A Drilling Parameter Recommendation Method and Apparatus Based on AI Models

Abstract

This invention proposes an AI-based drilling parameter recommendation method and apparatus, relating to the field of petroleum engineering technology. The method includes: cleaning raw data; inputting the cleaned data into a parameter recommendation subsystem; calculating drilling performance metrics based on a physical model, aiming to optimize these metrics, and constructing a functional relationship between drilling parameters and the performance metrics through dynamic learning; and recommending optimized drilling parameters to the driller based on this functional relationship. This invention applies an AI model to drilling parameter recommendation, enabling it to identify different operating conditions and clean the raw data. It employs a combination of linear exploration and nonlinear application to recommend drilling parameters that optimize the target metrics under different conditions. This invention overcomes the shortcomings of traditional drilling parameter recommendation methods, such as cumbersome processes, long processing times, and poor accuracy. It can dynamically and efficiently recommend optimal drilling parameters, optimize drilling speed, and thus achieve cost reduction and efficiency improvement in the drilling process.

Description

Technical Field

[0001] This invention relates to the field of petroleum engineering technology, and more particularly to a method and apparatus for recommending drilling parameters based on an AI model. Background Technology

[0002] This section is intended to provide background or context for the embodiments of the invention set forth in the claims. The description herein is not an admission that it is prior art simply because it is included in this section.

[0003] Drilling parameter optimization and recommendation is a crucial part of drilling engineering, directly impacting safe drilling, high-quality production, and efficient utilization. How to quickly and accurately recommend drilling parameters conducive to achieving target indicators has become one of the key research directions in modern drilling control optimization technology. However, the complexity and diversity of drilling parameters lead to problems such as large-scale, high-risk, and high-investment drilling projects. Methods based on empirical models to recommend drilling parameters are difficult to meet drilling construction objectives, severely impacting drilling efficiency.

[0004] In summary, there is an urgent need for a technical solution that can overcome the above-mentioned shortcomings and enable the analysis, optimization, and recommendation of drilling parameters. Summary of the Invention

[0005] To address the problems existing in the prior art, this invention proposes a drilling parameter recommendation method and apparatus based on an AI model.

[0006] In a first aspect of this invention, a drilling parameter recommendation method based on an AI model is proposed, comprising:

[0007] Obtain the raw data and clean it.

[0008] The cleaned data is input into the parameter recommendation subsystem; among which, drilling performance metrics are calculated based on the physical model, with the goal of optimizing the drilling performance metrics, and the functional relationship between drilling parameters and drilling performance metrics is constructed through dynamic learning;

[0009] Based on the aforementioned functional relationship, optimized drilling parameters are recommended to the driller.

[0010] In a second aspect of the present invention, a drilling parameter recommendation device based on an AI model is proposed, comprising:

[0011] A data preprocessing subsystem is used to acquire raw data and clean the raw data.

[0012] The parameter recommendation subsystem is used to calculate drilling performance metrics based on a physical model, with the goal of optimizing these metrics. It constructs a functional relationship between drilling parameters and drilling performance metrics through dynamic learning, and recommends optimized drilling parameters to the driller based on this functional relationship.

[0013] In a third aspect of the present invention, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement a drilling parameter recommendation method based on an AI model.

[0014] In a fourth aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing a computer program that, when executed by a processor, implements a drilling parameter recommendation method based on an AI model.

[0015] In a fifth aspect of the present invention, a computer program product is provided, the computer program product comprising a computer program that, when executed by a processor, implements a drilling parameter recommendation method based on an AI model.

[0016] This invention discloses a drilling parameter recommendation method and apparatus based on an AI model. It collects raw well site data and performs corresponding preprocessing before providing it to a parameter recommendation system. The parameter recommendation subsystem receives data from the data processing subsystem, calculates various metrics on the data, learns and constructs a functional relationship between drilling parameters and target metrics based on the AI ​​model, and recommends drilling operation parameters that can improve the target metrics based on this functional relationship. This invention applies an AI model to drilling parameter recommendation, enabling it to identify different operating conditions and clean the raw data. It employs a combination of linear exploration and nonlinear application to recommend drilling parameters that optimize the target metrics under different modes. This invention overcomes the shortcomings of traditional drilling parameter recommendation methods, such as cumbersome processes, long processing times, and poor accuracy. It can dynamically and efficiently recommend optimal drilling parameters and optimize drilling speed, thereby achieving the goal of reducing costs and increasing efficiency in the drilling process. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a schematic diagram of a drilling parameter recommendation method based on an AI model according to an embodiment of the present invention.

[0019] Figure 2 This is a schematic diagram of the process of cleaning raw data according to an embodiment of the present invention.

[0020] Figure 3 This is a schematic diagram of a data preprocessing process based on working conditions according to an embodiment of the present invention.

[0021] Figure 4 This is a schematic diagram of the process of cleaning data using the expert threshold method according to an embodiment of the present invention.

[0022] Figure 5 This is a schematic diagram of the data cleaning process using the speed constraint method according to an embodiment of the present invention.

[0023] Figure 6 This is a flowchart illustrating the parameter recommendation process of a parameter recommendation subsystem according to an embodiment of the present invention.

[0024] Figure 7 This is a schematic diagram of a process for optimizing drilling performance according to an embodiment of the present invention.

[0025] Figure 8 This is a schematic diagram illustrating the principle of the three stages of exploration, learning, and application of the parameter recommendation subsystem according to an embodiment of the present invention.

[0026] Figure 9 This is a schematic diagram of the mode switching process of a parameter recommendation subsystem according to an embodiment of the present invention.

[0027] Figure 10 This is a schematic diagram of the velocity value distribution according to an embodiment of the present invention.

[0028] Figure 11 This is a schematic diagram of a support vector regression model according to an embodiment of the present invention.

[0029] Figure 12 This is a schematic diagram of the adjustment (reduction) of exploration mode parameters according to an embodiment of the present invention.

[0030] Figure 13 This is a schematic diagram of the adjustment (increase) of the exploration mode parameters according to an embodiment of the present invention.

[0031] Figure 14 This is a schematic diagram of a random forest algorithm according to an embodiment of the present invention.

[0032] Figure 15 This is a schematic diagram of the architecture of a drilling parameter recommendation device based on an AI model according to an embodiment of the present invention.

[0033] Figure 16 This is a schematic diagram of a computer device structure according to an embodiment of the present invention. Detailed Implementation

[0034] The principles and spirit of the invention will now be described with reference to several exemplary embodiments. It should be understood that these embodiments are given merely to enable those skilled in the art to better understand and implement the invention, and are not intended to limit the scope of the invention in any way. Rather, these embodiments are provided to make this disclosure more thorough and complete, and to fully convey the scope of this disclosure to those skilled in the art.

[0035] Those skilled in the art will recognize that embodiments of the present invention can be implemented as a system, apparatus, device, method, or computer program product. Therefore, this disclosure can be specifically implemented in the following forms: entirely hardware, entirely software (including firmware, resident software, microcode, etc.), or a combination of hardware and software.

[0036] According to an embodiment of the present invention, a drilling parameter recommendation method and apparatus based on an AI model are proposed, relating to the field of petroleum engineering technology. The present invention cleans the raw data using an expert threshold method and a velocity constraint method through a data preprocessing subsystem based on the operating conditions. The processed data is then transmitted to a parameter recommendation subsystem (ROP-OPT-AI), which calculates drilling performance metrics based on a physical model, dynamically learns and constructs a functional relationship between drilling parameters and drilling performance improvement targets, and recommends optimal drilling parameters to the driller based on this functional relationship.

[0037] The principles and spirit of the present invention will be explained in detail below with reference to several representative embodiments.

[0038] Figure 1 This is a schematic diagram of a drilling parameter recommendation method based on an AI model according to an embodiment of the present invention. Figure 1 As shown, the method includes:

[0039] Step 101: Obtain the raw data and clean the raw data;

[0040] Step 102: Input the cleaned data into the parameter recommendation subsystem; wherein, based on the physical model, the drilling performance metrics are calculated, and with the goal of optimizing the drilling performance metrics, a functional relationship between drilling parameters and drilling performance metrics is constructed through dynamic learning.

[0041] Step 103: Recommend optimized drilling parameters to the driller based on the aforementioned functional relationship.

[0042] In practical applications, the drilling parameter recommendation method based on an AI model proposed in this invention works as follows: The data preprocessing subsystem uses an independent thread to obtain raw data via HTTP protocol and performs corresponding data preprocessing according to the working conditions. The data is cleaned using expert thresholding and velocity constraint methods, and the rate of penetration (ROP) is calculated according to well depth intervals. The processed micro-level well section data is saved to the raw data buffer, and when the data accumulates to a certain extent, it is sent to the parameter recommendation subsystem.

