A method and system for optimizing drilling control parameters

Abstract

The application discloses a method and system for optimizing drilling control parameters, comprising: obtaining real-time logging data of a drilling site and converting the real-time logging data into well depth logging data; obtaining stratum characteristic parameters of a to-be-drilled stratum by using a pre-constructed stratum characteristic parameter prediction model according to the well depth logging data; and solving a pre-constructed drilling pressure and rotating speed target optimization model based on the stratum characteristic parameters of the to-be-drilled stratum and site correlation parameters of the drilling, so as to obtain optimal drilling control parameters of the to-be-drilled stratum. The application can be effectively applied to drilling operations, improves drilling efficiency and reduces drilling cost.

Description

Technical Field

[0001] This invention relates to the field of oil and gas drilling rig control technology, and in particular to a method and system for optimizing drilling control parameters. Background Technology

[0002] The Sichuan Basin is rich in oil and gas resources and is an important area for natural gas exploration and development in China. However, the geological conditions of the oil and gas extraction areas in the Sichuan Basin are complex and diverse, and the oil and gas extraction process faces challenges such as deep burial of oil and gas resources and complex geological pressure systems. Due to the coexistence and superposition of multiple factors, current drilling operations face problems such as long cycles, high costs, and high dependence on experts.

[0003] In current drilling operations, the selection of drilling control parameters such as drilling pressure and rotation speed mainly relies on the experience of on-site drillers and experts. This often leads to a lag in response to complex and diverse geological conditions, and the selection of overly conservative drilling control parameters when faced with unknown formation information, resulting in long drilling operation cycles and high costs.

[0004] Following a series of studies, most oil and gas exploration and development drilling companies have established relevant drilling data management platforms, capable of collecting, monitoring, and recording various drilling engineering information and geological information in real time. However, existing platforms have shortcomings in the effective utilization of drilling data, and there is still room for improvement in drilling parameter optimization and decision support. There is an urgent need to develop drilling control parameter optimization technology based on artificial intelligence.

[0005] Drilling parameter optimization refers to the process of optimizing and adjusting a series of controllable parameters, including drilling pressure and drill bit rotation speed, during drilling to achieve the desired drilling results. Existing drilling parameter optimization methods mainly construct optimization models based on drilling process mechanism models, and then use various optimization algorithms to calculate and solve the target optimization model.

[0006] Common drilling models based on mechanistic models include the Young's mechanical rate of penetration (MRP) model, the multivariate linear rate of penetration (MRP) model, and the bit energy specificity model. Methods for solving objective optimization models mainly include genetic algorithms, particle swarm optimization (PSO) algorithms, and pattern search methods. Existing methods for constructing and solving optimization models have high data requirements and suffer from low algorithm stability and poor model generalization ability when dealing with complex and diverse formations. Summary of the Invention

[0007] The purpose of this invention is to provide a scheme for optimizing drilling control parameters so that the method for determining drilling optimization parameters can effectively overcome the uncertainty, nonlinearity and parallelism of the drilling process.

[0008] To address the aforementioned technical problems, this invention provides a method for optimizing drilling control parameters, comprising: acquiring real-time logging data from the drilling site and converting it into well depth logging data; obtaining formation characteristic parameters of the formation to be drilled using a pre-constructed formation characteristic parameter prediction model based on the well depth logging data; and solving a pre-constructed drill pressure and rotational speed target optimization model based on the formation characteristic parameters of the formation to be drilled and the field correlation parameters of the drilling operation to obtain the optimal drilling control parameters of the formation to be drilled.

[0009] Preferably, the well depth logging data includes, but is not limited to, well depth, drilling speed, drilling pressure, rotational speed, torque, standpipe pressure, displacement, drilling fluid density, and drill bit diameter. The step of converting the real-time logging data into the well depth logging data includes: performing multi-source data fusion processing on the real-time logging data and the drill bit information during drilling to obtain logging data to be processed; deleting invalid logging data from the logging data to be processed; and resampling the logging data to be processed after deleting invalid logging data, using the well depth as an index, to obtain the well depth logging data.

[0010] Preferably, the invalid logging data includes, but is not limited to: abnormal logging data, missing logging data, and logging data under non-drilling conditions. The step of deleting invalid logging data from the logging data to be processed includes: identifying abnormal and missing logging data, including: setting a sliding window and a fitting method; and fitting the logging data captured in the sliding window during the window sliding process according to the fitting method to identify abnormal and missing logging data in the logging data to be processed.

[0011] Preferably, the formation feature parameter prediction model is constructed through the following steps: a training dataset is constructed based on historical drilling data from multiple drilled wells; a preset model to be trained is trained based on the training dataset to obtain the formation feature parameter prediction model; wherein, the historical drilling data of the drilled wells includes: well depth logging data and corresponding formation feature parameters; the model to be trained is a long short-term memory recurrent neural network model.

