An interpretable drilling parameter intelligent optimization method and device

Abstract

This specification provides an interpretable intelligent optimization method and apparatus for drilling parameters. The method includes: acquiring first drilling parameters for a first target well; training a neural network model based on the first drilling parameters to obtain a drilling speed prediction model; determining a drilling speed prediction equation; acquiring second drilling parameters for a second target well; inputting the second drilling parameters into the drilling speed prediction equation to obtain a target drilling speed; determining the influence value of the second drilling parameters on the target drilling speed according to the weight parameters in the drilling speed prediction equation; comparing the influence value with a preset threshold to determine adjustable parameters within a preset parameter range in the second drilling parameters; determining the target drilling parameters and the corresponding second target drilling speed value based on the preset parameter range and the drilling speed prediction equation; and optimizing the drilling process of the second target well based on the target drilling parameters. This method can effectively optimize drilling parameters and improve the efficiency of oil exploration and development.

Description

Technical Field

[0001] This manual pertains to the field of petroleum exploration and development, and in particular relates to an interpretable intelligent optimization method and apparatus for drilling parameters. Background Technology

[0002] Drilling parameter optimization typically studies the impact of various drilling parameters on the rate of penetration (RLP) under certain baseline conditions. Optimizing drilling parameters can effectively improve drilling efficiency. However, existing methods suffer from uncontrollable factors in RLP prediction models, leading to unpredictable prediction results and hindering effective optimization of drilling parameters, ultimately resulting in reduced efficiency in oil and gas exploration and development.

[0003] There is currently no effective solution to the above problems. Summary of the Invention

[0004] This specification provides an interpretable intelligent optimization method and apparatus for drilling parameters, which can effectively optimize drilling parameters, improve drilling efficiency, and reduce drilling costs.

[0005] On the one hand, the embodiments of this specification provide an interpretable intelligent optimization method for drilling parameters, including:

[0006] First drilling parameters of the first target well are obtained, and a fully connected neural network model is trained based on the first drilling parameters to obtain a drilling rate prediction model; wherein, the drilling rate prediction model adopts a target activation function;

[0007] The drilling rate prediction model is processed according to the target activation function and the first drilling parameters to obtain the drilling rate prediction equation;

[0008] Obtain the second drilling parameters of the second target well, and input the second drilling parameters into the drilling rate prediction equation to obtain the target drilling rate;

[0009] Based on the weighted parameters in the drilling speed prediction equation, determine the influence value of the second drilling parameter on the target drilling speed;

[0010] The influence value is compared with a preset threshold. When the influence value is determined to be greater than the preset threshold, an adjustable parameter in the second drilling parameters is determined; wherein the adjustable parameter is within the preset parameter range.

[0011] Based on the preset parameter range and the drilling rate prediction equation, the target drilling parameters and the corresponding second target drilling rate value are determined; wherein, the second target drilling rate value is greater than the preset drilling rate threshold.

[0012] The drilling process of the second target well is optimized based on the target drilling parameters.

[0013] Furthermore, before obtaining the first drilling parameters of the first target well, the process further includes:

[0014] Obtain the initial drilling parameters for the first target well;

[0015] The initial drilling parameters are cleaned and normalized to obtain the first drilling parameters of the first target well.

[0016] Furthermore, the first drilling parameters include one or more of the following: rotational speed, drilling pressure, torque, riser pressure, inlet flow rate, hook load, drilling fluid density, drilling fluid viscosity, drilling fluid plastic viscosity, and yield strength.

[0017] Furthermore, the drilling rate prediction equation is constructed according to the following formula:

[0018] ROP=A×RP+B×WOB+C×Tor+D×SPP+E×IF+F×HL+G×Den-H×Vis-I×PV+J×YP

[0019] Where RP is rotational speed, WOB is drilling pressure, Tor is torque, SPP is riser pressure, IF is inlet flow rate, HL is hook load, Den is drilling fluid density, Vis is drilling fluid viscosity, PV is drilling fluid plastic viscosity, YP is yield strength, and A, B, C, D, E, F, G, H, I, and J are the weighting parameters corresponding to each drilling parameter.

[0020] Furthermore, based on the weighted parameters in the drilling speed prediction equation, the influence of the second drilling parameter on the target drilling speed is determined, including:

[0021] Determine the average value of each drilling parameter in the second drilling parameters;

[0022] Each weight parameter is multiplied by the average value to obtain a product set.

[0023] The influence of the second drilling parameter on the target drilling rate is determined based on the comparison results by comparing each product value in the product set.

[0024] Further, determining the target drilling parameters and the corresponding second target drilling rate value based on the preset parameter range and the drilling rate prediction equation includes:

[0025] The adjustable parameters are adjusted according to the preset parameter range and preset adjustment rules to obtain multiple sets of adjusted drilling parameters;

[0026] The multiple sets of drilling parameters are sequentially input into the drilling rate prediction equation to obtain multiple sets of drilling rate values;

[0027] The drilling parameters are compared with the preset drilling speed thresholds to determine the target drilling parameter and the second target drilling speed value corresponding to the target drilling parameter.

[0028] Furthermore, the optimization process for drilling the second target well based on the target drilling parameters includes:

[0029] The drilling speed increase value is determined based on the target drilling speed and the second target drilling speed value;

[0030] Based on the drilling speed improvement value and the target drilling parameters, a recommended plan is generated and the recommended plan is pushed to the drilling equipment in real time.

[0031] The drilling equipment receives the recommended plan and optimizes the drilling process of the second target well based on the recommended plan.

[0032] On the other hand, embodiments of this specification also provide an interpretable intelligent drilling parameter optimization device, including:

[0033] The data management module is used to obtain the first drilling parameters of the first target well;

[0034] The drilling speed prediction model module is used to train a fully connected neural network model based on the first drilling parameters to obtain a drilling speed prediction model; wherein, the drilling speed prediction model adopts a target activation function;

[0035] The drilling speed prediction interpretation module is used to process the drilling speed prediction model according to the target activation function and the first drilling parameters to obtain the drilling speed prediction equation; obtain the second drilling parameters of the second target well, input the second drilling parameters into the drilling speed prediction equation to obtain the target drilling speed; and determine the influence value of the second drilling parameters on the target drilling speed according to the weight parameters in the drilling speed prediction equation.

[0036] The drilling parameter optimization module is used to compare the influence value with a preset threshold, and when the influence value is determined to be greater than the preset threshold, to determine an adjustable parameter in the second drilling parameters; wherein the adjustable parameter is within the preset parameter range; and to determine the target drilling parameter and the second target drilling speed value corresponding to the target drilling parameter according to the preset parameter range and the drilling speed prediction equation; wherein the second target drilling speed value is greater than the preset drilling speed threshold.

[0037] The optimization processing module is used to optimize the drilling process of the second target well based on the target drilling parameters.

[0038] Furthermore, the device also includes:

[0039] The cleaning module is used to obtain the initial drilling parameters of the first target well; and to perform data cleaning and data normalization processing on the initial drilling parameters to obtain the first drilling parameters of the first target well.

[0040] In another aspect, this application also provides a computer-readable storage medium storing computer instructions thereon, wherein the computer-readable storage medium implements the above-described interpretable intelligent optimization method for drilling parameters when the instructions are executed.

