parameters obtained during the drilling process; A time-series correlation analysis is performed on the time-series feature vectors to extract the relative change trends and phase shift relationships between parameters; based on the change trends and shift relationships, it is identified whether there are characteristic patterns that characterize the evolution of the drilling system from a stable state to an unstable state; when the characteristic patterns are detected, the corresponding time interval is determined as a potential dynamic mode switching risk interval.

[0009] In a preferred embodiment, identifying whether there exists a characteristic pattern representing the evolution of the drilling system from a stable state to an unstable state based on the changing trend and offset relationship includes: constructing a parameter co-evolution relationship matrix by combining the rate of change and the phase difference; extracting features from the co-evolution relationship matrix to obtain a co-instability feature quantity used to characterize the system stability; determining whether the drilling system has evolved from a stable state to an unstable state based on the co-instability feature quantity; and when the unstable evolution feature persists in the time dimension, determining the corresponding time interval as a potential dynamic mode switching risk interval.

[0010] In a preferred embodiment, determining the candidate power mode switching trigger condition includes: within the potential switching risk interval, constructing a power state sequence characterizing the state evolution of the drilling process based on operating parameters; calculating the state change rate and energy accumulation rate of the power state trajectory to obtain an instability approach index characterizing the degree to which the system approaches an unstable state; based on the instability approach index and combined with the minimum response time required for power mode switching, determining the time interval for the system state to evolve from stable to unstable; using the time interval as a candidate trigger interval for power mode switching, and generating corresponding power mode switching trigger conditions accordingly.

[0011] In a preferred embodiment, the real-time prediction of the impact of transient power mode switching on drill string vibration based on the triggering condition, and the generation of prediction results, includes: acquiring the drilling pressure, rotational speed, torque, and vibration signals within a preset time window before the switching triggering time, and constructing a state reference vector characterizing the current dynamic characteristics of the drill string based on the signals; determining the corresponding power mode switching method based on the triggering condition, and parameterizing the power output change characteristics during the switching process to form a switching feature vector; constructing a mapping model to characterize the impact of transient power mode switching on drill string vibration using the state reference vector and the switching input vector as joint inputs; and outputting the drill string vibration response prediction result under the corresponding power mode switching condition based on the prediction model.

[0012] In a preferred embodiment, determining the corresponding power mode switching method based on the triggering condition includes: during the system initialization phase, pre-constructing a power mode set containing multiple power output control methods to describe the selectable power mode types during drilling; after detecting the power mode switching triggering condition, judging the current drilling state based on the instability approach index to determine the power mode switching type corresponding to the current instability characteristics; according to the switching type, quantifying and setting key control parameters during the power mode switching process to form a switching parameter set describing the power mode change process; and uniformly parameterizing and encoding the switching parameter set to generate a switching input vector characterizing the transient characteristics of the power mode switching.

[0013] In a preferred embodiment, the step of generating different power mode switching schemes based on the prediction results using a multi-objective weighted optimization algorithm includes: analyzing the temporal variation characteristics of the drill string during the power mode switching process based on the prediction results, and determining the variation law of each vibration response index during the switching process; dividing the power mode switching process into several transient stages according to the variation law, and determining the dominant influence range of each stage; constructing corresponding evaluation indices for each transient stage, and assigning weights according to their influence on system stability to form a multi-objective weighted evaluation function; based on the set of stage evaluation indices, and combined with the differences in the influence of each transient sub-stage on system stability, assigning corresponding weight coefficients to different evaluation indices to construct a multi-objective weighted evaluation function; and jointly optimizing different power modes and their switching parameters under the constraints of the multi-objective weighted evaluation function to obtain a set of power mode switching schemes.