[0043] The parameter recommendation subsystem calculates drilling performance metrics such as mechanical specific energy and optimization targets based on data. The mechanical specific energy calculation process is divided into three cases according to the working conditions: composite drilling with screw, composite drilling without screw, and sliding drilling. The construction of optimization targets considers both maximizing mechanical drilling speed and minimizing mechanical specific energy.

[0044] With the goal of improving drilling performance metrics, a combined approach of linear exploration and nonlinear application is used for drilling parameter recommendation. This approach not only reveals data trends but also accurately fits historical data distributions. After startup, the parameter recommendation subsystem enters exploration mode, exploring and collecting data in the direction of increasing ROP based on the trend fitted by the linear model. After sufficient exploration, it enters learning mode, learning the functional relationship between ROP and drilling parameters in the current exploration data based on the model, and providing the optimal combination of drilling parameters. Subsequently, maintaining the optimal drilling parameters, it enters application mode, efficiently and steadily breaking rock while continuously monitoring the current drilling status. Based on the degree of ROP performance degradation, it automatically switches between learning and exploration modes to adapt to the current drilling conditions.

[0045] The following section provides a detailed explanation of the AI ​​model-based drilling parameter recommendation method proposed in this invention, based on the aforementioned working principle.

[0046] Step 101, Data Cleaning:

[0047] refer to Figure 2 In step 101, the raw data is obtained, and the specific process for cleaning the raw data is as follows:

[0048] Step 1011: The raw data is obtained via HTTP protocol using an independent thread through the data preprocessing subsystem.

[0049] Step 1012: Perform data preprocessing according to the working conditions, clean the data using the expert threshold method and the velocity constraint method, calculate the mechanical drilling rate according to the well depth interval, and save the processed micro-element well section data to the original data buffer.

[0050] Step 1013: Once the data has accumulated to a certain extent, it is sent to the parameter recommendation subsystem.

[0051] The following section provides a detailed explanation of each processing step in step 1012.

[0052] In one embodiment, the detailed process of data preprocessing based on operating conditions is as follows: Figure 3 As shown.

[0053] Pre-processing of operating conditions:

[0054] Based on the current operating conditions of the received data and the switching status between previous and subsequent operating conditions, the raw data is preprocessed accordingly.

[0055] In the preprocessing stage, the drilling status is first determined based on the well site data; if it is a drilling status, the sliding drilling status is further determined; if it is not a drilling status, the status parameters are updated and the data buffer and mechanical drilling rate calculation queue are cleared.

[0056] When determining the sliding drilling state, if it is a sliding drilling state, the screw speed and screw torque are recalculated and the state parameters are updated; if it is not a sliding drilling state, no action is taken.

[0057] In one embodiment, the detailed process of cleaning data using the expert threshold method and the velocity constraint method is as follows: Figure 4 and Figure 5 As shown.

[0058] Expert threshold method:

[0059] Abnormal values ​​in the data are monitored by setting expert thresholds, wherein the expert thresholds include at least: drilling pressure range, composite drilling torque range, sliding drilling torque range, composite drilling ground speed range, sliding drilling screw speed range, inlet flow range, riser pressure range, and drilling speed range.

[0060] Velocity constraint method:

[0061] Outliers are cleaned using the expert threshold method, and the remaining data are further evaluated using the velocity constraint method to clean up data that does not conform to the upper and lower bounds of the velocity value.

[0062] In one embodiment, the specific process for calculating the mechanical drilling rate according to the well depth interval is as follows:

[0063] The mechanical drilling rate is determined based on well depth intervals, and a queue is maintained according to these intervals. The window size of the queue dynamically changes, and each time a new mechanical drilling rate is calculated, the calculated data is added to the end of the queue. The mechanical drilling rate is calculated as follows:

[0064]

[0065] Among them, ROP now This refers to the newly added tail data; Depth1, Depth nowThese are the first and last data points in the well depth data queue, respectively; Time1, Time... now These are the first and last data points in the time data queue, respectively.

[0066] In one embodiment, the specific process for saving the processed micro-element well section data to the original data buffer is as follows:

[0067] The processed data is stored in a dynamic buffer of micro-well segments. When the interval between the well depth of the last data in the dynamic buffer of micro-well segments and the well depth of the first data is greater than or equal to the size of the buffer, all data in the dynamic buffer of micro-well segments is sent to the parameter recommendation subsystem. After the sending is completed, the dynamic buffer of micro-well segments is cleared and the data is re-accumulated.

[0068] refer to Figure 6 This is a flowchart illustrating the parameter recommendation process of a parameter recommendation subsystem according to an embodiment of the present invention. The parameter recommendation subsystem (ROP-OPT-AI) acquires micro-level well segment data from the data preprocessing subsystem, filters and merges it into a single data point, calculates metrics characterizing drilling performance (Mechanical Energy Specific Energy (MSE), Optimization Objective) based on a predefined physical model, and adds the calculation results to a predefined data structure. As data collection reaches a certain level, the algorithm dynamically learns the functional relationship between drilling operation parameters (such as WOB, RPM, Qpdm) and the optimization objectives and ROP. Based on this functional relationship, it recommends better drilling operation parameters to the driller, gradually improving drilling efficiency.

[0069] Step 102, Optimize drilling performance:

[0070] refer to Figure 7 In step 102, the cleaned data is input into the parameter recommendation subsystem; wherein, drilling performance metrics are calculated based on the physical model, with the goal of optimizing the drilling performance metrics, and a functional relationship between drilling parameters and drilling performance metrics is constructed through dynamic learning. The specific process is as follows:

[0071] Step 1021: Calculate drilling performance metrics based on the cleaned data, wherein the drilling performance metrics include at least: mechanical specific energy and optimization target;

[0072] Step 1022: Set the goal of improving drilling performance metrics, and adopt a combination of linear exploration and nonlinear application to recommend drilling parameters. After the parameter recommendation subsystem is started, it enters exploration mode, exploring and collecting data in the direction of increasing mechanical drilling rate based on the trend of linear model fitting. After sufficient exploration, it enters learning mode, learning the functional relationship between mechanical drilling rate and drilling parameters in the current exploration data based on the model, and determining the optimal combination of drilling parameters. Maintaining the optimal drilling parameters, it enters application mode, monitoring the current drilling status while breaking rock, and automatically switching between learning mode and exploration mode based on the degree of performance degradation of mechanical drilling rate.

[0073] In one embodiment, the method further includes:

[0074] The time and well depth of the last data point in the micro-element well segment data are extracted and used as the time and well depth of the merged data. The other data are averaged using Kalman filtering to obtain the merged data. The merged data is then input into the parameter recommendation subsystem.

[0075] In one embodiment, drilling performance metrics are calculated based on the cleaned data, including:

[0076] The formula for calculating mechanical specific energy is:

[0077]

[0078] Where MSE is mechanical specific energy; EFF s Efficiency coefficient; WOB is drilling pressure; T is torque; RPM is rotational speed; ROP is mechanical drilling speed; d B The drill bit diameter is the mechanical energy; mechanical energy is the work done by the drill bit to break a unit volume of rock, used to evaluate drilling efficiency. A higher mechanical energy value indicates lower rock breaking efficiency, poor adaptability of the drill bit to the formation, and the need to optimize drilling parameters.

[0079] The formula for calculating the optimization objective is as follows:

[0080]

[0081] Wherein, Objective is the optimization objective, which is to maximize the mechanical drilling rate while minimizing the mechanical specific energy; ROP0 and MSE0 are normalization quantities used to remove the influence of dimensions.

[0082] In one embodiment, the modes of the parameter recommendation subsystem include: exploration mode, learning mode, and application mode. (See reference) Figure 8 This is a schematic diagram illustrating the principle of the three stages of exploration, learning, and application of the parameter recommendation subsystem according to an embodiment of the present invention.

[0083] In one embodiment, reference Figure 9The mode switching process of the parameter recommendation subsystem is as follows:

[0084] Determine whether the parameter recommendation subsystem is in exploration mode; if it is in exploration mode, further determine whether the exploration is sufficient; if it is not in exploration mode, evaluate the drilling performance.

[0085] When determining whether the exploration is sufficient, if the exploration is sufficient, switch to learning mode and use a nonlinear model to learn the optimal parameters; if the exploration is insufficient, switch to exploration mode and use a linear model to explore the space.

[0086] When evaluating drilling performance metrics, determine whether drilling performance has deteriorated; if no deterioration has occurred, switch to application mode and maintain the current parameters for rock breaking; if deterioration has occurred, determine whether the deterioration index has reached the deterioration threshold.