[0012] Preferably, the formation characteristic parameters of the formation to be drilled include: a formation drillability index, wherein the formation characteristic parameter prediction model is represented by the following expression:

[0013] K d =f(H,v pc ,W,N,T,p m ,Q,ρ,d,A)

[0014] Among them, K d The formation drillability index is represented by H, which represents the current well depth, and v. pcW represents drilling speed, N represents drilling pressure, T represents drilling speed, and p represents torque. m Q represents riser pressure, ρ represents drilling fluid density, d represents drill bit diameter, and A represents drill bit type.

[0015] Preferably, the historical drilling data of multiple drilled wells is obtained by following these steps: multi-source data fusion processing is performed based on the comprehensive logging data, daily drilling and completion reports, and drilling and completion reports of multiple drilled wells at different logging times to obtain logging feature data; invalid logging data is deleted from the logging feature data; the logging feature data after deleting invalid logging data is resampled using well depth as an index to obtain well depth logging data for each drilled well; and formation characteristic parameters for each drilled well are calculated based on the logging feature data.

[0016] Preferably, the drilling pressure and rotational speed target optimization model is constructed through the following steps: constructing optimization variables based on the drilling control parameters of the formation to be drilled; constructing an objective function based on the optimization variables, with the goal of maximizing the drilling speed of the drilling machinery and minimizing the drill bit wear rate in the drilling operation; constructing constraint conditions based on the range of drilling pressure, rotational speed, and drill bit tooth wear in the drilling operation; and constructing the drilling pressure and rotational speed target optimization model based on the objective function and the constraint conditions.

[0017] Preferably, the objective function is represented by the following expression:

[0018]

[0019] Where F1 represents the drilling speed of the drilling machinery, K d C represents the formation drillability index. p C represents the pressure difference influence coefficient. H The hydraulic purification coefficient is represented by W, the drilling pressure by M, the threshold drilling pressure by N, the rotational speed by λ, the rotational speed exponent by C2, the drill bit tooth wear coefficient by h, the amount of drill bit tooth wear by F2, and the drill bit wear rate by A. f The coefficients represent rock abrasiveness, a1 and a2 are both rotation speed influence coefficients, N represents rotation speed, Z2 and Z1 are both drilling pressure influence coefficients, W represents drilling pressure, C1 represents tooth wear reduction coefficient, and h represents drill bit tooth wear amount.

[0020] The constraint condition is expressed using the following expression:

[0021]

[0022] Where M represents the threshold drill pressure, W represents the drill pressure, and W max N represents the maximum drilling pressure. min N represents the minimum rotational speed, and N represents the rotational speed. maxThe value of represents the maximum rotational speed, and h represents the amount of wear on the drill bit teeth.

[0023] Preferably, the optimal drilling control parameters for the formation to be drilled are the optimal combination of drilling pressure and rotation speed. A genetic algorithm is used to solve the target optimization model for drilling pressure and rotation speed, including: S1: setting the population size and maximum number of iterations; S2: initializing the number of iterations and performing population initialization operations based on optimization variables and constraints; S3: performing fast non-dominated sorting on the individuals in the initial population, followed by selection, crossover, and mutation operations to obtain the parent population; S4: incrementing the iteration count by 1 and selecting individuals from the parent population... S5: Perform selection, crossover, and mutation operations to obtain the offspring population; S6: Merge the parent and offspring populations, and perform fast non-dominated sorting and crowding calculation on the individuals in the merged population to obtain a new parent population; S7: Determine if the number of iterations is less than the maximum number of iterations. If yes, proceed to S7; otherwise, use the new parent population as the parent population and return to S4; S8: Calculate the fitness value of each individual in the new parent population based on the objective function, and use the optimization variable corresponding to the individual with the highest fitness value as the optimal drilling control parameter.

[0024] On the other hand, embodiments of the present invention provide a system for optimizing drilling control parameters, comprising: a field data acquisition module configured to acquire real-time logging data from the drilling site and convert it into well depth logging data; a formation feature prediction module configured to obtain formation feature parameters of the formation to be drilled based on the well depth logging data and using a pre-built formation feature parameter prediction model; and an optimal drilling control parameter generation module configured to solve a pre-built target optimization model for drilling pressure and rotational speed based on the formation feature parameters of the formation to be drilled and the field correlation parameters of the drilling operation to obtain the optimal drilling control parameters of the formation to be drilled.

[0025] Compared with the prior art, one or more embodiments of the above solutions may have the following advantages or beneficial effects:

[0026] This invention proposes a method and system for optimizing drilling control parameters. The method and system acquire real-time logging data and convert it into depth logging data, reducing the requirements for on-site data acquisition. Furthermore, it employs a formation characteristic parameter prediction model to predict the geological characteristics of the formation to be drilled, and calculates the corresponding drilling control parameters based on the predicted geological characteristics, effectively overcoming the uncertainties, nonlinearities, and parallel processes in the drilling process. This solution can be effectively applied to drilling operations, improving drilling efficiency and reducing drilling costs.

[0027] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the description, claims, and drawings. Attached Figure Description

[0028] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with the embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0029] Figure 1 This is a schematic diagram illustrating the steps of a method for optimizing drilling control parameters according to an embodiment of this application.