[0014] In a preferred embodiment, the step of dividing the power mode switching process into several transient stages according to the change pattern and determining the dominant influence range of each stage includes: extracting the vibration amplitude, energy change rate, and vibration attenuation characteristics characterizing the drill string dynamic state based on the vibration response results output by the mapping model, as vibration response indicators; analyzing the temporal changes of each vibration response indicator during the power mode switching process, calculating its change intensity and cumulative influence in different time intervals; comparing the change intensity of each vibration response indicator in the same time interval to determine the vibration response indicator with the greatest impact on drill string stability; and determining the dominant influence range of an indicator when a certain vibration response indicator is continuously in a dominant state in a continuous time interval.

[0015] In a preferred embodiment, the step of performing power mode switching based on the optimal switching scheme and continuously monitoring drill string vibration and operating condition changes includes: adjusting power output according to the optimal power mode switching scheme and collecting process data during the switching process; comparing and analyzing the process data with the corresponding switching prediction results to calculate the deviation between the actual vibration response and the predicted vibration response; based on the deviation results, correcting the mapping model parameters used to describe the relationship between power mode switching and vibration response, so that the model output gradually approaches the actual response characteristics; adaptively adjusting the weights and constraint parameters of each stage in the multi-objective optimization according to the actual vibration performance at each stage during the switching process, so that the optimization results are consistent with the actual stability requirements; and using the updated model parameters and optimization parameters as input for subsequent power mode switching prediction and decision-making to achieve continuous adaptive optimization of the power mode switching strategy.

[0016] On the other hand, the adaptive drilling control system based on multi-mode power switching includes the following modules: The system comprises the following modules: a potential switching risk identification module (for collecting sensor data during drilling, identifying potential switching risk ranges and determining candidate power mode switching trigger conditions through feature extraction and joint pattern recognition), a switching impact prediction module (for predicting the impact of different power mode switching transients on drill string vibration based on the trigger conditions, and generating prediction results), a switching scheme optimization decision module (for generating different power mode switching schemes based on the prediction results using a multi-objective weighted optimization algorithm, and selecting the optimal switching scheme), and a power mode switching module (for executing power mode switching based on the optimal switching scheme, continuously monitoring drill string vibration and operating condition changes, and using process data to update the prediction model and optimization strategy to improve the performance of the next switching).

[0017] The technical effects and advantages of the adaptive drilling control method and system based on multi-mode power switching of this invention are as follows: This invention introduces a predictive mechanism for the transient impact of power mode switching during drilling. Before switching, the drill string vibration response under different power modes is evaluated, and a multi-objective weighted optimization method is used to select the optimal switching scheme, achieving coordinated optimization of power mode switching and drill string vibration control. Compared to existing control methods that rely on experience or single-parameter triggering, this invention can identify potential switching risks in advance under nonlinear, time-varying drilling conditions, avoiding sudden increases in vibration energy during power switching, thereby improving the stability and controllability of the switching process. Attached Figure Description

[0018] Figure 1 This is a flowchart illustrating the adaptive drilling control method based on multi-mode power switching of the present invention. Figure 2 This is a schematic diagram of the adaptive drilling control system based on multi-mode power switching of the present invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0020] Example 1, Figure 1 The present invention provides an adaptive drilling control method based on multi-mode power switching, comprising the following steps: S1: Collect sensor data during the drilling process, and determine the potential switching risk range during the drilling process and the candidate power mode switching trigger conditions through feature extraction and joint pattern recognition. In this embodiment, determining the potential switching risk range during the drilling process through feature extraction and joint pattern recognition includes: Based on the drilling pressure, torque, rotational speed and vibration signals collected during the drilling process, a time-series feature vector reflecting the relationship between the changes of each parameter and time is constructed. Perform time-series correlation analysis on the time-series feature vectors to extract the relative change trends and phase shift relationships between parameters; Based on the changing trend and offset relationship, identify whether there is an unstable evolution feature pattern formed by the coordinated change of multiple parameters, wherein the unstable evolution feature pattern is used to characterize the process of the drilling state shifting from the stable zone to the dynamic instability zone; When the unstable evolutionary feature pattern is detected, the corresponding time interval is determined as a potential switching risk interval.