[0087] When determining whether the indicators of decline have reached the decline threshold, if they have not, switch to learning mode and use a nonlinear model to learn the optimal parameters; if they have, switch to exploration mode and use a linear model to explore the space.

[0088] In one embodiment, the mode switching method in exploration mode includes:

[0089] The exploration mode of the parameter recommendation subsystem is used to guide the driller to explore in the direction of increasing mechanical drilling speed. By exploring the parameter space, the correspondence between operating parameters and mechanical drilling speed is obtained, and the area with the maximum mechanical drilling speed is found.

[0090] When exploration is sufficient, switch from exploration mode to learning mode. The switching condition is to determine whether a predetermined number of explorations has been completed. The minimum number of explorations and the maximum number of explorations are set. When the minimum number of explorations is reached, the current drilling speed is checked before each exploration. If the drilling speed exceeds the drilling speed threshold, exploration continues; otherwise, switch to learning mode. When the maximum number of explorations is exceeded, the exploration ends.

[0091] In one embodiment, the mode switching method in learning mode or application mode includes:

[0092] After the parameter recommendation subsystem switches to learning mode or application mode, it enters the dynamic monitoring of drilling performance stage; if the drilling performance reaches the performance threshold, the current operating parameters are maintained and the system enters application mode; if the drilling performance deteriorates but does not reach the deterioration threshold, the drilling parameters are adjusted and the system enters learning mode; if the drilling performance deteriorates and reaches the deterioration threshold, the system switches to the exploration stage.

[0093] The mechanical drilling speed score is used to determine the current drilling performance. Two threshold values ​​are set to evaluate the current drilling effect. If the mechanical drilling speed score falls below the first threshold, the drilling performance is considered to have severely deteriorated. If the mechanical drilling speed score falls above the second threshold, the drilling performance is considered to be stable. If the mechanical drilling speed score falls between the first and second thresholds, the drilling performance is considered to have slightly deteriorated. The mechanical drilling speed score is calculated as follows:

[0094]

[0095] ROP score ROP (Recovery Points) is the score for mechanical drilling speed. now This represents the current mechanical drilling speed; ROP history The historical mechanical drilling rate is used; the parameter recommendation subsystem calculates the mechanical drilling rate score for the current anchor point and the historical anchor point every 10 micro-element well sections of data received.

[0096] To improve the robustness of the judgment results, the method also includes:

[0097] Calculate the mechanical drilling rate score for multiple consecutive tests;

[0098] Based on the mechanical drilling speed score threshold, multiple mechanical drilling speed scores are divided into different intervals, and the interval with the most mechanical drilling speed scores is selected as the judgment result for this time.

[0099] Step 103, Recommended drilling parameters:

[0100] In step 103, optimized drilling parameters are recommended to the driller based on the functional relationship, including parameter recommendations under different working conditions and parameter recommendations under different modes.

[0101] 1) Recommended parameters for different operating conditions:

[0102] When recommending parameters under different operating conditions, for composite drilling, the functional relationship between drilling operation parameters and drilling performance improvement targets is as follows:

[0103] ROP = f(WOB, RPM) surface (Qpdm)

[0104] OBJ = f(WOB, RPM) surface (Qpdm)

[0105] ROP is the mechanical drilling rate; OBJ is the optimization target; WOB is the drilling pressure; RPM surface Where is the ground rotational speed; Qpdm is the displacement; recommended parameters are given based on the aforementioned functional relationship;

[0106] For sliding drilling, the functional relationship between drilling operation parameters and drilling performance improvement targets is as follows:

[0107] ROP = f(WOB, Qpdm)

[0108] OBJ = f(WOB, Qpdm)

[0109] Recommended parameters are given based on the aforementioned functional relationship.

[0110] 2) Recommended parameters for exploration mode:

[0111] When recommending parameters in the exploration mode, a support vector regression model is used to fit the relationship between drilling operation parameters, mechanical drilling rate and optimization target, and the recommended parameters are determined based on the fitted relationship.

[0112] For composite drilling or sliding drilling, a support vector regression model is used to fit the linear relationship of the target calculation method. Based on the slope obtained by parameter fitting, and using the boundary values ​​of the explored parameter interval as a basis, the recommended parameters are adjusted in the direction of increasing mechanical drilling rate according to the slope change. If the adjusted recommended parameters exceed the values ​​specified in the drilling design, the drilling design is used as the recommended parameters.

[0113] 3) Recommended learning mode parameters:

[0114] When recommending parameters in the learning mode, a random forest regression model is used for nonlinear fitting to fit the relationship between drilling operation parameters, mechanical drilling rate and optimization objective, and the recommended parameters are determined based on the fitted relationship.

[0115] For composite drilling or sliding drilling, a random forest regression model is used to fit the linear relationship of the optimization target calculation method. After fitting, a grid search is performed in the parameter space based on the random forest regression model to obtain the parameters corresponding to the maximum point of the mechanical drilling rate or optimization target. It is then determined whether the parameters meet the conditions, and if they do, recommendations are made.

[0116] 4) Recommended application mode parameters:

[0117] When recommending parameters in the application mode, since the drilling performance of this mode is relatively stable, the previously recommended parameters are used to keep the system in a steady state.

[0118] To provide a clearer explanation of the above-mentioned AI model-based drilling parameter recommendation method, a specific embodiment will be used for detailed explanation below.

[0119] The drilling parameter recommendation method based on AI model proposed in this invention is mainly implemented through a data preprocessing subsystem and a parameter recommendation subsystem.

[0120] 1. Data Preprocessing Subsystem

[0121] The data preprocessing subsystem uses an independent thread to continuously obtain raw data from the well site equipment via the HTTP protocol, preprocesses it according to the working conditions, cleans the data to remove outliers, calculates the ROP, and puts the processed data into the raw data buffer. When the data in the buffer accumulates to a certain level, it is sent to the parameter recommendation system.

[0122] 1.1 Pre-processing of operating conditions

[0123] Working condition preprocessing involves preprocessing the raw data according to the current working condition (drilling, non-drilling, sliding drilling, and combined drilling) and the switching between previous and subsequent working conditions (drilling to non-drilling, non-drilling to drilling). (Reference) Figure 3 The diagram shows the preprocessing flow for the working conditions. This flow includes: determining the drilling state, determining the sliding drilling state, and calculating the screw speed and screw torque under sliding drilling.

[0124] Upon receiving a data message, the system first determines if it is drilling data. If not, it indicates a non-drilling state, skips processing the current data, and clears the data buffer and ROP calculation queue. If it is drilling data, it first determines whether it is in a sliding drilling state based on observed values ​​of screw speed and torque. If it is in a sliding drilling state, the screw speed and torque are recalculated as the actual speed and torque for the current data, replacing the original values. If it is in a combined drilling state, no processing is performed.

[0125] 1.1.1 Drilling Status Judgment

[0126] If the received data simultaneously meets the two conditions that the absolute values ​​of well depth and drill bit position differ by 0.05 meters and the drilling pressure is greater than or equal to 10 kN, it indicates that the current state is drilling; otherwise, it is non-drilling.

[0127] 1.1.2 Determine if it is in a sliding drilling state

[0128] If the current data shows that the rotary table speed is less than 5 revolutions or the torque is less than 0.8 kN·m, then it is in the sliding drilling state.

[0129] 1.1.3 Calculate the screw speed (RPM) and screw torque (Torque) during sliding drilling.

[0130] In sliding drilling mode, the screw speed and screw torque need to be recalculated. Screw speed (RPM) motor The calculation formula is as follows:

[0131]

[0132] Where Qpdm represents displacement and Kn is the speed-to-flow ratio (Qpdm / RPM).

[0133] Screw torque Torque motor The calculation formula is as follows:

[0134]

[0135] Where SPP represents riser pressure, CSPP represents circulating riser pressure, and T max P is the maximum rated torque of the screw. max This is the maximum rated differential pressure.

[0136] 1.2 Data Cleaning

[0137] To mitigate the impact of data anomalies, the received data undergoes cleaning to remove obvious outliers. An expert threshold algorithm combined with speed constraints is used to monitor for outliers. Data is first processed by the expert threshold cleaning algorithm; if it is found to be an outlier, the speed constraint cleaning algorithm is skipped and the data is output directly. Otherwise, the data proceeds to the speed constraint cleaning algorithm for further evaluation. (Refer to...) Figure 4 This is a flowchart illustrating the data cleaning process.