[0030] Figure 2 This is a structural diagram of the formation characteristic parameter prediction model in the method for optimizing drilling control parameters according to an embodiment of this application.

[0031] Figure 3 This is a schematic diagram illustrating the specific process of using a genetic algorithm to solve the target optimization model of drilling pressure and rotational speed in the method for optimizing drilling control parameters according to an embodiment of this application.

[0032] Figure 4 This is a schematic diagram of the structure of a system for optimizing drilling control parameters according to an embodiment of this application. Detailed Implementation

[0033] The embodiments of the present invention will be described in detail below with reference to the accompanying drawings and examples, so that the process of how the present invention uses technical means to solve technical problems and achieve technical effects can be fully understood and implemented accordingly. It should be noted that, as long as there is no conflict, the various embodiments and features in the various embodiments of the present invention can be combined with each other, and the resulting technical solutions are all within the protection scope of the present invention.

[0034] Furthermore, the steps illustrated in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Also, although a logical order is shown in the flowcharts, in some cases the steps shown or described may be performed in a different order than that shown here.

[0035] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments. Unless the context clearly indicates otherwise, the singular forms “a” and “an” as used herein are also intended to include the plural. It should also be understood that the terms “comprising” and / or “including” as used herein specify the presence of the stated features, integers, steps, operations, units, and / or components, without excluding the presence or addition of one or more other features, integers, steps, operations, units, components, and / or combinations thereof.

[0036] The Sichuan Basin is rich in oil and gas resources and is an important area for natural gas exploration and development in China. However, the geological conditions of the oil and gas extraction areas in the Sichuan Basin are complex and diverse, and the oil and gas extraction process faces challenges such as deep burial of oil and gas resources and complex geological pressure systems. Due to the coexistence and superposition of multiple factors, current drilling operations face problems such as long cycles, high costs, and high dependence on experts.

[0037] In current drilling operations, the selection of drilling control parameters such as drilling pressure and rotation speed mainly relies on the experience of on-site drillers and experts. This often leads to a lag in response to complex and diverse geological conditions, and the selection of overly conservative drilling control parameters when faced with unknown formation information, resulting in long drilling operation cycles and high costs.

[0038] Following a series of studies, most oil and gas exploration and development drilling companies have established relevant drilling data management platforms, capable of collecting, monitoring, and recording various drilling engineering information and geological information in real time. However, existing platforms have shortcomings in the effective utilization of drilling data, and there is still room for improvement in drilling parameter optimization and decision support. There is an urgent need to develop drilling control parameter optimization technology based on artificial intelligence.

[0039] Drilling parameter optimization refers to the process of optimizing and adjusting a series of controllable parameters, including drilling pressure and drill bit rotation speed, during drilling to achieve the desired drilling results. Existing drilling parameter optimization methods mainly construct optimization models based on drilling process mechanism models, and then use various optimization algorithms to calculate and solve the target optimization model.

[0040] Common drilling models based on mechanistic models include the Young's mechanical rate of penetration (MRP) model, the multivariate linear rate of penetration (MRP) model, and the bit energy specificity model. Methods for solving objective optimization models mainly include genetic algorithms, particle swarm optimization (PSO) algorithms, and pattern search methods. Existing methods for constructing and solving optimization models have high data requirements and suffer from low algorithm stability and poor model generalization ability when dealing with complex and diverse formations.

[0041] Figure 1 This is a schematic diagram illustrating the steps of a method for optimizing drilling control parameters according to an embodiment of this application. The following refers to... Figure 1 The specific steps and flow of the method for optimizing drilling control parameters described in the embodiments of the present invention will be explained.

[0042] Step S110: Obtain real-time logging data from the drilling site and convert the real-time logging data into well depth logging data.

[0043] The method for optimizing drilling control parameters disclosed in this embodiment requires predicting formation characteristic parameters of the formation to be drilled based on real-time logging data from the drilling site. However, the real-time logging data acquired from the drilling site is time-indexed, while predicting formation characteristic parameters requires inputting data indexed by well depth. Therefore, it is necessary to acquire the real-time logging data from the drilling site and convert it into well depth logging data. Specifically, well depth logging data includes, but is not limited to: well depth, drilling speed, drilling pressure, rotational speed, torque, standpipe pressure, displacement, drilling fluid density, and drill bit diameter.

[0044] Optionally, to make the converted well depth logging data more consistent with the actual application needs of the drilling site, real-time logging data and drill bit information during drilling can be processed to obtain data to be processed. Invalid data in the data to be processed is then deleted, and the data to be processed after deleting invalid data is resampled to obtain well depth logging data. Specifically, step S110 may include the following sub-steps A1-A3:

[0045] Sub-step A1 involves performing multi-source data fusion processing on real-time logging data and drill bit information during drilling to obtain logging data to be processed.

[0046] Sub-step A2: Delete invalid logging data from the logging data to be processed.

[0047] Sub-step A3 involves resampling the logging data to be processed after deleting invalid logging data, using well depth as the index, to obtain well depth logging data.