[0021] The process of identifying whether an unstable evolutionary feature pattern exists, formed by the coordinated changes of multiple parameters, based on the aforementioned change trends and offset relationships, includes: The rate of change and the phase difference are combined to construct a parameter co-evolution relationship matrix, which is used to characterize the dynamic coupling state between the parameters; Feature extraction is performed on the co-evolutionary relationship matrix to obtain co-instability feature quantities used to characterize system stability. The co-instability feature quantities include at least: the synchronous decrease magnitude of multi-parameter correlation; the degree of parameter response timing misalignment; and the degree of consistency of multi-parameter energy change direction. The cooperative instability feature quantity is compared with the feature distribution corresponding to the historical stable drilling state. When at least one of the following conditions is met, it is determined that there is an unstable evolution feature mode: the correlation of multiple parameters rapidly decays from the stable range to below the preset threshold; the phase difference between parameters changes from a stable state to a continuously expanding offset state; the direction of energy change of multiple parameters changes from consistent evolution to mutual cancellation or alternating enhancement. When the unstable evolutionary feature pattern persists for multiple consecutive time windows, the corresponding time period is marked as a potential switching risk interval.

[0022] The conditions for determining the candidate power mode switching trigger include: Within the potential switching risk range, continuous change sequences of parameters such as drilling pressure, rotational speed, torque, and vibration are extracted to construct a dynamic state trajectory reflecting the evolution of the drilling system from a stable state to an unstable state, which is used to characterize the dynamic evolution trend of the current working condition. The rate of state change and the rate of energy accumulation are calculated on the dynamic state trajectory to obtain an instability approach index that characterizes the degree to which the system approaches an unstable state. The instability approach index is used to characterize the remaining margin of the system from severe vibration or instability. The instability approach index is obtained by weighting the rate of state change and the rate of energy accumulation. Based on the instability approach index and the minimum response time required for power mode switching, a time window is determined in which the system can still achieve a stable transition through power mode adjustment. This time window is defined as the candidate power mode switching window. When the instability approach index enters the candidate dynamic mode switching window, and the system has not yet entered an irreversible instability state, the corresponding dynamic mode switching trigger condition is generated.

[0023] The specific formula for calculating the rate of state change is as follows:

[0024] The energy accumulation rate is specifically calculated using the following formula:

[0025]

[0026] in, The rate of change of state. The dynamic state vector at time t (composed of drilling pressure, torque, rotational speed, axial vibration characteristic quantity, and torsional vibration characteristic quantity). For the time difference, The rate of energy accumulation. Vibrational energy, The length of the sliding time window. This represents the axial vibration amplitude. This represents the amplitude of torsional vibration.

[0027] S2, based on the triggering conditions, the influence of transient switching of different power modes on drill string vibration is predicted in real time, and prediction results are generated; In this embodiment, the real-time prediction of the impact of transient switching between different power modes on drill string vibration based on the triggering conditions, and the generation of prediction results, includes: Obtain drilling pressure, rotation speed, torque and vibration signals within a preset time window before the switching trigger time, and construct a state reference vector based on the signals to characterize the current drill string dynamics, which is used to describe the energy distribution state and dynamic response characteristics of the system before the switching occurs; Based on the triggering conditions, the corresponding power mode switching method is determined, and the magnitude, rate of change and duration of power output change during the switching process are parameterized to form a switching input vector to characterize the transient characteristics of the switching. Using the state reference vector and the switching input vector as joint inputs, a mapping model is constructed to characterize the impact of the transient dynamic mode switching on the drill string vibration. The mapping model is used to output the vibration response characteristics of the drill string during the switching process, including the vibration amplitude change trend, energy growth rate and decay characteristics. Substituting the switching input vector into the mapping model, the vibration response prediction results under the corresponding dynamic mode switching conditions are obtained.