[0138] 1.2.1 Expert Threshold Cleaning Algorithm

[0139] Anomalies in the data are monitored by setting expert thresholds. Each expert threshold can be set as follows:

[0140] Drilling pressure (WOB) range: 0~300kN;

[0141] Combined drilling torque range: 0~50kN·m;

[0142] Sliding drilling torque range: 0~30kN·m;

[0143] Composite drilling surface rotation speed (RPM) range: 0~150r;

[0144] Sliding drill screw rotation speed range: 0~300r;

[0145] Inlet flow rate range: 0~100L / s;

[0146] Riser pressure (SPP) range: 0~100MPa;

[0147] Drilling rate (ROP) range: 0–60 m / hr.

[0148] 1.2.2 Velocity Constraint Cleaning Algorithm

[0149] After being cleaned using an expert thresholding algorithm, the data will be further evaluated using a velocity constraint algorithm. (Reference) Figure 5 This is a flowchart illustrating the velocity constraint algorithm. Let the data point received at the current time t be x, then the corresponding velocity value Δx is: Δx = (x... t -x t-k ) / k, where k is the step size between the two data points. Data analysis shows that the velocity values ​​follow a Gaussian distribution. Figure 10 This diagram illustrates the velocity distribution. The dashed line on the left represents the upper bound of the selected velocity, and the dashed line on the right represents the lower bound. The solid black line represents a Gaussian distribution. To select the upper and lower bounds of the velocity values ​​based on the distribution, the mean and standard deviation of the data within the window are calculated, and then combined with the n-sigma criterion to determine the upper and lower bounds of the velocity values. The mean (μ) of the window data... t ) and standard deviation (σ) t The calculation formula is as follows:

[0150]

[0151]

[0152] Based on experience and data testing, n=5 is chosen, and the calculated upper bound of the velocity constraint Δx is obtained. ub and lower bound Δx lb :

[0153] Δx ub =μ t +n×σ t (5)

[0154] Δx lb =μ t -n×σ t (6)

[0155] The window size W affects the cleaning effect. When the amount of data is too small, the Gaussian distribution fitting effect is not good, resulting in poor cleaning effect. Therefore, it is necessary to set the window size in combination with the well site data. It is preferable that W > 9000 (considering that when the time step is 1s, at least 2.5H of velocity data is used for fitting).

[0156] 1.3 Calculate ROP

[0157] ROP is calculated based on well depth intervals, and a queue is maintained according to the well depth intervals. The window size of the queue changes dynamically (the queue window size is given by the dynamic buffer of the micro-element well segment). Each time the ROP of a new data point is calculated, the data is added to the end of the queue.

[0158] In the ROP calculation queue, well depth data can be represented as [Depth1, ..., Depth]. nowTime data can be represented as [Time1, ..., Time] now If ], then the ROP of the newly added tail data is:

[0159]

[0160] ΔDepth and ΔTime represent the well depth difference and time difference, respectively.

[0161] 1.4 Design of dynamic buffer zones for micro-element well sections

[0162] The dynamic buffer of the micro-element well section stores the processed data and defines the frequency at which the parameter recommendation system acquires data for calculation.

[0163] The size of the buffer is accumulated according to the well depth interval and changes dynamically with ROP. When the interval between the well depth of the last data in the buffer and the well depth of the first data is greater than or equal to the size of the buffer, all the data in the buffer is provided to the recommendation algorithm, and then the data buffer is cleared and data is re-accumulated.

[0164] If the anchor point ROP is less than or equal to 6 m / s, the buffer size is set to 0.05 m; if the anchor point ROP is greater than 6 m / s and less than or equal to 10 m / s, the buffer size is 0.08 m; if the anchor point ROP is greater than 10 m / s, the buffer size is 0.1 m. The dynamic buffer size of the micro-element well section is expressed as follows:

[0165]

[0166] The buffer stores logically continuous drilling data. When a change in operating conditions occurs, the data in the data buffer needs to be cleared, including the switch between drilling and non-drilling, and between sliding drilling and combined drilling.

[0167] 2. Parameter Recommendation Subsystem (ROP-OPT-AI)

[0168] refer to Figure 6 This is a flowchart illustrating the parameter recommendation process of the parameter recommendation subsystem. The subsystem acquires micro-level well segment data from the data preprocessing subsystem, filters and merges it into a single data point, calculates metrics characterizing drilling performance (Mechanical Energy Specific Energy (MSE), Optimization Objective) based on a predefined physical model, and adds the results to a predefined data structure. As data collection progresses, the algorithm dynamically learns the functional relationship between drilling operation parameters (such as WOB, RPM, Qpdm) and the optimization objectives Objective and ROP. Based on this relationship, it recommends better drilling operation parameters to the driller, gradually improving drilling efficiency.

[0169] 2.1 Data Processing of Micro-element Well Sections

[0170] After receiving data from the data preprocessing subsystem, the time, well depth, and drilling speed of the last data in the micro-element well segment are extracted as the time and well depth of the merged data. The other data are averaged using Kalman filtering. The merged data segment is the input for the recommendation algorithm.

[0171] 2.2 Calculation of Measurement Indicators

[0172] Mechanical specific energy (MSE) and optimization objective are calculated based on physical models and used as indicators of drilling performance.

[0173] Mechanical specific energy, or specific energy, is the work done by the drill bit to break a unit volume of rock (i.e., the mechanical energy required). It provides a method for evaluating drilling efficiency; a higher specific energy value indicates lower rock-breaking efficiency, poorer adaptability of the drill bit to the formation, and the need for optimization of drilling parameters. The model calculation formula is as follows:

[0174]

[0175] Among them, EFF s The efficiency coefficient is a constant, typically taken as EFF. s Equal to 35%; MSE is mechanical specific energy, MPa; WOB is drilling pressure, N; T is torque, Nm; RPM is rotational speed, rev / min; ROP is mechanical drilling rate, m / hr; d B The value is the drill bit diameter, in mm.

[0176] The calculation of mechanical specific energy (MSE) requires the total rotational speed and total torque. The calculation of total rotational speed is divided into the following three cases:

[0177] Composite drilling with screw: RPM = RPM surface +RPM motor

[0178] Composite drilling without screw: RPM = RPM surface

[0179] Sliding drilling: RPM = RPM motor = (Qpdm / Kn) × 60

[0180] Total torque can be divided into the following two cases according to operating conditions:

[0181] Composite drilling: Torque = Torque surface

[0182] Sliding drilling: Torque = Torque motor

[0183] The optimization objective is to maximize the rate of drilling while minimizing the specific energy of the machine. Normalized quantities ROP0 and MSE0 are defined to remove the influence of dimensions; currently, ROP0 = 60 and MSE0 = 20000 are chosen. The optimization objective is calculated using the following formula:

[0184]

[0185] 2.3 Mode Switching

[0186] The current system is divided into three modes: exploration, learning, and application. Mode switching includes switching between exploration mode and switching between learning and application modes.

[0187] 2.3.1 Mode Switching in Exploration Mode

[0188] The task of the exploration mode is to guide the driller to continuously explore in the direction of increasing ROP, fully exploring the parameter space to obtain as many correspondences between operating parameters and ROP as possible, and finding the area with the largest ROP. When exploration is sufficient, the system switches from exploration mode to learning mode, and the switching condition is whether a predetermined number of explorations has been completed. Considering the time requirements of actual drilling operations, the system adopts a dynamic exploration number based on drilling speed. When the drilling speed is high, more explorations are performed, and when the drilling speed is low, the number of explorations is reduced to reduce the impact on drilling operations. Specifically, the system sets a minimum number of explorations (5 times) and a maximum number of explorations (9 times). When the minimum number of explorations is reached, the current drilling speed is judged before each exploration. If the drilling speed is fast (judgment threshold is 6 m / h), exploration continues; otherwise, it switches to learning mode (a switching judgment is performed every 5 micro-element well sections of data).

[0189] 2.3.2 Mode switching in learning and application modes

[0190] After entering the learning and application mode, the system is in the dynamic monitoring stage of drilling performance. If the drilling performance is good, the current operating parameters are maintained and the system enters the application mode. If the drilling performance deteriorates slightly, the drilling parameters are fine-tuned and the system enters the learning mode to adapt to the changes in the current drilling status. If the drilling performance deteriorates severely, it indicates that a formation switch may be taking place or a complex downhole situation has occurred. The current model is no longer suitable for the new formation environment, and the system needs to be switched back to the exploration stage.