[0048] In sub-step A1, real-time logging data is acquired by a comprehensive logging instrument at the drilling site. The real-time logging data is data collected in real time at the drilling site, including data corresponding to 72 features such as drilling pressure, rotation speed and drilling speed.

[0049] In sub-step A1, based on real-time logging data, unnecessary features are removed, and the drill bit information of the ongoing drilling is extracted, transformed into new features for the real-time logging data, and added to it. This results in a data table containing 15 features, from which the logging data to be processed is obtained. Specifically, the features included in the data table are: well depth, vertical well depth, drilling pressure, torque, rotational speed, pump flush, standpipe pressure, drilling fluid type, drilling fluid density, drilling fluid inlet flow rate, drilling fluid inlet temperature, drilling time, drill bit type, drill bit diameter, and drill bit wear per unit footage.

[0050] In sub-step A2, invalid logging data includes, but is not limited to: abnormal logging data, missing logging data, and logging data under non-drilling conditions.

[0051] In sub-step A3, resampling the logging data after deleting invalid data, using well depth as an index, not only standardizes the data but also removes a large amount of duplicate data, significantly reducing the data size. Specifically, all parameters corresponding to the real-time logging data are piecewise fitted at the well depth scale, and well depth logging data is obtained by sampling at the same well depth interval.

[0052] Optionally, in order to ensure that the converted well depth logging data meets the actual application requirements, a well depth interval of 0.1m can be preferred.

[0053] Furthermore, abnormal and missing logging data in the invalid logging data need to be identified using the Savitzky-Golay smoothing filter method and then deleted; while logging data from non-drilling conditions in the invalid logging data are directly deleted. Specifically, sub-step A2 includes identifying abnormal and missing logging data, which specifically includes the following steps B1-B2:

[0054] Step B1: Preset the sliding window and fitting method.

[0055] Step B2: Fit the logging data captured in the sliding window during the window sliding process according to the fitting method to identify abnormal logging data and missing logging data in the logging data to be processed.

[0056] The Savitzky-Golay smoothing filter method is a sliding window weighted average algorithm that performs k-order polynomial fitting on data points within a certain length window to obtain the fitted result.

[0057] Optionally, in step B1, considering that the actual logging data is continuous data sampled every 3 seconds at the drilling site, in order to significantly filter out abnormal and missing logging data in the logging data to be processed, a sliding window length of 11 and a second-order polynomial fitting method are preferred.

[0058] Step S120: Based on the well depth logging data, the formation characteristic parameters of the formation to be drilled are obtained using a pre-constructed formation characteristic parameter prediction model.

[0059] First, by mining historical drilling data from multiple drilled wells and constructing a formation characteristic parameter prediction model based on deep learning, the formation characteristic parameters of the formation to be drilled are predicted using the pre-constructed formation characteristic parameter prediction model based on well depth logging data. Predicting formation characteristic parameters of the formation to be drilled through the model does not require sophisticated measurement-while-drilling equipment, thus reducing application costs.

[0060] In this embodiment, the formation characteristic parameter prediction model is constructed through the following steps C1-C2, and in one embodiment, the structure of the constructed formation characteristic parameter prediction model is as follows: Figure 2 As shown.

[0061] Step C1: Construct a training dataset based on historical drilling data from multiple drilled wells.

[0062] Step C2: Based on the training dataset, train the preset model to be trained to obtain the formation feature parameter prediction model.

[0063] In step C1, the historical drilling data of the drilled wells includes: well depth logging data and corresponding formation characteristic parameters.

[0064] In step C2, LSTM is a special type of recurrent neural network. Compared to traditional RNNs, LSTM has better memory and can retain information from further back when processing continuous data with well depth as the sequence. It can predict the formation characteristic parameters of the formation to be drilled at the next moment using multiple drilled well data. Therefore, considering the temporal nature of well depth logging data and ensuring that the constructed formation characteristic parameter prediction model can accurately predict the formation characteristic parameters of the formation to be drilled, it is preferable to construct the formation characteristic parameter prediction model based on the LSTM principle. Specifically, the model to be trained is a Long Short-Term Memory (LSTM) recurrent neural network model.

[0065] Furthermore, historical drilling data for multiple drilled wells are obtained through the following steps D1-D4:

[0066] Step D1 involves performing multi-source data fusion processing based on the comprehensive logging data of multiple drilled wells, daily drilling and completion construction data, and drilling and completion reports at different logging times to obtain logging feature data.

[0067] Step D2: Delete invalid logging data from the logging feature data.

[0068] Step D3: Using well depth as an index, resample the logging feature data after deleting invalid logging data to obtain the well depth logging data for each drilled well.

[0069] Step D4: Calculate the formation characteristic parameters for each drilled well based on the logging characteristic data.

[0070] In step D1, the integrated logging data, drilling and completion daily data, and drilling and completion report are all downloaded from the EPBP (Electronic Power Platform) database. Furthermore, in this embodiment of the invention, step D2 is implemented using a method similar to step A2, and similarly, step D3 is implemented using a method similar to step A3.