[0028] It should be noted that determining the corresponding power mode switching method based on the triggering condition includes: During the system initialization phase, at least two switchable power modes are predefined based on the power output structure and control method of the drilling equipment. The power modes include different speed regulation methods, torque distribution strategies or power output control methods, and a power mode candidate set is formed to describe possible power switching schemes. After detecting the power mode switching trigger condition, the current drilling state is classified and judged in combination with the instability approach index and its changing trend to determine the type of current instability evolution. Based on different instability types, a matching power mode switching strategy type is selected. The types include: vibration enhancement type dominated by rotational speed fluctuation; energy accumulation type caused by torque mutation; and composite instability type caused by the imbalance of drilling pressure and rotational speed coupling. After determining the switching strategy type, the key control parameters during the power mode switching process are quantified and set according to the magnitude and rate of change of the current instability approach index, and the above parameters are combined to form a switching parameter set to describe the dynamic characteristics of the switching process; the key control parameters include: the magnitude of power output change; the rate of change of power adjustment; and the duration of power mode transition. The switching parameter group is uniformly encoded according to a preset parameter dimension to form a switching input vector used to characterize the transient characteristics of power mode switching.

[0029] The step of constructing a mapping model to characterize the transient impact of dynamic mode switching on drill string vibration by using the state reference vector and the switching input vector as joint inputs includes: The state reference vector and the switching input vector are uniformly normalized and combined according to physical correlation to form a joint input feature vector containing system operating state characteristics and dynamic switching characteristics, which is used to characterize the comprehensive dynamic state of the drill string when the switching occurs. Based on the dynamic characteristics of the drill string, the joint input feature vector is mapped to the vibration response space to construct a causal mapping structure with state features and switching features as inputs and vibration response features as outputs, which is used to characterize the evolution law of drill string vibration response under different switching conditions. In the mapping structure, a switching sensitivity factor is introduced to characterize the transient characteristics of the switching process. The switching sensitivity factor is composed of switching amplitude, switching rate and switching duration, and is used to characterize the difference in excitation intensity of the drill string dynamic response to different switching modes. Based on the aforementioned joint input features and switching sensitivity factors, a mapping model is established to output the vibration response characteristics of the drill string. The mapping model is used to output the vibration amplitude variation trend, energy growth rate, and vibration attenuation characteristics, thereby characterizing the degree of influence of dynamic mode switching transients on drill string stability.

[0030] The mapping model can be implemented using a machine learning model trained on historical drilling data, or using a fast dynamic simulation model based on the drill string dynamics equations, to establish the mapping relationship between the power mode switching input and the drill string vibration response.

[0031] S3. Based on the prediction results, a multi-objective weighted optimization algorithm is used to generate different power mode switching schemes, and the comprehensive optimal switching scheme is selected. In this embodiment, the step of generating different power mode switching schemes based on the prediction results using a multi-objective weighted optimization algorithm includes: Based on the prediction results, the temporal variation characteristics of vibration amplitude, vibration energy growth rate and vibration attenuation characteristics of the drill string during the power mode switching process are analyzed to determine the variation law of each vibration response index during the switching process. Based on the changing patterns, the dominant range of each vibration response index in the dynamic mode switching process is identified, and the dynamic mode switching process is divided into multiple transient sub-stages, including at least the initial switching stage, the energy injection stage, and the stable recovery stage. For different transient sub-stages, vibration response indices that play a dominant role in the stability of the system in that stage are selected, and corresponding set of stage evaluation indices are constructed so that each evaluation index participates in the optimization calculation only in its corresponding transient sub-stage. Based on the set of stage evaluation indicators, and considering the differences in the degree of influence of each transient sub-stage on system stability, corresponding weight coefficients are assigned to different evaluation indicators to construct a multi-objective weighted evaluation function. Under the constraints of the multi-objective weighted evaluation function, a consistency constraint condition across transient sub-stages is introduced to limit the abrupt changes in vibration amplitude, energy change trend and attenuation characteristics between adjacent transient sub-stages, so as to avoid the local optimal switching scheme from causing vibration amplification or stability degradation in subsequent stages. Based on the set of stage evaluation indicators and consistency constraints, different power modes and their switching parameter combinations are optimized in stages to generate a set of power mode switching schemes that meet the stability requirements of each transient sub-stage. Within the set of switching schemes, the cumulative vibration risk of each scheme during the entire switching process is comprehensively evaluated. The switching scheme with the optimal comprehensive evaluation result is selected as the final power mode switching scheme to be executed. The multi-objective weighted optimization algorithm takes vibration response characteristics as the core constraint, while also considering the comprehensive impact of drilling efficiency, power consumption, and drill string wear on system operating performance. By assigning different weights to each optimization objective and constructing a unified weighted evaluation function, the algorithm achieves a comprehensive optimal solution for the power mode switching scheme.