[0191] Use ROP score (ROP scoreThe system assesses current drilling performance by setting two ROP (Recovery Point of Performance) score thresholds. A score below the minimum threshold (0.7) indicates a severe decline in drilling performance, while a score above the maximum threshold (0.9) indicates stable performance. A score between the two thresholds indicates a minor decline in drilling performance. The system calculates the ROP score for each anchor point and historical anchor points received every 10 micro-element well segments. score The calculation formula is as follows:

[0192]

[0193] To make the evaluation more robust, a "three consecutive ROP scores" evaluation method is adopted. The ROP scores of the current ROP anchor point are calculated by comparing it with the three most recent ROP anchor points in history. A voting method is used, and the mode is entered according to the ROP score of the interval that falls into.

[0194] The voting details are shown in Table 1. If the ROP scores for the three intervals are 1:1:1, considering that the drilling status fluctuates greatly at this time, the system will maintain the application mode and wait for one round.

[0195] In practical applications, the specific number of calculations can be selected and adjusted according to the actual situation.

[0196] Table 1. ROP scores for three consecutive times

[0197]

[0198]

[0199] 2.4 Parameter Recommendations

[0200] 2.4.1 Parameter Recommendations for Different Operating Conditions

[0201] Composite drilling: The functional relationship between drilling operation parameters and drilling performance improvement targets, and recommended parameters based on this functional relationship:

[0202] ROP = f(WOB, RPM) surface (Qpdm)

[0203] OBJ = f(WOB, RPM) surface (12)

[0204] Where ROP is the mechanical drilling rate; OBJ is the optimization target; WOB is the drilling pressure; RPM surface Where is the ground rotational speed; Qpdm is the displacement; recommended parameters are given based on the aforementioned functional relationship;

[0205] Sliding drilling: The functional relationship between drilling operation parameters and drilling performance improvement targets, and recommended parameters based on this functional relationship:

[0206] ROP = f(WOB, Qpdm)

[0207] OBJ=f(WOB,Qpdm) (13)

[0208] 2.4.2 Recommended Parameters for Exploration Mode

[0209] A support vector regression (SVR) model was used to fit the relationship between drilling operation parameters and ROP and OBJ. Figure 11 This is a schematic diagram of the support vector regression model.

[0210] The fitting objective of the SVR model is to maximize the interval distance d (i.e., While minimizing the loss (i.e., The mathematical expression of the model is as follows:

[0211]

[0212]

[0213] Taking composite drilling (without considering rotational speed during sliding drilling) as an example, the linear relationship of formula (10) is fitted using the SVR model. Based on the slope obtained by fitting each parameter, the recommended parameters are adjusted in the direction of increasing ROP according to the boundary value of the explored parameter interval. The recommended parameter size should not exceed the value specified in the drilling design; otherwise, the drilling design will be used as the recommended parameter.

[0214] The explored parameter range boundary values ​​refer to the maximum and minimum anchor points of WOB, RPM, and Qpdm received during the exploration phase. Taking drilling pressure as an example, if the model requires a reduction in drilling pressure, the minimum historical drilling pressure anchor point during the exploration phase will be used as the benchmark value for drilling pressure adjustment; if the model requires an increase in drilling pressure, the maximum historical drilling pressure anchor point during the exploration phase will be used as the benchmark value for current drilling pressure adjustment. Figure 12 To explore the schematic diagram of adjusting (reducing) mode parameters, Figure 13 A schematic diagram for exploring the adjustment (increase) of mode parameters.

[0215] 2.4.3 Recommended Learning Mode Parameters

[0216] A random forest regression model was used for nonlinear fitting to fit the relationship between drilling operation parameters and ROP and OBJ. Figure 14 This is a schematic diagram of the Random Forest algorithm. Figure 14The letters in the model name represent Random Forest, Instance, Tree (Tree1, 2, 3), Value (Value A, B, C), Average, and FinalValue, respectively. The model trains multiple decision trees on the dataset, each capable of independently performing the fitting task. It then independently resamples the data multiple times and combines the results from these multiple decision trees to obtain the model's output. This makes the output less susceptible to noise and outliers, resulting in stronger robustness and generalization.

[0217] Taking composite drilling (sliding drilling without considering rotation speed) as an example, the nonlinear relationship of formula (10) is fitted using a random forest regression model. After fitting, a grid search is performed on the parameter space based on the model to obtain the parameters corresponding to the maximum point of OBJ (ROP). It is then determined whether the parameters meet the conditions. If they do, recommendations are made.

[0218] The method for determining the parameter ranges and step sizes in the grid search is as follows: Calculate the maximum and minimum values ​​of WOB, RPM, and Qpdm from historical data. The range for WOB (unit: kN) is [minimum - 3, maximum + 3], with an exploration step size of 1; the range for RPM (unit: r) is [minimum - 2, maximum + 2], with an exploration step size of 0.5; and the range for Qpdm (unit: L / s) is [minimum - 1, maximum + 1], with an exploration step size of 0.2. When recommending parameters, select the WOB, RPM, and Qpdm values ​​corresponding to the maximum value of OBJ (ROP) during the grid search process.

[0219] 2.4.4 Recommended Application Mode Parameters

[0220] When the system is in application mode, the drilling performance is relatively stable. At this time, the previously recommended parameters are kept unchanged to keep the system in a steady state.

[0221] The drilling parameter recommendation method based on an AI model proposed in this invention is implemented through a data preprocessing subsystem and a parameter recommendation subsystem. The data preprocessing subsystem is mainly connected to the drilling equipment, collects raw well site data, performs corresponding preprocessing, and provides it to the parameter recommendation subsystem. The parameter recommendation subsystem receives data from the data processing subsystem, calculates various metrics on the data, learns and constructs a functional relationship between drilling parameters and target metrics based on the AI ​​model, and recommends drilling operation parameters that can improve the target metrics based on this functional relationship. This invention applies the AI ​​model to drilling parameter recommendation, which can identify different working conditions and clean the raw data. It adopts a scheme combining linear exploration and nonlinear application to recommend drilling parameters that optimize the target metrics under different modes. This invention overcomes the shortcomings of traditional drilling parameter recommendation methods, such as cumbersome processes, long processing times, and poor accuracy. It can dynamically and efficiently recommend optimal drilling parameters and optimize drilling speed, thereby achieving the goal of reducing costs and increasing efficiency in the drilling process.

[0222] It should be noted that although the operation of the method of the present invention has been described in a specific order in the above embodiments and figures, this does not require or imply that the operations must be performed in that specific order, or that all the operations shown must be performed to achieve the desired result. Additionally or alternatively, certain steps may be omitted, multiple steps may be combined into one step, and / or one step may be broken down into multiple steps.

[0223] After introducing the method of exemplary embodiments of the present invention, the following references are made. Figure 15 An AI model-based drilling parameter recommendation device according to an exemplary embodiment of the present invention will be introduced.

[0224] The implementation of the drilling parameter recommendation device based on the AI ​​model can refer to the implementation of the method described above, and the repetitions will not be repeated. The term "module" or "unit" used below can refer to a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0225] Based on the same inventive concept, this invention also proposes a drilling parameter recommendation device based on an AI model, such as... Figure 15 As shown, the device includes:

[0226] A data preprocessing subsystem is used to acquire raw data and clean the raw data.

[0227] The parameter recommendation subsystem is used to calculate drilling performance metrics based on a physical model, with the goal of optimizing these metrics. It constructs a functional relationship between drilling parameters and drilling performance metrics through dynamic learning, and recommends optimized drilling parameters to the driller based on this functional relationship.

[0228] In one embodiment, the data preprocessing subsystem acquires raw data and cleans the raw data, including:

[0229] Use a separate thread to retrieve the raw data via the HTTP protocol;

[0230] Data preprocessing is performed according to the working conditions. The data is cleaned using the expert threshold method and the velocity constraint method. The mechanical drilling rate is calculated according to the well depth interval. The processed micro-element well section data is saved to the original data buffer.

[0231] Once the data has accumulated to a certain level, it will be sent to the parameter recommendation subsystem.

[0232] In one embodiment, the data preprocessing subsystem performs data preprocessing according to operating conditions, including:

[0233] Based on the current operating conditions of the received data and the switching status between previous and subsequent operating conditions, the raw data is preprocessed accordingly.

[0234] In the preprocessing stage, the drilling status is first determined based on the well site data; if it is a drilling status, the sliding drilling status is further determined; if it is not a drilling status, the status parameters are updated and the data buffer and mechanical drilling rate calculation queue are cleared.

[0235] When determining the sliding drilling state, if it is a sliding drilling state, the screw speed and screw torque are recalculated and the state parameters are updated; if it is not a sliding drilling state, no action is taken.