[0071] In this embodiment, the formation characteristic parameters of the formation to be drilled include: the formation drillability index. The formation characteristic parameter prediction model is represented by the following expression:

[0072] K d =f(H,v pc ,W,N,T,p m ,Q,ρ,d,A) (1)

[0073] Among them, K d The formation drillability index is a unitless parameter used to characterize the influence of factors other than drill pressure, rotation speed, drill bit parameters, and hydraulic factors on the drilling rate; H represents the current well depth in meters; v pc The value represents the drilling speed, the depth the drill bit advances per hour; W represents the drilling pressure, in kN; N represents the rotational speed, in r / min; T represents the torque, in N·m; p m The unit represents riser pressure, in bar; Q represents displacement, in L / min; ρ represents drilling fluid density, in g / cm³. 3 ;d represents the drill bit diameter in mm;A represents the drill bit type, and A is one-hot encoded as the model input.

[0074] Step S130: Based on the formation characteristic parameters of the formation to be drilled and the field correlation parameters of the drilling operation, the pre-constructed target optimization model of drilling pressure and rotation speed is solved to obtain the optimal drilling control parameters of the formation to be drilled.

[0075] First, an optimization model for drilling pressure and rotation speed is constructed by solving the optimization objective. Then, based on the formation characteristic parameters of the formation to be drilled and the field correlation parameters of the drilling operation, the constructed optimization model for drilling pressure and rotation speed is used to solve the optimal drilling control parameters of the formation to be drilled. The drilling operation is then guided to be carried out efficiently according to the obtained optimal drilling control parameters.

[0076] In this embodiment, the field-related parameters of the drilling operation are calculated or read from real-time logging data, geological data, drill bit-related information, and other data at the drilling site.

[0077] In this embodiment, the target optimization model for drilling pressure and rotational speed is constructed through the following steps E1-E4:

[0078] Step E1: Construct optimization variables based on the drilling control parameters of the formation to be drilled.

[0079] Step E2: Based on the optimization variables, construct the objective function with the goal of maximizing the drilling speed of the drilling machinery and minimizing the drill bit wear rate during the drilling operation.

[0080] Step E3: Based on the range of drilling pressure, rotational speed, and drill bit tooth wear in the drilling operation, construct the constraints.

[0081] Step E4: Based on the objective function and constraints, construct the target optimization model for drilling pressure and rotational speed.

[0082] In step E1, the drilling control parameters include drilling pressure and rotational speed.

[0083] In step E2, the objective function is constructed through the following steps:

[0084] (1) Establish an objective function with drilling speed and drill bit wear as the targets, aiming to increase the drilling speed of the drilling machinery while ensuring the lowest possible drill bit wear rate, thereby reducing the increase in drilling costs caused by drill bit wear. The objective function is shown in the following formula:

[0085] F = [maxF1, minF2] (2)

[0086] F1 is the first optimization objective, representing the drilling speed of the drilling machinery, and drilling parameters are calculated to improve the drilling speed of the drilling machinery; F2 is the second optimization objective, representing the drill bit wear rate.

[0087] (2) The first optimization objective function is shown in the following equation:

[0088]

[0089] (3) The second optimization objective function is shown in the following equation:

[0090]

[0091] (4) Based on the above equations (2), (3), and (4), the objective function of the drilling pressure rotation speed target optimization model can be obtained as shown in the following equation:

[0092]

[0093] Where W represents drilling pressure; N represents rotational speed; K d The formation drillability index is predicted using a formation characteristic parameter prediction model; C p Indicates the pressure difference influence coefficient; C H λ represents the hydraulic purification coefficient; M represents the threshold drilling pressure, obtained through a five-point field test; C2 represents the drill bit tooth wear coefficient; h represents the drill bit tooth wear amount; λ represents the rotational speed index, generally less than 1. When purification is sufficient, λ = 1; when insufficient, λ < 1. The value of λ is obtained through a five-point field test; A f The coefficients represent rock abrasiveness, obtained from field drill bit data; a1 and a2 are both rotation speed influence coefficients; Z2 and Z1 are both drill pressure influence coefficients, obtained directly from the table based on the drill bit model and drill bit; C1 represents the tooth wear reduction coefficient; h represents the amount of drill bit tooth wear.

[0094] Specifically, C p and C H In practical applications, C2 and h are obtained through statistical analysis of historical drilling data; C2 and h are determined based on the type, model, and usage time of the drill bit used at the drilling site; a1, a2, and C1 are all related to the drill bit tooth profile and are obtained by looking up the table based on the drill bit model.

[0095] In step E3, the constraint conditions are expressed using the following expression:

[0096]

[0097] Where M represents the threshold drill pressure, W represents the drill pressure, and W max N represents the maximum drilling pressure. min N represents the minimum rotational speed, and N represents the rotational speed. max The value of represents the maximum rotational speed, and h represents the amount of wear on the drill bit teeth.

[0098] In this embodiment, the optimal drilling control parameters for the formation to be drilled are the optimal combination of drilling pressure and rotational speed. The NSGA-II algorithm is used to solve the target optimization model for drilling pressure and rotational speed, and outputs the most suitable recommended values ​​for drilling pressure and rotational speed. Field technicians control the drilling pressure and rotational speed according to the output recommended values ​​to improve the drilling speed of the drilling machinery and reduce the drill bit wear rate.