[0032] Furthermore, the identification of the dominant influence range of each vibration response index during the dynamic mode switching process based on the changing patterns includes: Based on the vibration response characteristics output by the mapping model, the vibration amplitude, vibration energy growth rate and vibration attenuation characteristics that characterize the vibration state of the drill string are obtained and used as vibration response indicators. The time series changes of various vibration response indicators during the dynamic mode switching process were analyzed, and their rate of change and cumulative change in different time periods were calculated. By comparing the rate of change and cumulative change of each vibration response index at the same time, the vibration response index that has the greatest impact on the stability of the drill string during that time period can be identified. When the change in a certain vibration response index within a continuous time window exceeds a preset threshold as a proportion of its historical change range, that time interval is determined as the dominant range of the vibration response index. Based on the dominant action range corresponding to each vibration response index, the entire process of dynamic mode switching is divided into time periods to form a set of transient sub-stages for subsequent phased optimization.

[0033] The process involves comprehensively evaluating the cumulative vibration risk of each scheme within the set of switching schemes throughout the switching process, and selecting the switching scheme with the optimal comprehensive evaluation result as the final dynamic mode switching scheme to be executed. This includes: Obtain the predicted vibration response values ​​for each candidate dynamic mode switching scheme in each transient sub-stage, including the vibration amplitude change, vibration energy increase, and vibration attenuation characteristic parameters; Within each transient sub-stage, the predicted vibration response value is subjected to risk quantification based on the dominant vibration response index corresponding to that stage, thereby obtaining a stage risk value characterizing the degree of vibration instability in that stage. The stage risk values ​​corresponding to each transient sub-stage of the same switching scheme are cumulatively calculated to form a vibration risk accumulation index that reflects the switching scheme throughout the entire switching process. When accumulating risks, a stage weighting factor is introduced to weight and correct the risk contribution of different transient sub-stages, so as to highlight the weight of the stage with a greater impact on system stability in the comprehensive evaluation. The cumulative vibration risk indices of each switching scheme are compared and analyzed. The switching scheme with the smallest cumulative vibration risk index and that meets the preset stability constraints is selected as the final dynamic mode switching scheme to be implemented.

[0034] S4 executes the power mode switch based on the optimal switching scheme, continuously monitors drill string vibration and changes in operating conditions, and uses process data to update the prediction model and optimization strategy to improve the performance of the next switch.

[0035] In this embodiment, the process of performing power mode switching based on the optimal switching scheme, continuously monitoring drill string vibration and changes in operating conditions, and using process data to update the prediction model and optimization strategy to improve the performance of the next switching includes: The power output is adjusted according to the optimal power mode switching scheme. During the switching process, drilling pressure, rotation speed, torque and vibration signals are collected in real time to obtain process data reflecting the switching effect. The process data is compared and analyzed with the corresponding switching prediction results, and the deviation between the actual vibration response and the predicted vibration response is calculated to characterize the error level of the prediction model under the current working condition. Based on the aforementioned deviation results, the parameters of the mapping model used to describe the relationship between dynamic mode switching and vibration response are corrected so that the model output gradually approximates the actual response characteristics. Meanwhile, based on the actual vibration performance of each transient sub-stage during the switching process, the stage weights and constraint parameters used in the multi-objective optimization process are adaptively adjusted to correct the degree of influence of different stages on the switching stability. The updated mapping model parameters and optimized parameters are used as inputs for subsequent power mode switching prediction and decision-making, enabling the system to gradually develop adaptive control capabilities that match the actual operating conditions during multiple switching processes.