[0236] In one embodiment, the data preprocessing subsystem cleans the data using expert thresholding and velocity constraint methods, including:

[0237] By setting expert thresholds to monitor abnormal values ​​in the data, the expert thresholds include at least: drilling pressure range, composite drilling torque range, sliding drilling torque range, composite drilling surface rotation speed range, sliding drilling screw rotation speed range, inlet flow range, riser pressure range, and drilling speed range.

[0238] Outliers are cleaned using the expert threshold method, and the remaining data are further evaluated using the velocity constraint method to clean up data that does not conform to the upper and lower bounds of the velocity value.

[0239] In one embodiment, the data preprocessing subsystem calculates the mechanical drilling rate according to well depth intervals, including:

[0240] The mechanical drilling rate is determined based on well depth intervals, and a queue is maintained according to these intervals. The window size of the queue dynamically changes, and each time a new mechanical drilling rate is calculated, the calculated data is added to the end of the queue. The mechanical drilling rate is calculated as follows:

[0241]

[0242] Among them, ROP now This refers to the newly added tail data; Depth1, Depth now These are the first and last data points in the well depth data queue, respectively; Time1, Time... now These are the first and last data points in the time data queue, respectively.

[0243] In one embodiment, the data preprocessing subsystem saves the processed micro-element well section data to the original data buffer, including:

[0244] The processed data is stored in a dynamic buffer of micro-well segments. When the interval between the well depth of the last data in the dynamic buffer of micro-well segments and the well depth of the first data is greater than or equal to the size of the buffer, all data in the dynamic buffer of micro-well segments is sent to the parameter recommendation subsystem. After the sending is completed, the dynamic buffer of micro-well segments is cleared and the data is re-accumulated.

[0245] In one embodiment, a parameter recommendation subsystem is used to calculate drilling performance metrics based on a physical model, aiming to optimize the drilling performance metrics, and to construct a functional relationship between drilling parameters and drilling performance metrics through dynamic learning; and to recommend optimized drilling parameters to the driller based on the functional relationship, including:

[0246] Drilling performance metrics are calculated based on the cleaned data, wherein the drilling performance metrics include at least: mechanical specific energy and optimization targets;

[0247] With the goal of improving drilling performance metrics, a combined approach of linear exploration and nonlinear application is used to recommend drilling parameters. After the parameter recommendation subsystem is activated, it enters exploration mode, exploring and collecting data in the direction of increasing mechanical drilling rate based on the trend of linear model fitting. After sufficient exploration, it enters learning mode, learning the functional relationship between mechanical drilling rate and drilling parameters in the current exploration data based on the model to determine the optimal combination of drilling parameters. Maintaining the optimal drilling parameters, it enters application mode, monitoring the current drilling status while breaking rock, and automatically switching between learning mode and exploration mode based on the degree of performance degradation of mechanical drilling rate.

[0248] In one embodiment, the parameter recommendation subsystem is further used for:

[0249] The time and well depth of the last data point in the micro-element well segment data are extracted and used as the time and well depth of the merged data. The other data are averaged using Kalman filtering to obtain the merged data. The merged data is then input into the parameter recommendation subsystem.

[0250] In one embodiment, the parameter recommendation subsystem calculates drilling performance metrics based on the cleaned data, including:

[0251] The formula for calculating mechanical specific energy is:

[0252]

[0253] Where MSE is mechanical specific energy; EFF s Efficiency coefficient; WOB is drilling pressure; T is torque; RPM is rotational speed; ROP is mechanical drilling speed; d B The drill bit diameter is the mechanical energy; mechanical energy is the work done by the drill bit to break a unit volume of rock, used to evaluate drilling efficiency. A higher mechanical energy value indicates lower rock breaking efficiency, poor adaptability of the drill bit to the formation, and the need to optimize drilling parameters.

[0254] The formula for calculating the optimization objective is as follows:

[0255]

[0256] Wherein, Objective is the optimization objective, which is to maximize the mechanical drilling rate while minimizing the mechanical specific energy; ROP0 and MSE0 are normalization quantities used to remove dimensions.

[0257] In one embodiment, a parameter recommendation subsystem is provided, the modes of which include: exploration mode, learning mode, and application mode.

[0258] In one embodiment, the parameter recommendation subsystem has a mode switching method as follows:

[0259] Determine whether the parameter recommendation subsystem is in exploration mode; if it is in exploration mode, further determine whether the exploration is sufficient; if it is not in exploration mode, evaluate the drilling performance.

[0260] When determining whether the exploration is sufficient, if the exploration is sufficient, switch to learning mode and use a nonlinear model to learn the optimal parameters; if the exploration is insufficient, switch to exploration mode and use a linear model to explore the space.

[0261] When evaluating drilling performance metrics, determine whether drilling performance has deteriorated; if no deterioration has occurred, switch to application mode and maintain the current parameters for rock breaking; if deterioration has occurred, determine whether the deterioration index has reached the deterioration threshold.

[0262] When determining whether the indicators of decline have reached the decline threshold, if they have not, switch to learning mode and use a nonlinear model to learn the optimal parameters; if they have, switch to exploration mode and use a linear model to explore the space.

[0263] In one embodiment, the parameter recommendation subsystem includes a mode switching method in exploration mode, comprising:

[0264] The exploration mode of the parameter recommendation subsystem is used to guide the driller to explore in the direction of increasing mechanical drilling speed. By exploring the parameter space, the correspondence between operating parameters and mechanical drilling speed is obtained, and the area with the maximum mechanical drilling speed is found.

[0265] When exploration is sufficient, switch from exploration mode to learning mode. The switching condition is to determine whether a predetermined number of explorations has been completed. The minimum number of explorations and the maximum number of explorations are set. When the minimum number of explorations is reached, the current drilling speed is checked before each exploration. If the drilling speed exceeds the drilling speed threshold, exploration continues; otherwise, switch to learning mode. When the maximum number of explorations is exceeded, the exploration ends.

[0266] In one embodiment, the parameter recommendation subsystem includes a mode switching method in learning mode or application mode, comprising:

[0267] After the parameter recommendation subsystem switches to learning mode or application mode, it enters the dynamic monitoring of drilling performance stage; if the drilling performance reaches the performance threshold, the current operating parameters are maintained and the system enters application mode; if the drilling performance deteriorates but does not reach the deterioration threshold, the drilling parameters are adjusted and the system enters learning mode; if the drilling performance deteriorates and reaches the deterioration threshold, the system switches to the exploration stage.

[0268] The mechanical drilling speed score is used to determine the current drilling performance. Two threshold values ​​are set to evaluate the current drilling effect. If the mechanical drilling speed score falls below the first threshold, the drilling performance is considered to have severely deteriorated. If the mechanical drilling speed score falls above the second threshold, the drilling performance is considered to be stable. If the mechanical drilling speed score falls between the first and second thresholds, the drilling performance is considered to have slightly deteriorated. The mechanical drilling speed score is calculated as follows:

[0269]

[0270] ROP score ROP (Recovery Points) is the score for mechanical drilling speed. now This represents the current mechanical drilling speed; ROP history The historical mechanical drilling rate is used; the parameter recommendation subsystem calculates the mechanical drilling rate score for the current anchor point and the historical anchor point every 10 micro-element well sections of data received.

[0271] In one embodiment, a parameter recommendation subsystem is included, and the method further includes:

[0272] Calculate the mechanical drilling rate score for multiple consecutive tests;

[0273] Based on the mechanical drilling speed score threshold, multiple mechanical drilling speed scores are divided into different intervals, and the interval with the most mechanical drilling speed scores is selected as the judgment result for this time.

[0274] In one embodiment, the parameter recommendation subsystem recommends optimized drilling parameters to the driller based on the functional relationship, including:

[0275] When recommending parameters under different operating conditions, for composite drilling, the functional relationship between drilling operation parameters and drilling performance improvement targets is as follows:

[0276] ROP = f(WOB, RPM) surface (Qpdm)

[0277] OBJ = f(WOB, RPM) surface (Qpdm)

[0278] ROP is the mechanical drilling rate; OBJ is the optimization target; WOB is the drilling pressure; RPM surface Where is the ground rotational speed; Qpdm is the displacement; recommended parameters are given based on the aforementioned functional relationship;

[0279] For sliding drilling, the functional relationship between drilling operation parameters and drilling performance improvement targets is as follows:

[0280] ROP = f(WOB, Qpdm)

[0281] OBJ = f(WOB, Qpdm)

[0282] Recommended parameters are given based on the aforementioned functional relationship.

[0283] In one embodiment, the parameter recommendation subsystem recommends optimized drilling parameters to the driller based on the functional relationship, including:

[0284] When recommending parameters in the exploration mode, a support vector regression model is used to fit the relationship between drilling operation parameters, mechanical drilling rate and optimization target, and the recommended parameters are determined based on the fitted relationship.