[0099] Among them, reference Figure 3 The genetic algorithm is used to solve the target optimization model for drilling pressure rotation speed. The specific steps include:

[0100] Step S1: Set the population size and maximum number of iterations;

[0101] Step S2: Initialize the number of iterations and perform population initialization based on optimization variables and constraints;

[0102] Step S3: First, perform fast non-dominated sorting on the individuals in the initial population, and then perform selection, crossover and mutation operations to obtain the parent population.

[0103] Step S4: Increment the iteration count by 1, and perform selection, crossover, and mutation operations on the individuals in the parent population to obtain the offspring population;

[0104] Step S5: Merge the parent population and the offspring population, and perform fast non-dominated sorting and crowding calculation on the individuals in the merged population to obtain a new parent population.

[0105] Step S6: Determine if the number of iterations is less than the maximum number of iterations. If yes, proceed to step S7. If no, use the new parent population as the parent population and return to step S4.

[0106] Step S7: Calculate the fitness value of each individual in the new parent population based on the objective function, and take the optimization variable corresponding to the individual with the largest fitness value as the optimal drilling control parameter.

[0107] Optionally, in step S1, the population size is set to popsize = 200 and the maximum number of iterations is set to maxgen = 160, depending on the actual application. In particular, when the calculation accuracy of the variables to be optimized, drilling pressure and rotation speed, is 0.001, setting the binary code length to 10 is sufficient to meet the accuracy requirements.

[0108] In step S2, the constraint bars of the effective solution space formed by the rotation speed and drilling pressure are used as the conditions for generating the initial population. A uniformly distributed initial solution is adopted. The specific conditions for generating the initial population are shown in the following formula:

[0109]

[0110] Among them, W i N i These represent the binary bits of the i-th individual in the initial population (i = 1, 2, ..., popsize), representing drilling pressure and rotation speed, respectively; W max W min These are the maximum and minimum limits of drilling pressure, N. max N min These are the maximum and minimum speed limits, respectively.

[0111] Optionally, in step S3, a roulette wheel selection method is used to select individuals from the initial population, where the probability P of selecting the i-th individual in the initial population is... i Determined by the following formula:

[0112]

[0113] Among them, fitness (W) i N i ) represents the fitness value of the i-th individual in the population, and popsize represents the population size.

[0114] Optionally, the crossover probability P c It refers to the probability of corresponding gene segments being exchanged on chromosomes, usually P. c ∈[0, 1]. A higher crossover probability means that individuals in the population can repeatedly crossover in each generation, covering a more complete solution space. A low crossover probability can lead to the next generation moving on before the genes have fully crossed over, resulting in low algorithm iteration efficiency. Therefore, considering the application scenarios of this invention, the optimal crossover probability P can be selected. c =0.75.

[0115] Optionally, the mutation probability P mThis refers to the probability of gene mutation on a chromosome. Gene mutation leads to genetic diversity in a population, and in algorithms, it enhances the global optimization of the algorithm, preventing it from getting trapped in local optima. Therefore, considering the application scenarios of this invention, a strategy of continuously adjusting the mutation probability is adopted during the iterative evolution of the population. That is, a higher mutation probability is selected in the early stage of iteration, and a lower mutation probability is selected in the later stage of iteration to improve the convergence speed. The optimal mutation probability P can be... m The setting is adjustable within the range [0, 0.1].

[0116] In step S5, the crowding degree of the i-th individual is calculated using the following expression:

[0117]

[0118] In the crowding calculation, individuals within the same level are sorted from left to right and from top to bottom according to the magnitude of objective functions F1 and F2, where f1 and f2 are the normalized calculation formulas of objective functions F1 and F2, respectively.

[0119] Specifically, It is calculated using the following expression:

[0120]

[0121] It is calculated using the following expression:

[0122]

[0123] It is calculated using the following expression:

[0124]

[0125] It is calculated using the following expression:

[0126]

[0127] In equation (9) above, i d This represents the crowding level of the i-th individual. This represents the normalized position of the next adjacent individual to the i-th individual in the population, measured by the F1 metric. This represents the normalized position of the preceding neighbor of the i-th individual in the population, measured by the F1 metric. This represents the normalized difference in the F1 direction between two adjacent individuals of the i-th individual in the population. This represents the normalized position of the next adjacent individual to the i-th individual in the population, measured by F2. This represents the normalized position of the preceding neighbor of the i-th individual in the population, measured by F2. This represents the normalized difference in the F2 direction between two adjacent individuals of the i-th individual in the population.

[0128] In equation (10) above, F1(W i+1 N i+1 F1(W,N) represents the first objective function value of the next adjacent individual of the i-th individual in the population, maxF1(W,N) represents the maximum value of the first objective function value of all individuals in the population, and minF1(W,N) represents the minimum value of the first objective function value of all individuals in the population.

[0129] In equation (11) above, F1(W i-1 N i-1 Let ) denote the first optimization objective function of the preceding individual adjacent to the i-th individual in the population.