[0041] This specification provides an interpretable intelligent optimization method and apparatus for drilling parameters. First, first drilling parameters of a first target well are obtained, and a fully connected neural network model is trained based on these parameters to obtain a drilling speed prediction model. The drilling speed prediction model employs a target activation function. The drilling speed prediction model is processed according to the target activation function and the first drilling parameters to obtain a drilling speed prediction equation. Second, second drilling parameters of a second target well are obtained and input into the drilling speed prediction equation to obtain a target drilling speed. The influence value of the second drilling parameters on the target drilling speed is determined based on the weight parameters in the drilling speed prediction equation. Further, the influence value is compared with a preset threshold. When the influence value is greater than the preset threshold, an adjustable parameter in the second drilling parameters is determined. The adjustable parameter is within a preset parameter range. Based on the preset parameter range and the drilling speed prediction equation, the target drilling parameters and the corresponding second target drilling speed value are determined. The second target drilling speed value is greater than the preset drilling speed threshold. Finally, the drilling process of the second target well is optimized based on the target drilling parameters. The above scheme can make the drilling rate prediction results controllable, effectively optimize drilling parameters, improve the efficiency of oil exploration and development, and reduce drilling costs. Attached Figure Description

[0042] To more clearly illustrate the embodiments of this specification, the accompanying drawings used in the embodiments will be briefly introduced below. The drawings described below are only some embodiments recorded in this specification. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0043] Figure 1 This is a flowchart illustrating an embodiment of the intelligent optimization method for drilling parameters provided in this specification.

[0044] Figure 2 This is a schematic diagram illustrating one embodiment of the intelligent optimization method for interpretable drilling parameters provided in the embodiments of this specification, within a scenario example.

[0045] Figure 3This is a schematic diagram illustrating one embodiment of the intelligent optimization method for interpretable drilling parameters provided in the embodiments of this specification, within a scenario example.

[0046] Figure 4 This is a schematic diagram of the structural composition of an interpretable intelligent drilling parameter optimization device provided in one embodiment of this specification;

[0047] Figure 5 This is a schematic diagram of the structural composition of an electronic device provided in one embodiment of this specification. Detailed Implementation

[0048] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this specification, and not all embodiments. Based on the embodiments in this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this specification.

[0049] Drilling parameter optimization typically studies the impact of various drilling parameters on the rate of penetration (RLP) under certain baseline conditions. Optimizing drilling parameters can effectively improve drilling efficiency. However, existing methods suffer from uncontrollable factors in RLP prediction models, leading to unpredictable prediction results and hindering effective drilling parameter optimization, ultimately reducing the efficiency of oil exploration and development.

[0050] Based on the above approach, this specification proposes an interpretable intelligent optimization method for drilling parameters. First, the first drilling parameters of a first target well are obtained, and a fully connected neural network model is trained based on these parameters to obtain a drilling speed prediction model. The drilling speed prediction model employs a target activation function. The drilling speed prediction model is processed according to the target activation function and the first drilling parameters to obtain a drilling speed prediction equation. Second, the second drilling parameters of a second target well are obtained and input into the drilling speed prediction equation to obtain a target drilling speed. The influence value of the second drilling parameters on the target drilling speed is determined based on the weight parameters in the drilling speed prediction equation. Further, the influence value is compared with a preset threshold. When the influence value is greater than the preset threshold, an adjustable parameter in the second drilling parameters is determined. The adjustable parameter is within a preset parameter range. Based on the preset parameter range and the drilling speed prediction equation, the target drilling parameters and the corresponding second target drilling speed value are determined. The second target drilling speed value is greater than the preset drilling speed threshold. Finally, the drilling process of the second target well is optimized based on the target drilling parameters. See Figure 1As shown in the embodiments of this specification, an interpretable intelligent optimization method for drilling parameters is provided. In specific implementation, this method may include the following:

[0051] S101: Obtain the first drilling parameters of the first target well, and train the fully connected neural network model based on the first drilling parameters to obtain the drilling speed prediction model; wherein the drilling speed prediction model adopts a target activation function.

[0052] S102: Process the drilling rate prediction model according to the target activation function and the first drilling parameters to obtain the drilling rate prediction equation.

[0053] S103: Obtain the second drilling parameters of the second target well, input the second drilling parameters into the drilling rate prediction equation, and obtain the target drilling rate.

[0054] S104: Determine the influence value of the second drilling parameter on the target drilling rate based on the weight parameters in the drilling rate prediction equation.

[0055] S105: Compare the influence value with a preset threshold, and when it is determined that the influence value is greater than the preset threshold, determine the adjustable parameter in the second drilling parameters; wherein the adjustable parameter is within the preset parameter range.

[0056] S106: Based on the preset parameter range and the drilling rate prediction equation, determine the target drilling parameters and the corresponding second target drilling rate value; wherein the second target drilling rate value is greater than a preset drilling rate threshold.

[0057] S107: Optimize the drilling process of the second target well based on the target drilling parameters.

[0058] In some embodiments, prior to obtaining the first drilling parameters of the first target well, the specific implementation may include:

[0059] S1: Obtain the initial drilling parameters for the first target well;

[0060] S2: Perform data cleaning and normalization on the initial drilling parameters to obtain the first drilling parameters of the first target well.

[0061] In some embodiments, the first target well can be an already drilled well, and the second target well can be a well to be optimized. The first target well may contain first drilling parameters, and the second target well may contain second drilling parameters. The first drilling parameters are used to train a fully connected neural network model to obtain a drilling rate prediction model. It should be noted that the first drilling parameters are parameters that have undergone data cleaning, data normalization, or standardization. Data cleaning refers to removing outliers, null values, invalid values, contradictory values, etc., from the initial drilling parameters. Standardization refers to normalizing the parameters due to significant differences in their values. The second drilling parameters can also be data that has undergone data cleaning and normalization. By performing data cleaning and normalization on the drilling parameters, the training time of the drilling rate prediction model can be effectively reduced, and the prediction accuracy of the drilling rate prediction model can be improved.

[0062] In some embodiments, the first drilling parameter may include one or more of the following: rotational speed, drilling pressure, torque, standpipe pressure, inlet flow rate, hook load, drilling fluid density, drilling fluid viscosity, drilling fluid plastic viscosity, and yield strength. It should be noted that the second drilling parameter may also include one or more of the following: rotational speed, drilling pressure, torque, standpipe pressure, inlet flow rate, hook load, drilling fluid density, drilling fluid viscosity, drilling fluid plastic viscosity, and yield strength.

[0063] In some embodiments, the aforementioned rotational speed, drilling pressure, torque, riser pressure, inlet flow rate, and hook load are engineering data, while the aforementioned drilling fluid density, drilling fluid viscosity, drilling fluid plastic viscosity, and yield strength are drilling fluid data. These data constitute drilling parameters. It should be noted that drilling parameters are not limited to the examples above. Those skilled in the art, inspired by the technical essence of the embodiments in this specification, may make other modifications. However, as long as the achieved function and effect are the same as or similar to the embodiments in this specification, they should be covered within the protection scope of the embodiments in this specification. For example, depending on actual needs, drilling parameters may also include: drill bit size, number of drill bit teeth, acoustic logging, gamma-ray logging, etc. Obtaining drilling parameters lays the data foundation for further optimization of drilling parameters. The optimization of drilling parameters will be explained separately later, and will not be repeated here.