[0285] For composite drilling or sliding drilling, a support vector regression model is used to fit the linear relationship of the target calculation method. Based on the slope obtained by parameter fitting, and using the boundary values ​​of the explored parameter interval as a basis, the recommended parameters are adjusted in the direction of increasing mechanical drilling rate according to the slope change. If the adjusted recommended parameters exceed the values ​​specified in the drilling design, the drilling design is used as the recommended parameters.

[0286] In one embodiment, the parameter recommendation subsystem recommends optimized drilling parameters to the driller based on the functional relationship, including:

[0287] When recommending parameters in the learning mode, a random forest regression model is used for nonlinear fitting to fit the relationship between drilling operation parameters, mechanical drilling rate and optimization objective, and the recommended parameters are determined based on the fitted relationship.

[0288] For composite drilling or sliding drilling, a random forest regression model is used to fit the linear relationship of the optimization target calculation method. After fitting, a grid search is performed in the parameter space based on the random forest regression model to obtain the parameters corresponding to the maximum point of the mechanical drilling rate or optimization target. It is then determined whether the parameters meet the conditions, and if they do, recommendations are made.

[0289] In one embodiment, the parameter recommendation subsystem recommends optimized drilling parameters to the driller based on the functional relationship, including:

[0290] When recommending parameters in the application mode, the previously recommended parameters are used.

[0291] It should be noted that although several modules of the drilling parameter recommendation device based on the AI ​​model have been mentioned in the detailed description above, this division is merely exemplary and not mandatory. In fact, according to embodiments of the present invention, the features and functions of two or more modules described above can be embodied in one module. Conversely, the features and functions of one module described above can be further divided and embodied by multiple modules.

[0292] Based on the aforementioned inventive concept, such as Figure 16 As shown, the present invention also proposes a computer device 1600, including a memory 1610, a processor 1620, and a computer program 1630 stored in the memory 1610 and executable on the processor 1620. When the processor 1620 executes the computer program 1630, it implements the aforementioned drilling parameter recommendation method based on the AI ​​model.

[0293] Based on the aforementioned inventive concept, the present invention proposes a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the aforementioned drilling parameter recommendation method based on an AI model.

[0294] Based on the aforementioned inventive concept, this invention proposes a computer program product, which includes a computer program that, when executed by a processor, implements a drilling parameter recommendation method based on an AI model.

[0295] This invention discloses a drilling parameter recommendation method and apparatus based on an AI model. It collects raw well site data and performs corresponding preprocessing before providing it to a parameter recommendation system. The parameter recommendation subsystem receives data from the data processing subsystem, calculates various metrics on the data, learns and constructs a functional relationship between drilling parameters and target metrics based on the AI ​​model, and recommends drilling operation parameters that can improve the target metrics based on this functional relationship. This invention applies an AI model to drilling parameter recommendation, enabling it to identify different operating conditions and clean the raw data. It employs a combination of linear exploration and nonlinear application to recommend drilling parameters that optimize the target metrics under different modes. This invention overcomes the shortcomings of traditional drilling parameter recommendation methods, such as cumbersome processes, long processing times, and poor accuracy. It can dynamically and efficiently recommend optimal drilling parameters and optimize drilling speed, thereby achieving the goal of reducing costs and increasing efficiency in the drilling process.

[0296] The acquisition, storage, use, and processing of data in this application comply with relevant laws and regulations.

[0297] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, apparatus, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0298] This invention is described with reference to flowchart illustrations and / or block diagrams of methods and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0299] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0300] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0301] Finally, it should be noted that the above-described embodiments are merely specific implementations of the present invention, used to illustrate the technical solutions of the present invention, and not to limit it. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments within the technical scope disclosed in the present invention, or make equivalent substitutions for some of the technical features; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A drilling parameter recommendation method based on an AI model, characterized in that, include: Obtain the raw data and clean it. The cleaned data is input into the parameter recommendation subsystem; among which, drilling performance metrics are calculated based on the physical model, with the goal of optimizing the drilling performance metrics, and the functional relationship between drilling parameters and drilling performance metrics is constructed through dynamic learning; Based on the aforementioned functional relationship, optimized drilling parameters are recommended to the driller; The cleaned data is input into the parameter recommendation subsystem; among which, drilling performance metrics are calculated based on the physical model, with the goal of optimizing the drilling performance metrics, and a functional relationship between drilling parameters and drilling performance metrics is constructed through dynamic learning, including: Drilling performance metrics are calculated based on the cleaned data, wherein the drilling performance metrics include at least: mechanical specific energy and optimization targets; With the goal of improving drilling performance metrics, a combined approach of linear exploration and nonlinear application is used to recommend drilling parameters. After the parameter recommendation subsystem is activated, it enters exploration mode, exploring and collecting data in the direction of increasing mechanical drilling rate based on the trend of linear model fitting. After sufficient exploration, it enters learning mode, learning the functional relationship between mechanical drilling rate and drilling parameters in the current exploration data based on the model to determine the optimal combination of drilling parameters. Maintaining the optimal drilling parameters, it enters application mode, monitoring the current drilling status while breaking rock, and automatically switching between learning mode and exploration mode based on the degree of performance degradation of mechanical drilling rate. The methods for switching modes in learning mode or application mode include: After the parameter recommendation subsystem switches to learning mode or application mode, it enters the dynamic monitoring of drilling performance stage; if the drilling performance reaches the performance threshold, the current operating parameters are maintained and the system enters application mode; if the drilling performance deteriorates but does not reach the deterioration threshold, the drilling parameters are adjusted and the system enters learning mode; if the drilling performance deteriorates and reaches the deterioration threshold, the system switches to the exploration stage. The mechanical drilling speed score is used to determine the current drilling performance. Two threshold values ​​are set to evaluate the current drilling effect. If the mechanical drilling speed score falls below the first threshold, the drilling performance is considered to have severely deteriorated. If the mechanical drilling speed score falls above the second threshold, the drilling performance is considered to be stable. If the mechanical drilling speed score falls between the first and second thresholds, the drilling performance is considered to have slightly deteriorated. The mechanical drilling speed score is calculated as follows: ROP score Score for mechanical drilling speed; ROP now This represents the current mechanical drilling speed; ROP history The historical mechanical drilling rate is used; the parameter recommendation subsystem calculates the mechanical drilling rate score for the current anchor point and the historical anchor point every 10 micro-element well sections of data received. The method also includes: Calculate the mechanical drilling rate score for multiple consecutive tests; Based on the mechanical drilling speed score threshold, multiple mechanical drilling speed scores are divided into different intervals, and the interval with the most mechanical drilling speed scores is selected as the judgment result for this time.

2. The method according to claim 1, characterized in that, Obtain raw data and clean the raw data, including: The data preprocessing subsystem uses an independent thread to obtain raw data via the HTTP protocol. Data preprocessing is performed according to the working conditions. The data is cleaned using the expert threshold method and the velocity constraint method. The mechanical drilling rate is calculated according to the well depth interval. The processed micro-element well section data is saved to the original data buffer. Once the data has accumulated to a certain level, it will be sent to the parameter recommendation subsystem.

3. The method according to claim 2, characterized in that, Data preprocessing is performed based on operating conditions, including: Based on the current operating conditions of the received data and the switching status between previous and subsequent operating conditions, the raw data is preprocessed accordingly. In the preprocessing stage, the drilling status is first determined based on the well site data; if it is a drilling status, the sliding drilling status is further determined; if it is not a drilling status, the status parameters are updated and the data buffer and mechanical drilling rate calculation queue are cleared. When determining the sliding drilling state, if it is a sliding drilling state, the screw speed and screw torque are recalculated and the state parameters are updated; if it is not a sliding drilling state, no action is taken.

4. The method according to claim 2, characterized in that, Data cleaning was performed using expert thresholding and velocity constraint methods, including: By setting expert thresholds to monitor abnormal values ​​in the data, the expert thresholds include at least: drilling pressure range, composite drilling torque range, sliding drilling torque range, composite drilling surface rotation speed range, sliding drilling screw rotation speed range, inlet flow range, riser pressure range, and drilling speed range. Outliers are cleaned using the expert threshold method, and the remaining data are further evaluated using the velocity constraint method to clean up data that does not conform to the upper and lower bounds of the velocity value.

5. The method according to claim 2, characterized in that, Calculating the mechanical drilling rate according to well depth intervals includes: The mechanical drilling rate is determined based on well depth intervals, and a queue is maintained according to these intervals. The window size of the queue dynamically changes, and each time a new mechanical drilling rate is calculated, the calculated data is added to the end of the queue. The mechanical drilling rate is calculated as follows: in, ROP now This refers to the newly added data at the back of the queue; Depth 1. Depth now These are the first and last data points in the well depth data queue, respectively. Time 1. Time now These are the first and last data points in the time data queue, respectively.