[0130] In equation (12) above, F2(W i+1 N i+1 F2(W,N) represents the second objective function of the next adjacent individual of the i-th individual in the population, max F2(W,N) represents the maximum value of the second objective function of all individuals in the population, and min F2(W,N) represents the minimum value of the second objective function of all individuals in the population.

[0131] In equation (13) above, F2(W i-1 N i-1 Let ) represent the second optimization objective function of the preceding individual adjacent to the i-th individual in the population.

[0132] Based on the above-described method for optimizing drilling control parameters, the present invention also provides a system for optimizing drilling control parameters. This system for optimizing drilling control parameters is used to implement the method for optimizing drilling control parameters as described above.

[0133] Figure 4 This is a schematic diagram of a system for optimizing drilling control parameters according to an embodiment of this application. Figure 4 As shown, the system for optimizing drilling control parameters according to an embodiment of the present invention includes: a field data acquisition module 401, a formation feature prediction module 402, and an optimal drilling control parameter generation module 403.

[0134] Specifically, the field data acquisition module 401 is implemented according to the method described in step S110 above, configured to acquire real-time logging data from the drilling site and convert it into well depth logging data; the formation feature prediction module 402 is implemented according to the method described in step S120 above, configured to obtain the formation feature parameters of the formation to be drilled based on the well depth logging data and using a pre-built formation feature parameter prediction model; the optimal drilling control parameter generation module 403 is implemented according to the method described in step S130 above, configured to solve the pre-built drilling pressure and rotational speed target optimization model based on the formation feature parameters of the formation to be drilled and the field correlation parameters of the drilling operation, to obtain the optimal drilling control parameters of the formation to be drilled.

[0135] This invention discloses a method and system for optimizing drilling control parameters. The method and system acquire real-time logging data and convert it into depth logging data, reducing the requirements for on-site data acquisition. Furthermore, it employs a formation characteristic parameter prediction model to predict the geological characteristics of the formation to be drilled, and calculates the corresponding drilling control parameters based on the predicted geological characteristics, effectively overcoming the uncertainties, nonlinearities, and parallel processes in the drilling process. This invention can be effectively applied to drilling operations, improving drilling efficiency and reducing drilling costs.

[0136] The above description is merely a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included 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.

[0137] In the description of this invention, unless otherwise stated, "a plurality of" means two or more; the terms "upper," "lower," "left," "right," "inner," "outer," "front end," "rear end," "head," "tail," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, the terms "first," "second," "third," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0138] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "connected" and "linked" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0139] It should be understood that the embodiments disclosed herein are not limited to the specific structures, processing steps, or materials disclosed herein, but should be extended to equivalent substitutions of these features as understood by those skilled in the art. It should also be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting.

[0140] The phrase "an embodiment" or "an embodiment" used in this specification means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the invention. Therefore, the phrase "an embodiment" or "an embodiment" appearing in various places throughout the specification does not necessarily refer to the same embodiment.

[0141] The embodiments of the present invention are given for illustrative and descriptive purposes only, and are not intended to be exhaustive or to limit the invention to the forms disclosed. Many modifications and variations will be apparent to those skilled in the art. The embodiments were chosen and described in order to better illustrate the principles and practical application of the invention, and to enable those skilled in the art to understand the invention and to design various embodiments with various modifications suitable for a particular purpose.

[0142] While the embodiments disclosed in this invention are as described above, the content is merely for the purpose of facilitating understanding of the invention and is not intended to limit the invention. Any person skilled in the art to which this invention pertains may make any modifications and changes in form and detail of the implementation without departing from the spirit and scope disclosed herein; however, the scope of patent protection of this invention shall still be determined by the scope defined in the appended claims.

Claims

1. A method for optimizing drilling control parameters, characterized in that, include: Acquire real-time logging data from the drilling site and convert it into well depth logging data; Based on the well depth logging data, the formation characteristic parameters of the formation to be drilled are obtained using a pre-constructed formation characteristic parameter prediction model. Based on the formation characteristic parameters of the formation to be drilled and the field correlation parameters of the drilling operation, the pre-constructed drilling pressure and rotation speed target optimization model is solved to obtain the optimal drilling control parameters of the formation to be drilled.

2. The method according to claim 1, characterized in that, The well depth logging data includes, but is not limited to, well depth, drilling speed, drilling pressure, rotational speed, torque, standpipe pressure, displacement, drilling fluid density, and drill bit diameter. The step of converting the real-time logging data into the well depth logging data includes: The real-time logging data and the drill bit information during drilling are subjected to multi-source data fusion processing to obtain the logging data to be processed; Delete invalid logging data from the logging data to be processed; Using well depth as an index, the unprocessed logging data after deleting invalid logging data is resampled to obtain the well depth logging data.

3. The method according to claim 2, characterized in that, The invalid logging data includes, but is not limited to: abnormal logging data, missing logging data, and logging data under non-drilling conditions. The step of deleting invalid logging data from the logging data to be processed includes: identifying abnormal logging data and missing logging data, including: Preset sliding window and fitting method; According to the fitting method, the logging data captured in the sliding window during the window sliding process is fitted to identify abnormal logging data and missing logging data in the logging data to be processed.