[0064] In some embodiments, the drilling speed prediction equation can be calculated as follows:

[0065] ROP=A×RP+B×WOB+C×Tor+D×SPP+E×IF+F×HL+G×Den-H×Vis-I×PV+J×YP

[0066] Where RP is rotational speed, WOB is drilling pressure, Tor is torque, SPP is riser pressure, IF is inlet flow rate, HL is hook load, Den is drilling fluid density, Vis is drilling fluid viscosity, PV is drilling fluid plastic viscosity, YP is yield strength, and A, B, C, D, E, F, G, H, I, and J are the weighting parameters corresponding to each drilling parameter.

[0067] In some embodiments, the drilling speed prediction model can be processed according to the target activation function and the first drilling parameters to obtain the drilling speed prediction equation, wherein the target activation function can be the ReLU function. The first drilling speed prediction model can be obtained by training a fully connected neural network (FNN) model according to the first drilling parameters. For example, the acquired first drilling parameters can be divided, with 70% of the data divided into a training set and 30% into a test set. The training set can contain multiple drilling sample parameters. It should be noted that the division of the training set and the test set is not limited to the above examples. Those skilled in the art may make other changes based on the technical essence of the embodiments in this specification, but as long as the functions and effects achieved are the same as or similar to those in the embodiments of this specification, they should be covered within the protection scope of the embodiments of this specification. Secondly, the fully connected neural network prediction model is trained based on the drilling sample parameters mentioned above. Based on the frequency distribution of the activation function during the training of the prediction model, the hyperparameters of the prediction model are optimized. The drilling sample parameters are then input into the optimized prediction model until the number of training iterations reaches a preset threshold or the prediction accuracy of the test set reaches a certain level, such as 85%. At this point, the training ends, and the optimized model obtained after training is used as the drilling speed prediction model.

[0068] The FNN model described above can have at least one hidden layer, and when the data is complex, the number of hidden layers can even exceed 10. Taking the calculation process of a three-layer neural network as an example, it is shown in formula (1). The calculation of each neuron in the hidden layer mainly includes two parts: linear combination and activation calculation. In addition, the output layer usually does not perform activation calculation.

[0069]

[0070] Where, x i Given i input data points, y i 1 For the data corresponding to the input layer, y i 2 For the data corresponding to the hidden layer, y 3 The data output by the output layer. It is the weight of the output of the first layer neuron k when the second layer neurons i are linearly combined. It's a deviation. It is an activation function.

[0071] The activation functions of fully connected neural networks can be divided into two main categories: saturated activation functions and non-saturated activation functions, as shown in Equations (2) and (3). Equation (2) is the corresponding formula for the hyperbolic function (tanh) in saturated activation functions, and Equation (3) is the corresponding formula for the modified linear unit (ReLU) in non-saturated activation functions.

[0072]

[0073] ReLU(x) = max(0,x) (3)

[0074] The most important function of the activation function is to transform the linear combination of neurons into a nonlinear computation, thus enabling the fitting of complex data. Compared to the tanh activation function, the ReLU function has the advantage of fast convergence and is less prone to gradient vanishing. The aforementioned drilling speed prediction model uses the ReLU function as its activation function.

[0075] The hyperparameters mentioned above may include: the number of neural network layers, the number of neurons per layer, the learning rate, and the optimizer, etc. It should be noted that the hyperparameters are not limited to the examples above. Those skilled in the art may make other changes based on the technical essence of the embodiments in this specification. However, as long as the functions and effects achieved are the same as or similar to those in the embodiments of this specification, they should be covered within the protection scope of the embodiments of this specification. Since the training process of a neural network is a backpropagation process of error, the parameters of the model will become more and more stable as training continues, and the activation function matrix of the model will also gradually tend to stabilize. However, more training times are not necessarily better. Continuous training may lead to overfitting of the model, reducing the stability and transferability of the model. By combining the activation frequency distribution during the training of the prediction model, it is possible to determine whether the prediction model has converged, is underfitting, or is overfitting, so that the above hyperparameters can be adjusted in a timely manner to achieve the best prediction effect.

[0076] The convergence of the aforementioned prediction model can be evaluated using the mean relative error (MSE) to assess the degree of data variation. A smaller MSE value indicates better accuracy in describing the experimental data. For example, if the mean relative error (MRE) reaches 10%, it indicates that the activation frequency distribution is gradually stabilizing, indicating good convergence and high prediction accuracy. If the MRE exceeds 10%, it indicates that the activation frequency distribution is unstable and the convergence is poor. In this case, it is necessary to adjust the hyperparameters of the neural network to improve the prediction accuracy. It should be noted that the value of the mean relative error is not limited to the examples above. Those skilled in the art may make other modifications based on the technical essence of the embodiments in this specification. However, as long as the achieved function and effect are the same as or similar to the embodiments in this specification, they should be covered within the protection scope of the embodiments in this specification.

[0077] Finally, by converting the complex nonlinear function changes in neurons into matrix operations, such as converting it into Y = wx + b (w is the weight parameter, x is each drilling data, and b is the bias term), we can intuitively see why the drilling rate prediction model calculates these drilling rate values, and also intuitively show the magnitude of the impact of each drilling data on ROP.

[0078] Specifically, to further simplify formula (1), x, w, and b are represented by matrices, as shown in the following formula:

[0079]

[0080] Formula 1 can be rewritten in matrix form:

[0081]

[0082] The derivation of the ReLU function is shown in formula (6). When x is greater than 0, DR(x) is 1; when x is less than 0, DR(x) is 0. The calculation result of DR(x) is recorded as matrix DR.

[0083]

[0084] Here, ReLU is the activation function, and DR is the activation matrix of the neurons in the current layer.

[0085] ReLU(x) and DR(x) satisfy the following relationship:

[0086]

[0087] According to equation (7), the calculation process of ReLU can be transformed from function operation to matrix calculation. The DR matrix consists of 0 and 1. If 0 indicates that the neuron is not activated, and 1 indicates that the neuron is activated. Therefore, when the activation function of the FNN model is ReLU, equation (5) can be rewritten as:

[0088] y = (x * w 2 +b 2 )*DR(x*w 2 +b 2 )*w 3 +b 3 (8)

[0089] Based on the combination rules and left distribution rules of matrix operations, equation (8) can be simplified to:

[0090] y = w*x + b(9)

[0091] In equation (9), w is w 2 *DR(x*w 2 +b 2 )*w 3 b is b 2 *DR(x*w 2 +b 2 )*w 3 +b 3

[0092] The above changes represent this complex nonlinear function operation in the form of a linear regression equation.

[0093] The DR is a matrix containing only (0, 1), where 1 represents an activated neuron and 0 represents an unactivated neuron. By counting the 0s and 1s in each column of the DR matrix, we can know the activation frequency of each neuron in the current layer.

[0094] The DR matrix transforms the function operation of a neural network with ReLU activation into matrix operation and simplifies it into a linear combination of input parameters. This enables the extraction and linear representation of the weights of the neural network with ReLU activation, greatly improving the reliability of the intelligent drilling speed prediction model.

[0095] In some embodiments, each drilling parameter in the first drilling parameters has a corresponding weight parameter, such as: rotational speed corresponding to weight parameter A, drilling pressure corresponding to weight parameter B, torque corresponding to weight parameter C, etc. Substituting each drilling parameter and its corresponding weight parameter into formula (9) yields the above-mentioned drilling speed prediction equation.