[0036] It should be noted that the correction of the mapping model parameters used to describe the relationship between dynamic mode switching and vibration response based on the deviation results includes: Based on the process data and corresponding prediction results, an error vector is constructed between the actual vibration response and the predicted vibration response to characterize the prediction deviation characteristics of different vibration response indices during the dynamic mode switching process. Based on the distribution characteristics of the error vector in each transient sub-stage, identify the model parameters that contribute significantly to the prediction error and determine their sensitivity weights in the corresponding stages. With minimizing the error vector as the optimization objective, and while keeping the model structure unchanged, the highly sensitive parameters are incrementally corrected to obtain an updated set of model parameters. The updated model parameters are re-substituted into the dynamic mode switching mapping model to re-predict the vibration response under the same working conditions, and the consistency is verified with the current measured data. When the prediction error meets the preset convergence condition, the current model parameters are confirmed to be valid and used as the benchmark parameters for subsequent dynamic mode switching prediction and optimization calculations.

[0037] Furthermore, the adaptive adjustment of the stage weights and constraint parameters used in the multi-objective optimization process based on the actual vibration performance of each transient sub-stage during the switching process includes: After each transient sub-stage, based on the actual vibration amplitude, vibration energy growth rate, and decay rate within that stage, the stability evaluation index for that stage is calculated and compared with the target reference value set in the optimization model for that stage to obtain the stage stability deviation. The optimization model is used to describe the trade-off between vibration suppression effect and switching efficiency during power mode switching. It is established using power mode switching parameters as decision variables and the predicted drill string vibration response characteristics as evaluation criteria. The optimization model aims to minimize vibration amplitude, minimize vibration energy growth, and maximize switching stability, and uses the drill string allowable vibration threshold, power output change rate, and system stability constraints as constraints. Based on the magnitude and direction of the stability deviation of the stage, the risk contribution of the stage in the overall switching process is determined. When the deviation is greater than a preset threshold, the stage is determined to be the dominant risk stage in the current switching process. For transient sub-stages that are identified as the dominant risk stage, increase their weight coefficient in the multi-objective optimization function so that the vibration suppression objective corresponding to this stage is preferentially satisfied in subsequent optimizations. Meanwhile, based on the vibration amplitude growth rate and energy accumulation trend during this stage, the corresponding constraint parameters are dynamically tightened or relaxed. Specifically: when the vibration response shows a rapid amplification trend, the upper limit constraint of the vibration corresponding to this stage is tightened; when the vibration response is in a controllable decay state, the constraint is appropriately relaxed to improve the switching efficiency. The updated stage weights and constraint parameters are fed back into the multi-objective optimization model to resolve the subsequent power mode switching schemes. This allows the optimization results to gradually converge toward the optimal switching strategy under actual working conditions while meeting the overall stability requirements.

[0038] Example 2, Figure 2 The present invention provides an adaptive drilling control system based on multi-mode power switching, comprising the following modules: Potential switching risk identification module: used to collect sensor data during the drilling process, and through feature extraction and joint pattern recognition, to determine the potential switching risk range during the drilling process and to determine the candidate power mode switching trigger conditions; Switching Impact Prediction Module: Used to predict in real time the impact of transient switching between different power modes on drill string vibration based on the triggering conditions, and generate prediction results; The switching scheme optimization decision module is used to generate different power mode switching schemes based on the prediction results using a multi-objective weighted optimization algorithm, and select the comprehensive optimal switching scheme. Power mode switching module: Used to perform power mode switching based on the optimal switching scheme, continuously monitor drill string vibration and operating condition changes, and use process data to update the prediction model and optimization strategy to improve the performance of the next switching.