6. The method according to claim 2, characterized in that, The processed micro-element well section data is saved to the original data buffer, including: The processed data is stored in a dynamic buffer of micro-well segments. When the interval between the well depth of the last data in the dynamic buffer of micro-well segments and the well depth of the first data is greater than or equal to the size of the buffer, all data in the dynamic buffer of micro-well segments is sent to the parameter recommendation subsystem. After the sending is completed, the dynamic buffer of micro-well segments is cleared and the data is re-accumulated.

7. The method according to claim 1, characterized in that, The method also includes: The time and well depth of the last data point in the micro-element well segment data are extracted and used as the time and well depth of the merged data. The other data are averaged using Kalman filtering to obtain the merged data. The merged data is then input into the parameter recommendation subsystem.

8. The method according to claim 1, characterized in that, Drilling performance metrics are calculated based on the cleaned data, including: The formula for calculating mechanical specific energy is: in, MSE Mechanical specific energy; EFF s Efficiency coefficient; WOB For drilling pressure; T Torque; RPM Rotational speed; ROP This refers to the mechanical drilling speed; d B The drill bit diameter is the mechanical energy; mechanical energy is the work done by the drill bit to break a unit volume of rock, used to evaluate drilling efficiency. A higher mechanical energy value indicates lower rock breaking efficiency, poor adaptability of the drill bit to the formation, and the need to optimize drilling parameters. The formula for calculating the optimization objective is as follows: in, The optimization objective is to maximize the mechanical drilling rate while minimizing the mechanical specific energy. ROP 0、 MSE 0 is a normalized value, used to remove dimensions.

9. The method according to claim 1, characterized in that, The modes of the parameter recommendation subsystem include: exploration mode, learning mode, and application mode.

10. The method according to claim 9, characterized in that, The mode switching method of the parameter recommendation subsystem is as follows: Determine whether the parameter recommendation subsystem is in exploration mode; if it is in exploration mode, further determine whether the exploration is sufficient; if it is not in exploration mode, evaluate the drilling performance. When determining whether the exploration is sufficient, if the exploration is sufficient, switch to learning mode and use a nonlinear model to learn the optimal parameters; If the exploration is insufficient, switch to exploration mode and explore the space using a linear model. When evaluating drilling performance metrics, determine whether drilling performance has deteriorated; If no degradation occurs, switch to application mode and maintain the current rock breaking parameters; If a recession occurs, determine whether the indicators for assessing the recession have reached the recession threshold. When determining whether the indicators of decline have reached the decline threshold, if they have not, switch to learning mode and use a nonlinear model to learn the optimal parameters; if they have, switch to exploration mode and use a linear model to explore the space.

11. The method according to claim 10, characterized in that, The methods for switching modes in exploration mode include: The exploration mode of the parameter recommendation subsystem is used to guide the driller to explore in the direction of increasing mechanical drilling speed. By exploring the parameter space, the correspondence between operating parameters and mechanical drilling speed is obtained, and the area with the maximum mechanical drilling speed is found. When exploration is sufficient, switch from exploration mode to learning mode. The switching condition is to determine whether a predetermined number of explorations has been completed. The minimum number of explorations and the maximum number of explorations are set. When the minimum number of explorations is reached, the current drilling speed is checked before each exploration. If the drilling speed exceeds the drilling speed threshold, exploration continues; otherwise, switch to learning mode. When the maximum number of explorations is exceeded, the exploration ends.

12. The method according to claim 1, characterized in that, Based on the aforementioned functional relationship, optimized drilling parameters are recommended to the driller, including: When recommending parameters under different operating conditions, for composite drilling, the functional relationship between drilling operation parameters and drilling performance improvement targets is as follows: This refers to the mechanical drilling speed; To optimize the objective; For drilling pressure; Ground rotation speed; For displacement; recommended parameters are given based on the aforementioned functional relationship; For sliding drilling, the functional relationship between drilling operation parameters and drilling performance improvement targets is as follows: Recommended parameters are given based on the aforementioned functional relationship.

13. The method according to claim 1, characterized in that, Based on the aforementioned functional relationship, optimized drilling parameters are recommended to the driller, including: When recommending parameters in the exploration mode, a support vector regression model is used to fit the relationship between drilling operation parameters, mechanical drilling rate and optimization target, and the recommended parameters are determined based on the fitted relationship. For composite drilling or sliding drilling, a support vector regression model is used to fit the linear relationship of the target calculation method. Based on the slope obtained by parameter fitting, and using the boundary values ​​of the explored parameter interval as a basis, the recommended parameters are adjusted in the direction of increasing mechanical drilling rate according to the slope change. If the adjusted recommended parameters exceed the values ​​specified in the drilling design, the drilling design is used as the recommended parameters.

14. The method according to claim 1, characterized in that, Based on the aforementioned functional relationship, optimized drilling parameters are recommended to the driller, including: When recommending parameters in the learning mode, a random forest regression model is used for nonlinear fitting to fit the relationship between drilling operation parameters, mechanical drilling rate and optimization objective, and the recommended parameters are determined based on the fitted relationship. For composite drilling or sliding drilling, a random forest regression model is used to fit the linear relationship of the optimization target calculation method. After fitting, a grid search is performed in the parameter space based on the random forest regression model to obtain the parameters corresponding to the maximum point of the mechanical drilling rate or optimization target. It is then determined whether the parameters meet the conditions, and if they do, recommendations are made.

15. The method according to claim 1, characterized in that, Based on the aforementioned functional relationship, optimized drilling parameters are recommended to the driller, including: When recommending parameters in the application mode, the previously recommended parameters are used.

16. A drilling parameter recommendation device based on an AI model, characterized in that, include: A data preprocessing subsystem is used to acquire raw data and clean the raw data. The parameter recommendation subsystem is used to calculate drilling performance metrics based on a physical model, with the goal of optimizing these metrics. It constructs a functional relationship between drilling parameters and drilling performance metrics through dynamic learning, and recommends optimized drilling parameters to the driller based on this functional relationship. Specifically, the parameter recommendation subsystem is used for: Drilling performance metrics are calculated based on the cleaned data, wherein the drilling performance metrics include at least: mechanical specific energy and optimization targets; With the goal of improving drilling performance metrics, a combined approach of linear exploration and nonlinear application is used to recommend drilling parameters. After the parameter recommendation subsystem is activated, it enters exploration mode, exploring and collecting data in the direction of increasing mechanical drilling rate based on the trend of linear model fitting. After sufficient exploration, it enters learning mode, learning the functional relationship between mechanical drilling rate and drilling parameters in the current exploration data based on the model to determine the optimal combination of drilling parameters. Maintaining the optimal drilling parameters, it enters application mode, monitoring the current drilling status while breaking rock, and automatically switching between learning mode and exploration mode based on the degree of performance degradation of mechanical drilling rate. The methods for switching modes in learning mode or application mode include: After the parameter recommendation subsystem switches to learning mode or application mode, it enters the dynamic monitoring of drilling performance stage; if the drilling performance reaches the performance threshold, the current operating parameters are maintained and the system enters application mode; if the drilling performance deteriorates but does not reach the deterioration threshold, the drilling parameters are adjusted and the system enters learning mode; if the drilling performance deteriorates and reaches the deterioration threshold, the system switches to the exploration stage. The mechanical drilling speed score is used to determine the current drilling performance. Two threshold values ​​are set to evaluate the current drilling effect. If the mechanical drilling speed score falls below the first threshold, the drilling performance is considered to have severely deteriorated. If the mechanical drilling speed score falls above the second threshold, the drilling performance is considered to be stable. If the mechanical drilling speed score falls between the first and second thresholds, the drilling performance is considered to have slightly deteriorated. The mechanical drilling speed score is calculated as follows: ROP score Score for mechanical drilling speed; ROP now This represents the current mechanical drilling speed; ROP history The historical mechanical drilling rate is used; the parameter recommendation subsystem calculates the mechanical drilling rate score for the current anchor point and the historical anchor point every 10 micro-element well sections of data received. Calculate the mechanical drilling rate score for multiple consecutive tests; Based on the mechanical drilling speed score threshold, multiple mechanical drilling speed scores are divided into different intervals, and the interval with the most mechanical drilling speed scores is selected as the judgment result for this time.

17. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method of any one of claims 1 to 15.

18. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method of any one of claims 1 to 15.

19. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the method of any one of claims 1 to 15.

Images