[0096] In some embodiments, obtaining the drilling rate prediction equation can break the "black box" uncontrollability of the intelligent prediction model, enabling real-time understanding of the model's calculation process and results, and improving the reliability of the model's prediction results.

[0097] In some embodiments, determining the influence value of the second drilling parameter on the target drilling rate based on each weight parameter in the drilling rate prediction equation may, in specific implementation, include:

[0098] S1: Determine the average value of each drilling parameter in the second drilling parameters;

[0099] S2: Multiply each of the weight parameters with the average value to obtain a product set;

[0100] S3: Compare the product values ​​in the product set to determine the impact of the second drilling parameter on the target drilling rate based on the comparison results.

[0101] In some embodiments, determining the target drilling parameters and the second target drilling rate value corresponding to the target drilling parameters based on the preset parameter range and the drilling rate prediction equation may, in specific implementation, include:

[0102] S1: Adjust the adjustable parameters according to the preset parameter range and preset adjustment rules to obtain multiple sets of adjusted drilling parameters;

[0103] S2: Input the multiple sets of drilling parameters into the drilling rate prediction equation in sequence to obtain multiple sets of drilling rate values;

[0104] S3: Compare the multiple drilling rates corresponding to multiple sets of drilling parameters with the preset drilling rate threshold to determine the target drilling parameter among the multiple sets of drilling parameters and the second target drilling rate value corresponding to the target drilling parameter among the multiple sets of drilling rates.

[0105] In some embodiments, the above-described optimization of the drilling process of the second target well based on the target drilling parameters may, in specific implementation, include:

[0106] S1: Determine the drilling speed increase value based on the target drilling speed and the second target drilling speed value;

[0107] S2: Based on the drilling speed improvement value and the target drilling parameters, a recommended plan is generated and the recommended plan is pushed to the drilling equipment in real time;

[0108] S3: The drilling equipment receives the recommended plan and optimizes the drilling process of the second target well based on the recommended plan.

[0109] In some embodiments, the product set includes the product of the average value of multiple drilling parameters and the weight parameter corresponding to the drilling parameter. The determination of the influence value of the second drilling parameter on the target drilling rate based on the comparison results can take the maximum value in the comparison results as the maximum value of the influence value of the second drilling parameter on the target drilling rate; that is, the larger the product of the average value of the drilling parameter and the weight parameter corresponding to the drilling parameter, the greater the influence of the second drilling parameter on the target drilling rate.

[0110] In some embodiments, the above-mentioned preset parameter range can be determined based on economic benefit indicators and engineering difficulty. It should be noted that the method of determining the above-mentioned preset parameter range is not limited to the examples above. Those skilled in the art may make other changes under the guidance of the technical essence of the embodiments of this specification. However, as long as the functions and effects achieved are the same as or similar to those of the embodiments of this specification, they should all be covered within the protection scope of the embodiments of this specification.

[0111] In some embodiments, the preset adjustment rule can be to sequentially increase the adjustable parameters by one unit. For example, if the drilling pressure data is 100, according to the preset adjustment rule, the drilling pressure can be adjusted to 101. At this time, other drilling parameters in the adjustable parameters will also be increased by one unit accordingly, forming the first set of adjusted drilling parameters. Similarly, if the drilling pressure is adjusted from 101 to 102, other drilling parameters in the adjustable parameters will also be increased by one unit accordingly, forming the second set of adjusted drilling parameters. When the maximum limit in the adjustment range is reached, the adjustment of the drilling parameters to be optimized is stopped. It should be noted that the adjustable parameters can be adjustable data, and the adjustable parameters are not limited to the examples above. Those skilled in the art may make other changes based on the technical essence of the embodiments in this specification, but as long as the functions and effects achieved are the same as or similar to those in the embodiments of this specification, they should be covered within the protection scope of the embodiments of this specification.

[0112] In some embodiments, after determining the adjusted sets of drilling parameters, these parameters are sequentially input into the rate of drilling (RWD) prediction equation to obtain multiple RWD values. The RWD values ​​corresponding to these parameters are then compared with preset RWD thresholds to determine the maximum RWD. This maximum RWD value is used as the second target RWD value, and the drilling parameters corresponding to this second target RWD value are used as the target drilling parameters. By obtaining the second target RWD value, drilling at the maximum drilling rate can be ensured, effectively improving drilling efficiency and reducing drilling costs.

[0113] In some embodiments, the aforementioned increase in drilling speed can represent the amount by which the drilling speed has increased. For example, if the target drilling speed is 100 and the second target drilling speed is 150, then the drilling speed increase value is 50, which can represent an increase of 50 in drilling speed. The aforementioned increase in drilling speed and target drilling parameters are sent to the drilling equipment as a recommended scheme. The drilling equipment can optimize the drilling process based on this recommended scheme, thereby improving the efficiency of oil and gas exploration and development.

[0114] The above method will be described below with reference to a specific embodiment. However, it is worth noting that this specific embodiment is only for better illustration of this application and does not constitute an improper limitation of this application.

[0115] Before implementation, firstly, the initial drilling parameters of the first target well are obtained; secondly, the initial drilling parameters are cleaned and normalized to obtain the first drilling parameters of the first target well, such as: rotational speed, drilling pressure, torque, riser pressure, inlet flow rate, hook load, drilling fluid density, drilling fluid viscosity, drilling fluid plastic viscosity, and yield strength; further, the first drilling parameters are divided into training and testing sets to train and validate the fully connected neural network model, and the hyperparameters of the prediction model are optimized based on the activation function frequency distribution during training. The sample data is then further input... The training is continued until the number of training iterations reaches a preset threshold or the prediction accuracy on the test set reaches a certain level. Then, the training ends, and the optimized model obtained after training is used as the drilling speed prediction model. The drilling speed prediction model uses the ReLU activation function. Finally, by converting the complex nonlinear function changes in neurons into matrix operations, such as converting weight parameters, feature parameters, and bias terms into matrix form, the calculation process of the ReLU function can be transformed from function operation to matrix calculation. Finally, this complex nonlinear function operation is presented in the form of a linear regression equation, which yields the drilling speed prediction equation. In specific implementation, firstly, the second drilling parameters of the second target well are obtained, and these parameters are input into the drilling speed prediction equation to obtain the target drilling speed. Secondly, the average value of each drilling parameter in the second drilling parameters is determined, and each weighted parameter is multiplied by the average value to obtain a product set. The product values ​​in the product set are compared to determine the influence value of the second drilling parameters on the target drilling speed based on the comparison results. Further, the influence value is compared with a preset threshold. When the influence value is determined to be greater than the preset threshold, an adjustable parameter in the second drilling parameters is determined. The adjustable parameters are within a preset parameter range. Finally, the adjustable parameters are adjusted according to the preset parameter range and preset adjustment rules to obtain multiple sets of adjusted drilling parameters. These multiple sets of drilling parameters are sequentially input into the drilling speed prediction equation to obtain multiple sets of drilling speed values. The drilling speeds corresponding to these multiple sets of drilling parameters are compared with preset drilling speed thresholds to determine the target drilling parameters and the second target drilling speed value corresponding to the target drilling parameters. The second target drilling speed value is greater than the preset drilling speed threshold. The drilling process of the second target well is optimized based on the target drilling parameters. A drilling speed enhancement value is determined based on the target drilling speed and the second target drilling speed value. A recommended scheme is formed based on the drilling speed enhancement value and the target drilling parameters, and the recommended scheme is pushed to the drilling equipment in real time. The drilling equipment receives the recommended scheme and optimizes the drilling process of the second target well based on the recommended scheme.The above methods can improve the controllability and accuracy of drilling rate prediction, and determine the impact of various drilling parameters on drilling rate. This allows for targeted adjustment of drilling parameters, thereby improving drilling efficiency and reducing drilling costs.