[0039] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0040] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0041] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0042] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0043] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0044] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An adaptive drilling control method based on multi-mode power switching, characterized in that, Includes the following steps: Collect sensor data during the drilling process, and determine the potential switching risk range and candidate power mode switching trigger conditions through feature extraction and joint pattern recognition. Based on the aforementioned triggering conditions, the impact of transient switching between different power modes on drill string vibration is predicted in real time, and prediction results are generated. Based on the prediction results, a multi-objective weighted optimization algorithm is used to generate different power mode switching schemes, and the comprehensive optimal switching scheme is selected. The power mode switching is performed based on the optimal switching scheme, and the drill string vibration and working condition changes are continuously monitored. The process data is used to update the prediction model and optimization strategy to improve the performance of the next switching.

2. The adaptive drilling control method based on multi-mode power switching according to claim 1, characterized in that, The method of determining potential switching risk intervals during drilling through feature extraction and joint pattern recognition includes: Based on the multi-source operating parameters obtained during the drilling process, a time-series feature vector reflecting the relationship between the changes of each parameter over time is constructed. Perform time-series correlation analysis on the time-series feature vectors to extract the relative change trends and phase shift relationships between parameters; Based on the aforementioned trends and offset relationships, identify whether there are characteristic patterns that characterize the evolution of the drilling system from a stable state to an unstable state; When the aforementioned characteristic pattern is detected, the corresponding time interval is determined as a potential power mode switching risk interval.

3. The adaptive drilling control method based on multi-mode power switching according to claim 2, characterized in that, The identification of whether there are characteristic patterns representing the evolution of the drilling system from a stable state to an unstable state based on the changing trend and offset relationship includes: The rate of change and the phase difference are combined to construct a parameter co-evolution relationship matrix; Feature extraction is performed on the co-evolutionary relationship matrix to obtain co-instability feature quantities used to characterize system stability; Determine whether the drilling system has evolved from a stable state to an unstable state based on the cooperative instability characteristic quantities; When the unstable evolutionary characteristics persist over time, the corresponding time interval is identified as a potential risk interval for dynamic mode switching.

4. The adaptive drilling control method based on multi-mode power switching according to claim 3, characterized in that, The conditions for determining the candidate power mode switching trigger include: Within the potential switching risk range, a dynamic state sequence characterizing the state evolution of the drilling process is constructed based on the operating parameters; The rate of state change and the rate of energy accumulation are calculated on the dynamic state trajectory to obtain an instability approach index that characterizes the degree to which the system approaches an unstable state. Based on the instability approach index and combined with the minimum response time required for dynamic mode switching, the time interval for the system state to evolve from stable to unstable is determined. The time interval is used as a candidate trigger interval for power mode switching, and the corresponding power mode switching trigger condition is generated accordingly.

5. The adaptive drilling control method based on multi-mode power switching according to claim 4, characterized in that, The method of real-time prediction of the impact of transient switching between different power modes on drill string vibration based on the triggering conditions, and generating prediction results, includes: Obtain drilling pressure, rotation speed, torque and vibration signals within a preset time window before the switching trigger time, and construct a state reference vector characterizing the current drill string dynamics based on the signals; Based on the triggering conditions, the corresponding power mode switching method is determined, and the power output change characteristics during the switching process are parameterized to form a switching feature vector. By using the state reference vector and the switching input vector as joint inputs, a mapping model is constructed to characterize the impact of dynamic mode switching transients on drill string vibration. Based on the prediction model, the predicted results of drill string vibration response under the corresponding power mode switching conditions are output.