4. The method according to any one of claims 1 to 3, characterized in that, The formation characteristic parameter prediction model is constructed through the following steps: A training dataset was constructed based on historical drilling data from multiple drilled wells. Based on the training dataset, the preset model to be trained is trained to obtain a formation feature parameter prediction model. The historical drilling data of the drilled wells includes: well depth logging data and corresponding formation characteristic parameters; The model to be trained is a long short-term memory recurrent neural network model.

5. The method according to claim 4, characterized in that, The formation characteristic parameters of the formation to be drilled include: the formation drillability index, wherein the formation characteristic parameter prediction model is represented by the following expression: K d =f(H,v pc ,W,N,T,p m ,Q,ρ,d,A) Among them, K d The formation drillability index is represented by H, which represents the current well depth, and v. pc W represents drilling speed, N represents drilling pressure, T represents drilling speed, and p represents torque. m Q represents riser pressure, ρ represents drilling fluid density, d represents drill bit diameter, and A represents drill bit type.

6. The method according to claim 4 or 5, characterized in that, To obtain historical drilling data for multiple drilled wells, follow these steps: Based on the comprehensive logging data of multiple drilled wells, the daily drilling and completion construction data, and the drilling and completion report, multi-source data fusion processing based on different logging times is performed to obtain logging feature data; Delete invalid logging data from the logging feature data; Using well depth as an index, the logging feature data after deleting invalid logging data is resampled to obtain the well depth logging data for each drilled well; Based on the logging characteristic data, the formation characteristic parameters of each drilled well are calculated.

7. The method according to any one of claims 1 to 6, characterized in that, The target optimization model for drilling pressure and rotational speed is constructed through the following steps: Based on the drilling control parameters of the formation to be drilled, construct optimization variables; Based on the aforementioned optimization variables, an objective function is constructed with the goal of maximizing the drilling speed of the drilling machinery and minimizing the drill bit wear rate during the drilling operation. Constraints are constructed based on the range of drilling pressure, rotational speed, and drill bit tooth wear during the drilling operation. Based on the objective function and the constraints, an optimization model for the drilling pressure rotation speed target is constructed.

8. The method according to claim 7, characterized in that, The objective function is expressed by the following expression: Where F1 represents the drilling speed of the drilling machinery, K d C represents the formation drillability index. p C represents the pressure difference influence coefficient. H The hydraulic purification coefficient is represented by W, the drilling pressure by M, the threshold drilling pressure by N, the rotational speed by λ, the rotational speed exponent by C2, the drill bit tooth wear coefficient by h, the amount of drill bit tooth wear by F2, and the drill bit wear rate by A. f The coefficients represent rock abrasiveness, a1 and a2 are both rotation speed influence coefficients, N represents rotation speed, Z2 and Z1 are both drilling pressure influence coefficients, W represents drilling pressure, C1 represents tooth wear reduction coefficient, and h represents drill bit tooth wear amount. The constraint condition is expressed using the following expression: Where M represents the threshold drill pressure, W represents the drill pressure, and W max N represents the maximum drilling pressure. min N represents the minimum rotational speed, and N represents the rotational speed. max The value of represents the maximum rotational speed, and h represents the amount of wear on the drill bit teeth.

9. The method according to claim 7 or 8, characterized in that, The optimal drilling control parameters for the formation to be drilled are the optimal combination of drilling pressure and rotational speed. A genetic algorithm is used to solve the optimization model for the drilling pressure and rotational speed targets, which includes: S1: Set the population size and maximum number of iterations; S2: Initialize the number of iterations and perform population initialization operations based on optimization variables and constraints; S3: Perform a fast non-dominated sort on the individuals in the initial population, and then perform selection, crossover and mutation operations to obtain the parent population; S4: Increment the iteration count by 1, perform selection, crossover, and mutation operations on individuals in the parent population to obtain the offspring population; S5: Merge the parent and offspring populations, and perform fast non-dominated sorting and crowding calculation on the individuals in the merged population to obtain a new parent population. S6: Determine if the number of iterations is less than the maximum number of iterations. If yes, proceed to S7. If no, use the new parent population as the parent population and return to S4. S7: Calculate the fitness value of each individual in the new parent population based on the objective function, and take the optimization variable corresponding to the individual with the largest fitness value as the optimal drilling control parameter.

10. A system for optimizing drilling control parameters, characterized in that, include: The field data acquisition module is configured to acquire real-time logging data from the drilling site and convert it into well depth logging data. The formation feature prediction module is configured to obtain the formation feature parameters of the formation to be drilled by using a pre-built formation feature parameter prediction model based on the well depth logging data. The optimal drilling control parameter generation module is configured to solve a pre-built target optimization model for drilling pressure and rotational speed based on the formation characteristic parameters of the formation to be drilled and the field correlation parameters of the drilling operation, so as to obtain the optimal drilling control parameters for the formation to be drilled.

Images