[0116] In a specific scenario example, the interpretable intelligent optimization method for drilling parameters provided in the embodiments of this specification can be applied to improve drilling efficiency and reduce drilling costs. For specific implementation details, please refer to [the relevant documentation / reference]. Figure 2 As shown:

[0117] Data Input and Feature Selection: Drilling parameters are collected from the well site, such as: rotational speed (RP), weight on bit (WOB), torque (Tor), standpipe pressure (SPP), inlet flow rate (IF), hook load (HL), drilling fluid density (Den), drilling fluid viscosity (Vis), drilling fluid plastic viscosity (PV), and yield strength (YP). Bit size (BS), number of teeth (BF), acoustic logging (AC), gamma-ray logging (GR), etc., are also collected. While acquiring this data, the well to be optimized is selected on-site, and then the parameters required for predicting ROP (i.e., feature selection) are chosen. In this example, the selected parameters are: rotational speed (RP), weight on bit (WOB), torque (Tor), standpipe pressure (SPP), inlet flow rate (IF), hook load (HL), drilling fluid density (Den), drilling fluid viscosity (Vis), drilling fluid plastic viscosity (PV), and yield strength (YP).

[0118] Data processing: Drilling parameters are processed to remove null values ​​and outliers, and the drilling parameters are normalized to improve the accuracy of drilling rate prediction model training.

[0119] FNN hyperparameter selection: For example, you can choose from the number of neural network layers, the number of neurons per layer, the learning rate, and the optimizer.

[0120] FNN drilling rate prediction model training: The FNN model is trained using drilling sample parameters. 70% of the sample parameters can be divided into the training set and 30% into the test set.

[0121] Neuron activation interpretation, hyperparameter adjustment, and obtaining an optimized drilling speed prediction model: Based on the frequency distribution of activation functions during training, the prediction model is optimized by hyperparameters. Sample parameters are then input into the optimized prediction model until the number of training iterations reaches a preset threshold or the prediction accuracy of the test set reaches a certain level. Training is then terminated, and the optimized model obtained after training is used as the optimized drilling speed prediction model.

[0122] The drilling rate prediction model is explained by linearly representing the drilling rate prediction equation and displaying and ranking the weight parameters. The resulting drilling rate prediction model is represented and explained using linear equations. The explanation primarily focuses on whether the weight parameters conform to existing physical mechanisms, ensuring the reliability of the artificial intelligence model. For example, the influence weights of various drilling parameters on the rate of impact (ROP) can be displayed to analyze whether the relationship between a certain drilling parameter and ROP conforms to physical mechanisms. For instance, traditionally, the weight of borehole (WOB) is clearly proportional to the mechanical rate of impact (ROP). However, if the WOB weight parameter shows a negative value at a certain moment, the system will explain the WOB at the corresponding outlier. For example, if the WOB value at this point is 145 kN, far exceeding the average, the system analyzes that the increased WOB at this point accelerates drill bit wear and causes drill bit slippage, thus reducing ROP. Simultaneously, the weight parameters are used to determine whether the intelligent model is overfitting and how well it conforms to the mechanism. The results can be fed back to the ROP model to achieve optimal optimization results.

[0123] The module generates recommended drilling plans and conducts drilling development based on these plans. According to the optimization range of drilling parameters, the module automatically generates suggestions, such as how much to adjust the parameters and how much to increase the drilling speed (for example, if the WOB range is 100-150, and the WOB is only 100kN, the module inputs that adjusting the WOB to 130kN will result in the maximum drilling speed, and the mechanical drilling speed can be increased by 5%). At the same time, the suggestions are fed back to the expert support module to achieve the goal of recommending the final plan.

[0124] Figure 3 The diagram illustrates the drilling equipment, data management module, drilling speed prediction model module, drilling speed prediction interpretation module, and drilling parameter optimization module. The features of each module are as follows:

[0125] Drilling equipment: collects and transmits field data; determines the range of construction parameters.

[0126] Data Management Module: Accepts data input from drilling equipment, performs data cleaning and preprocessing, and allocates data according to module requirements.

[0127] Drilling speed prediction model module: The model is trained using a fully connected neural network, and the optimal drilling speed prediction model is selected.

[0128] The drilling rate prediction interpretation module interprets the neurons and weight parameters of the drilling rate prediction model, thereby interpreting the entire model, and finally linearly represents the drilling rate (ROP) prediction equation.

[0129] Drilling parameter optimization module: Based on the linear characterization equation and the system's predefined optimized drilling parameters, it establishes the relationship between control parameters and drilling rate, ranks the parameters by weight, and automatically recommends several drilling schemes. Based on the drilling rate prediction equation, it analyzes the importance of control parameters (pressure on drill bit, rotational speed, displacement, etc.) to the mechanical drilling rate, and adjusts the drilling parameters according to the set parameter thresholds to maximize the drilling rate, thereby improving drilling speed and efficiency.

[0130] The working principle of each module can be summarized as follows:

[0131] Drilling Equipment: Collects and adjusts drilling parameters in the drilling area to optimize drilling performance. Specifically, it can determine the optimal drilling parameter range based on economic indicators and engineering difficulty, and output this range to the drilling parameter optimization module. Upon receiving feedback from the optimization module, if the parameters meet economic and speed-up targets, it issues instructions to adjust the drilling equipment's operating parameters.

[0132] The data management module accepts data input and feature input. Because different drilling parameters have significantly different value ranges, directly using the raw parameter values ​​for analysis would emphasize the weight of high-value parameters in regression calculations while weakening the role of low-value parameters. Therefore, data cleaning and preprocessing are necessary to reduce model training time and improve prediction accuracy. After cleaning, the data is allocated according to module requirements, and this module always maintains the normality of the data.

[0133] The drilling rate prediction model module receives preprocessed data and features from the data management module during the calculation process. It then uses a fully connected neural network (FNN) to predict the rate of return (ROP). This module optimizes neural network hyperparameters (including the number of layers, neurons per layer, learning rate, and optimizer). After optimizing multiple neural network hyperparameters, the system automatically uses the optimal (high accuracy, no overfitting) drilling rate prediction model. After model calculation, the output data and parameters are transmitted to the drilling rate prediction interpretation module for model interpretation. The model only needs to retain the data input and output interfaces; its internal structure can be completely encapsulated, ensuring the module's stability and efficiency.