6. The adaptive drilling control method based on multi-mode power switching according to claim 5, characterized in that, The step of determining the corresponding power mode switching method based on the triggering condition includes: During the system initialization phase, a set of power modes containing multiple power output control methods is pre-built to describe the selectable power mode types during drilling. After detecting the power mode switching trigger condition, the current drilling state is judged based on the instability approach index, and the power mode switching type corresponding to the current instability characteristics is determined. Based on the switching type, the key control parameters in the power mode switching process are quantitatively set to form a set of switching parameters to describe the power mode change process. The switching parameter set is uniformly parameterized and encoded to generate a switching input vector that characterizes the transient characteristics of the power mode switching.

7. The adaptive drilling control method based on multi-mode power switching according to claim 6, characterized in that, The generation of different power mode switching schemes based on the prediction results using a multi-objective weighted optimization algorithm includes: Based on the prediction results, the temporal variation characteristics of the drill string during the power mode switching process are analyzed, and the variation law of each vibration response index during the switching process is determined. Based on the changing patterns, the power mode switching process is divided into several transient stages, and the dominant range of each stage is determined. For each transient stage, corresponding evaluation indicators are constructed, and weights are assigned according to their impact on system stability to form a multi-objective weighted evaluation function; Based on the set of stage evaluation indicators, and considering the differences in the degree of influence of each transient sub-stage on system stability, corresponding weight coefficients are assigned to different evaluation indicators to construct a multi-objective weighted evaluation function. Under the constraints of the multi-objective weighted evaluation function, different power modes and their switching parameters are jointly optimized to obtain a set of power mode switching schemes.

8. The adaptive drilling control method based on multi-mode power switching according to claim 7, characterized in that, Based on the changing patterns, the power mode switching process is divided into several transient phases, and the dominant influence range of each phase is determined, including: Based on the vibration response results output by the mapping model, the vibration amplitude, energy change rate and vibration attenuation characteristics that characterize the dynamic state of the drill string are extracted as vibration response indicators. The temporal changes of various vibration response indicators during the dynamic mode switching process are analyzed, and the intensity of their changes and the degree of cumulative impact in different time intervals are calculated. By comparing the intensity of change of each vibration response index within the same time interval, the vibration response index with the greatest impact on drill string stability is determined. When a certain vibration response index remains dominant over a continuous time interval, the corresponding time interval is defined as the dominant range of that index.

9. The adaptive drilling control method based on multi-mode power switching according to claim 8, characterized in that, The process of switching power modes based on the optimal switching scheme and continuously monitoring drill string vibration and changes in operating conditions includes: The power output is adjusted according to the optimal power mode switching scheme, and process data is collected during the switching process; The process data is compared and analyzed with the corresponding switching prediction results to calculate the deviation between the actual vibration response and the predicted vibration response. Based on the deviation results, the parameters of the mapping model used to describe the relationship between dynamic mode switching and vibration response are corrected so that the model output gradually approaches the actual response characteristics. Based on the actual vibration performance at each stage during the switching process, the weights and constraint parameters of each stage in the multi-objective optimization are adaptively adjusted to ensure that the optimization results are consistent with the actual stability requirements. The updated model parameters and optimized parameters are used as inputs for subsequent power mode switching prediction and decision-making, thereby achieving continuous adaptive optimization of the power mode switching strategy.

10. A system using the adaptive drilling control method based on multi-mode power switching as described in any one of claims 1-9, characterized in that, Includes the following modules: Potential switching risk identification module: used to collect sensor data during the drilling process, and through feature extraction and joint pattern recognition, to determine the potential switching risk range during the drilling process and to determine the candidate power mode switching trigger conditions; Switching Impact Prediction Module: Used to predict in real time the impact of transient switching between different power modes on drill string vibration based on the triggering conditions, and generate prediction results; The switching scheme optimization decision module is used to generate different power mode switching schemes based on the prediction results using a multi-objective weighted optimization algorithm, and select the comprehensive optimal switching scheme. Power mode switching module: Used to perform power mode switching based on the optimal switching scheme, continuously monitor drill string vibration and operating condition changes, and use process data to update the prediction model and optimization strategy to improve the performance of the next switching.

Images