[0134] The drilling rate prediction and interpretation module represents and interprets the drilling rate output of the AI ​​model using linear equations. It primarily interprets whether the weight parameters conform to existing physical mechanisms, ensuring the reliability of the AI ​​model. Weight parameter interpretation: This module receives all data from the drilling rate prediction model module and displays the influence weights of various drilling parameters on the ROP (Recovery Point Effect). It analyzes whether the relationship between a certain drilling parameter and ROP conforms to physical mechanisms. For example, traditionally, WOB (Wastewater Obscured) is clearly proportional to ROP, but if the WOB weight parameter shows a negative value at a certain moment, the system will interpret the WOB at the corresponding outlier. For example, if the WOB value at this point is 145kN, far exceeding the average, the system analyzes that the increased WOB at this point accelerates drill bit wear and causes drill bit slippage, thus reducing ROP. Simultaneously, it judges whether the intelligent model is overfitting and how well it conforms to the mechanism based on the weight parameters. The results can be fed back to the ROP model to achieve optimal optimization. This module also provides a linear representation of the intelligent drilling rate prediction model, intuitively displaying the linear equation of the intelligent drilling rate prediction model in a certain well section. The weighting parameter information and linear characterization equation are transmitted to the drilling parameter optimization module.

[0135] The drilling parameter optimization module: On one hand, it receives weighted parameter information and linearly represented drilling rate prediction equations from the drilling rate prediction and interpretation module. Based on the linear equations, it prioritizes the most important parameters and the effect of increasing a unit drilling parameter on the mechanical drilling rate. This module visually displays this processed information to the expert support module. On the other hand, it receives the drilling parameter optimization range provided by the expert support system and automatically generates suggestions—how much to adjust the parameter to and how much to increase the drilling rate (e.g., if the expert provides a WOB range of 100-150kN, and the WOB is currently only 100kN, the module inputs to adjust the WOB to 130kN, increasing the mechanical drilling rate by 5%). Simultaneously, it feeds back these suggestions to the expert support module, achieving the goal of recommending a final solution. It also receives the linear representation equations from the drilling rate prediction and interpretation module interpreting the current prediction results and quantitatively analyzes and judges the magnitude of the influence of each parameter on the mechanical drilling rate based on the weighted parameters. It sets parameter influence thresholds, such as 10% and 20%. To simplify the drilling process, the number of drilling parameters adjusted should be as small as possible. When the influence of adjusting a certain drilling parameter on the mechanical drilling rate is less than the influence threshold, that parameter is not adjusted. Determine the drilling parameters to be adjusted and combine them with the value range and threshold range of the adjustable control parameters of the current drilling equipment (such as drilling pressure 5-20kN), and adjust the drilling parameters to maximize the mechanical drilling rate.

[0136] The interaction process between the modules can be as follows: First, the drilling equipment acquires drilling parameters and inputs them into the data management module. The data management module then performs data cleaning and normalization on the input data. After processing, the data is output to the drilling speed prediction model module. The drilling speed prediction model module trains a fully connected neural network model based on the input data and feeds back the model parameters during training to the drilling speed prediction interpretation module. After interpretation by the drilling speed prediction interpretation module, the interpretation results are sent back to the drilling speed prediction model module. The drilling speed prediction model module then optimizes the trained prediction model, ultimately obtaining an optimized drilling speed prediction model. This optimized model is then sent back to the drilling speed prediction interpretation module, which performs linear characterization on the optimized model to obtain the drilling speed prediction equation. Second, based on the weight parameters of the obtained drilling speed prediction equation, the influence of each drilling parameter on the mechanical drilling speed is quantitatively analyzed. By determining the magnitude of the influence, adjustable parameters among the drilling parameters can be identified. These adjustable drilling parameters are then optimized in the drilling parameter optimization module. During optimization, adjustable drilling parameters can be combined with the threshold range of adjustable control parameters of the current drilling equipment to adjust the drilling parameters and maximize the mechanical drilling rate. Finally, the adjusted drilling parameters are sent to the drilling equipment, which can automatically generate recommended plans for drilling optimization.

[0137] Although this specification provides the following examples or appendices Figure 4 The methods, steps, or apparatus structures shown may include more or fewer combined operational steps or module units based on conventional or non-inventive methods. In steps or structures where there is no logically necessary causal relationship, the execution order of these steps or the module structure of the apparatus is not limited to the execution order or module structure shown in the embodiments or drawings of this specification. When the methods or module structures described are applied in actual devices, servers, or terminal products, they can be executed sequentially or in parallel according to the methods or module structures shown in the embodiments or drawings (e.g., in parallel processor or multi-threaded processing environments, or even distributed processing or server cluster implementation environments).

[0138] Based on the aforementioned drilling data optimization method, this specification also proposes an embodiment of a drilling data optimization device. For example... Figure 4 As shown, it includes: a data management module 401, a drilling speed prediction model module 402, a drilling speed prediction interpretation module 403, a drilling parameter optimization module 404, and an optimization processing module 405. The structure is described below.

[0139] The data management module 401 can be used to obtain the first drilling parameters of the first target well;

[0140] The drilling speed prediction model module 402 is used to train a fully connected neural network model based on the first drilling parameters to obtain a drilling speed prediction model; wherein, the drilling speed prediction model adopts a target activation function;

[0141] The drilling speed prediction interpretation module 403 can be specifically used to process the drilling speed prediction model according to the target activation function and the first drilling parameters to obtain the drilling speed prediction equation; obtain the second drilling parameters of the second target well, input the second drilling parameters into the drilling speed prediction equation to obtain the target drilling speed; and determine the influence value of the second drilling parameters on the target drilling speed according to the weight parameters in the drilling speed prediction equation.

[0142] The drilling parameter optimization module 404 can be specifically used to compare the influence value with a preset threshold, and when the influence value is determined to be greater than the preset threshold, determine the adjustable parameter in the second drilling parameters; wherein the adjustable parameter is within the preset parameter range; and determine the target drilling parameter and the second target drilling speed value corresponding to the target drilling parameter according to the preset parameter range and the drilling speed prediction equation; wherein the second target drilling speed value is greater than the preset drilling speed threshold.

[0143] The optimization processing module 405 can be used to optimize the drilling process of the second target well based on the target drilling parameters.

[0144] In some embodiments, the data management module 401 described above may include, before implementation, the following steps: obtaining the initial drilling parameters of the first target well; performing data cleaning and normalization on the initial drilling parameters to obtain the first drilling parameters of the first target well.

[0145] In some embodiments, the first drilling parameters in the data management module 401 may include one or more of the following: rotational speed, drilling pressure, torque, riser pressure, inlet flow rate, hook load, drilling fluid density, drilling fluid viscosity, drilling fluid plastic viscosity, and yield strength.

[0146] In some embodiments, the drilling speed prediction and interpretation module 403 described above may include obtaining the drilling speed prediction equation according to the following formula:

[0147] ROP=A×RP+B×WOB+C×Tor+D×SPP+E×IF+F×HL+G×Den-H×Vis-I×PV+J×YP

[0148] Where RP is rotational speed, WOB is drilling pressure, Tor is torque, SPP is riser pressure, IF is inlet flow rate, HL is hook load, Den is drilling fluid density, Vis is drilling fluid viscosity, PV is drilling fluid plastic viscosity, YP is yield strength, and A, B, C, D, E, F, G, H, I, and J are the weighting parameters corresponding to each drilling parameter.

[0149] In some embodiments, the drilling speed prediction and interpretation module 403 described above may include: determining the average value of each drilling parameter in the second drilling parameters; performing product processing on each weight parameter and the average value to obtain a product set; comparing each product value in the product set to determine the influence value of the second drilling parameters on the target drilling speed based on the comparison result.

[0150] In some embodiments, the drilling parameter optimization module 404 may include: adjusting the adjustable parameters according to a preset parameter range and a preset adjustment rule to obtain multiple sets of adjusted drilling parameters; sequentially inputting the multiple sets of drilling parameters into the drilling speed prediction equation to obtain multiple sets of drilling speed values; comparing the multiple sets of drilling speeds corresponding to the multiple sets of drilling parameters with a preset drilling speed threshold to determine the target drilling parameter in the multiple sets of drilling parameters and the second target drilling speed value corresponding to the target drilling parameter in the multiple sets of drilling speeds.

[0151] In some embodiments, the optimization processing module 405 may include: determining a drilling speed enhancement value based on the target drilling speed and the second target drilling speed value; forming a recommended scheme based on the drilling speed enhancement value and the target drilling parameters, and pushing the recommended scheme to the drilling equipment in real time; the drilling equipment receiving the recommended scheme and optimizing the drilling process of the second target well based on the recommended scheme.

[0152] It should be noted that the units, devices, or modules described in the above embodiments can be implemented by computer chips or physical entities, or by products with certain functions. For ease of description, the above devices are described by dividing them into various modules according to their functions. Of course, in implementing this specification, the functions of each module can be implemented in one or more software and / or hardware, or the module that implements the same function can be implemented by a combination of multiple sub-modules or sub-units, etc. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and there may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection between the devices or units shown or discussed can be through some interfaces, and the indirect coupling or communication connection between devices or units can be electrical, mechanical, or other forms.

[0153] As can be seen from the above, the interpretable drilling parameter intelligent optimization device provided in the embodiments of this specification, on the one hand, can overcome the uncontrollability of the prediction model by determining the drilling rate prediction equation based on the optimized drilling rate prediction model, thus making the prediction results of the model highly reliable. On the other hand, by optimizing drilling parameters, the efficiency of oil exploration and development can be improved and drilling costs reduced.

[0154] This specification also provides an electronic device, including a processor and a memory for storing processor-executable instructions. Specifically, the processor can perform the following steps according to the instructions: acquiring first drilling parameters of a first target well, training a fully connected neural network model based on the first drilling parameters to obtain a drilling speed prediction model; wherein the drilling speed prediction model uses a target activation function; processing the drilling speed prediction model according to the target activation function and the first drilling parameters to obtain a drilling speed prediction equation; acquiring second drilling parameters of a second target well, inputting the second drilling parameters into the drilling speed prediction equation to obtain a target drilling speed; determining the influence value of the second drilling parameters on the target drilling speed according to each weight parameter in the drilling speed prediction equation; comparing the influence value with a preset threshold, and determining an adjustable parameter in the second drilling parameters when the influence value is greater than the preset threshold; wherein the adjustable parameter is within a preset parameter range; determining the target drilling parameters and the corresponding second target drilling speed value according to the preset parameter range and the drilling speed prediction equation; wherein the second target drilling speed value is greater than the preset drilling speed threshold; and optimizing the drilling process of the second target well based on the target drilling parameters.

[0155] To execute the above instructions more accurately, please refer to... Figure 5 As shown in the embodiments of this specification, another specific electronic device is also provided, wherein the electronic device includes a network communication port 501, a processor 502, and a memory 503. The above structures are connected by internal cables so that the various structures can perform specific data interaction.

[0156] Specifically, the network communication port 501 can be used to acquire the first drilling parameters of the first target well, so as to train a fully connected neural network model based on the first drilling parameters to obtain a drilling speed prediction model; wherein the drilling speed prediction model adopts a target activation function.

[0157] The processor 502 can specifically be used to process the drilling rate prediction model according to the target activation function and the first drilling parameters to obtain a drilling rate prediction equation; obtain the second drilling parameters of the second target well, input the second drilling parameters into the drilling rate prediction equation to obtain the target drilling rate; determine the influence value of the second drilling parameters on the target drilling rate according to each weight parameter in the drilling rate prediction equation; compare the influence value with a preset threshold, and when the influence value is determined to be greater than the preset threshold, determine the adjustable parameters in the second drilling parameters; wherein the adjustable parameters are within the preset parameter range; determine the target drilling parameters and the second target drilling rate value corresponding to the target drilling parameters according to the preset parameter range and the drilling rate prediction equation; wherein the second target drilling rate value is greater than the preset drilling rate threshold; and optimize the drilling process of the second target well based on the target drilling parameters.

[0158] The memory 503 can be used to store the corresponding instruction program.

[0159] In this embodiment, the network communication port 501 can be a virtual port bound to different communication protocols, thereby enabling the sending or receiving of different data. For example, the network communication port can be a port responsible for web data communication, a port responsible for FTP data communication, or a port responsible for email data communication. Furthermore, the network communication port can also be a physical communication interface or communication chip. For example, it can be a wireless mobile network communication chip, such as GSM or CDMA; it can also be a Wi-Fi chip; or it can be a Bluetooth chip.

[0160] In this embodiment, the processor 502 can be implemented in any suitable manner. For example, the processor can take the form of a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro)processor, logic gates, switches, application-specific integrated circuits (ASICs), programmable logic controllers, and embedded microcontrollers, etc. This specification is not limiting.

[0161] In this embodiment, the memory 503 may include multiple layers. In a digital system, anything that can store binary data can be a memory. In an integrated circuit, a circuit with storage function but no physical form is also called a memory, such as RAM, FIFO, etc. In a system, a storage device with a physical form is also called a memory, such as a memory stick, TF card, etc.

[0162] This specification also provides a computer storage medium based on the above-described interpretable intelligent optimization method for drilling parameters. The computer storage medium stores computer program instructions that, when executed, implement the following: acquiring first drilling parameters of a first target well; training a fully connected neural network model based on the first drilling parameters to obtain a drilling rate prediction model; wherein the drilling rate prediction model employs a target activation function; processing the drilling rate prediction model according to the target activation function and the first drilling parameters to obtain a drilling rate prediction equation; acquiring second drilling parameters of a second target well; and inputting the second drilling parameters to... In the drilling speed prediction equation, the target drilling speed is obtained; based on the weight parameters in the drilling speed prediction equation, the influence value of the second drilling parameter on the target drilling speed is determined; the influence value is compared with a preset threshold, and when the influence value is determined to be greater than the preset threshold, an adjustable parameter in the second drilling parameter is determined; wherein, the adjustable parameter is within the preset parameter range; based on the preset parameter range and the drilling speed prediction equation, the target drilling parameter and the corresponding second target drilling speed value are determined; wherein, the second target drilling speed value is greater than the preset drilling speed threshold; the drilling process of the second target well is optimized based on the target drilling parameter.

[0163] In this embodiment, the storage medium includes, but is not limited to, Random Access Memory (RAM), Read-Only Memory (ROM), cache, hard disk drive (HDD), or memory card. The memory can be used to store computer program instructions. The network communication unit can be an interface configured according to standards specified in the communication protocol for network connection communication.

[0164] While this specification provides the steps of operation for the methods described in the embodiments or flowcharts, more or fewer steps may be included based on conventional or non-inventive means. The order of steps listed in the embodiments is merely one possible order of execution among many steps and does not represent the only possible order. In actual device or client product execution, the methods shown in the embodiments or drawings may be executed sequentially or in parallel (e.g., in a parallel processor or multi-threaded processing environment, or even a distributed data processing environment). The terms "comprising," "including," or any other variations thereof are intended to cover a non-exclusive inclusion, such that a process, method, product, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, product, or apparatus. Without further limitations, the presence of other identical or equivalent elements in a process, method, product, or apparatus that includes said elements is not excluded. The terms "first," "second," etc., are used to denote names and do not indicate any particular order.

[0165] Those skilled in the art will also know that, besides implementing the controller using purely computer-readable program code, the same functions can be achieved by logically programming the method steps, making the controller function as logic gates, switches, application-specific integrated circuits (ASICs), programmable logic controllers (PLCs), and embedded microcontrollers. Therefore, such a controller can be considered a hardware component, and the devices within it used to implement various functions can also be considered structures within that hardware component. Alternatively, the devices used to implement various functions can be considered as both software modules implementing the method and structures within a hardware component.

[0166] This specification can be described in the general context of computer-executable instructions that are executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, classes, etc., that perform a specific task or implement a specific abstract data type. This specification can also be practiced in distributed computing environments, where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0167] As can be seen from the above description of the embodiments, those skilled in the art can clearly understand that this specification can be implemented by means of software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solutions of this specification can essentially be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, mobile terminal, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments of this specification.

[0168] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to interchangeably. Each embodiment focuses on its differences from other embodiments. This specification can be used in numerous general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable electronic devices, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices, etc.

[0169] Although this specification has been described by way of examples, those skilled in the art will recognize that many variations of this specification are possible without departing from its spirit, and it is intended that the appended claims cover such variations without departing from the spirit of this specification.

Claims

1. An interpretable intelligent optimization method for drilling parameters, characterized in that, include: First drilling parameters of the first target well are obtained, and a fully connected neural network model is trained based on the first drilling parameters to obtain a drilling rate prediction model; wherein, the drilling rate prediction model adopts a target activation function; The drilling rate prediction model is processed according to the target activation function and the first drilling parameters to obtain the drilling rate prediction equation; Obtain the second drilling parameters of the second target well, and input the second drilling parameters into the drilling rate prediction equation to obtain the target drilling rate; Based on the weighted parameters in the drilling speed prediction equation, determine the influence value of the second drilling parameter on the target drilling speed; The influence value is compared with a preset threshold. When the influence value is determined to be greater than the preset threshold, an adjustable parameter in the second drilling parameters is determined; wherein the adjustable parameter is within the preset parameter range. Based on the preset parameter range and the drilling rate prediction equation, the target drilling parameters and the corresponding second target drilling rate value are determined; wherein, the second target drilling rate value is greater than the preset drilling rate threshold. The drilling process of the second target well is optimized based on the target drilling parameters. The first drilling parameters include one or more of the following: rotational speed, drilling pressure, torque, riser pressure, inlet flow rate, hook load, drilling fluid density, drilling fluid viscosity, drilling fluid plastic viscosity, and yield strength. The drilling speed prediction equation is obtained using the following formula: wherein RP is the rotary speed, WOB is the weight on bit, Tor is the torque, SPP is the standpipe pressure, IF is the inlet flow, HL is the hook load, Den is the drilling fluid density, Vis is the drilling fluid viscosity, PV is the plastic viscosity of the drilling fluid, YP is the yield strength, is a weight parameter corresponding to each drilling parameter; Based on the weighted parameters in the drilling speed prediction equation, the influence of the second drilling parameter on the target drilling speed is determined, including: Determine the average value of each drilling parameter in the second drilling parameters; Each weight parameter is multiplied by the average value to obtain a product set. The influence of the second drilling parameter on the target drilling rate is determined based on the comparison results by comparing each product value in the product set.

2. The method of claim 1, wherein, Before obtaining the first drilling parameters for the first target well, the following steps are also included: Obtain the initial drilling parameters for the first target well; The initial drilling parameters are cleaned and normalized to obtain the first drilling parameters of the first target well.

3. The method of claim 1, wherein, Based on the preset parameter range and the drilling rate prediction equation, the target drilling parameters and the corresponding second target drilling rate value are determined, including: The adjustable parameters are adjusted according to the preset parameter range and preset adjustment rules to obtain multiple sets of adjusted drilling parameters; The multiple sets of drilling parameters are sequentially input into the drilling rate prediction equation to obtain multiple sets of drilling rate values; The drilling parameters are compared with the preset drilling speed thresholds to determine the target drilling parameter and the second target drilling speed value corresponding to the target drilling parameter.

4. The method of claim 3, wherein, The drilling process of the second target well is optimized based on the target drilling parameters, including: The drilling speed increase value is determined based on the target drilling speed and the second target drilling speed value; Based on the drilling speed improvement value and the target drilling parameters, a recommended plan is generated and the recommended plan is pushed to the drilling equipment in real time. The drilling equipment receives the recommended plan and optimizes the drilling process of the second target well based on the recommended plan.

5. An interpretable drilling parameter intelligent optimization apparatus, characterized in that, include: The data management module is used to obtain the first drilling parameters of the first target well; The drilling speed prediction model module is used to train a fully connected neural network model based on the first drilling parameters to obtain a drilling speed prediction model; wherein, the drilling speed prediction model adopts a target activation function; The drilling speed prediction interpretation module is used to process the drilling speed prediction model according to the target activation function and the first drilling parameters to obtain the drilling speed prediction equation; obtain the second drilling parameters of the second target well, input the second drilling parameters into the drilling speed prediction equation to obtain the target drilling speed; and determine the influence value of the second drilling parameters on the target drilling speed according to the weight parameters in the drilling speed prediction equation. The drilling parameter optimization module is used to compare the influence value with a preset threshold, and when the influence value is determined to be greater than the preset threshold, to determine an adjustable parameter in the second drilling parameters; wherein the adjustable parameter is within the preset parameter range; and to determine the target drilling parameter and the second target drilling speed value corresponding to the target drilling parameter according to the preset parameter range and the drilling speed prediction equation; wherein the second target drilling speed value is greater than the preset drilling speed threshold. An optimization processing module is used to optimize the drilling process of the second target well based on the target drilling parameters; The first drilling parameters include one or more of the following: rotational speed, drilling pressure, torque, riser pressure, inlet flow rate, hook load, drilling fluid density, drilling fluid viscosity, drilling fluid plastic viscosity, and yield strength. The drilling speed prediction equation is obtained using the following formula: wherein RP is the rotary speed, WOB is the weight on bit, Tor is the torque, SPP is the standpipe pressure, IF is the inlet flow, HL is the hook load, Den is the drilling fluid density, Vis is the drilling fluid viscosity, PV is the plastic viscosity of the drilling fluid, YP is the yield strength, is a weight parameter corresponding to each drilling parameter; Based on the weighted parameters in the drilling speed prediction equation, the influence of the second drilling parameter on the target drilling speed is determined, including: Determine the average value of each drilling parameter in the second drilling parameters; Each weight parameter is multiplied by the average value to obtain a product set. The influence of the second drilling parameter on the target drilling rate is determined based on the comparison results by comparing each product value in the product set.

6. The apparatus of claim 5, wherein, Also includes: The cleaning module is used to obtain the initial drilling parameters of the first target well; The initial drilling parameters are cleaned and normalized to obtain the first drilling parameters of the first target well.

7. A computer readable storage medium characterized in that, It stores computer instructions that, when executed by a processor, implement the steps of the method according to any one of claims 1 to 4.

Images