Well track regulation and control method and device based on BP neural network

Abstract

According to the technical scheme, the method comprises the steps that at least one standard characteristic parameter is obtained from a target to-be-drilled well and a drilled well within a preset range of the target to-be-drilled well, wherein the standard characteristic parameter represents a parameter influencing a gamma value; inputting the characteristic parameters into a pre-trained gamma value prediction model based on a BP neural network to obtain an upper gamma value above the drill bit, a lower gamma value below the drill bit and a gamma value at the drill bit; and based on the upper gamma value, the lower gamma value and the gamma value at the drill bit, the drilling direction of the drill bit is adjusted to regulate and control the well track. According to the technical scheme, the drill bit gamma parameter prediction and well track regulation and control method based on the BP neural network technology is provided, the drilling tool is guided in real time to adjust the drilling track by predicting the drill bit gamma intensity and obtaining the change of the formation lithology, and the drilling well quality is improved.

Description

Technical Field

[0001] The invention relates to the technical field of petroleum and natural gas, and in particular to a wellbore trajectory control method and device based on a BP neural network. Background Art

[0002] Gamma prediction of the drill bit and precise control of the borehole trajectory are key links in oil and gas exploration and development, directly affecting the discovery of oil and gas layers and the effects of their subsequent development. There is a potential correspondence between the drilling engineering parameters and key geological parameters such as the downhole formation boundary, thickness and inclination, and gamma intensity. It is of great significance to complete the evaluation of parameters such as formation rock type, porosity, permeability, etc., help analyze the formation structure, and guide the completion of borehole trajectory control, so that the borehole trajectory can be more accurately maintained in the target layer and the oil and gas recovery rate can be improved.

[0003] Traditional prediction methods often rely on empirical formulas or models, and the regulation of wellbore trajectories is not precise enough, which can easily lead to trajectory control deviation from the reservoir.

[0004] In summary, a gamma prediction and precise wellbore trajectory control technology is needed to solve the problems of inaccurate positioning of drilling tools and trajectory control deviation leading to layer deviation in complex geological conditions and downhole environments. Summary of the invention

[0005] The present invention provides a wellbore trajectory control method and device based on BP neural network to solve the problems of inaccurate positioning of drilling tools and layer deviation caused by trajectory control deviation in complex geological conditions and downhole environment.

[0006] According to one aspect of the present invention, a wellbore trajectory control method based on a BP neural network is provided, comprising:

[0007] Acquire at least one standard characteristic parameter from a target well to be drilled and a well within a preset range of the target well to be drilled, wherein the standard characteristic parameter represents a parameter that has an impact on the gamma value;

[0008] Inputting the characteristic parameters into a pre-trained gamma value prediction model based on a BP neural network to obtain an upper gamma value above the drill bit, a lower gamma value below the drill bit, and a gamma value at the drill bit;

[0009] The drilling direction of the drill bit is adjusted based on the upper gamma value, the lower gamma value and the gamma value at the drill bit to control the wellbore trajectory.

[0010] Optionally, the gamma value prediction model is constructed in the following manner:

[0011] Acquire at least one sample standard characteristic parameter combination and construct a test set, a training set and a validation set, wherein the sample standard characteristic parameter combination includes a sample standard characteristic parameter and a corresponding upper gamma value, a lower gamma value and a gamma value at a drill bit;

[0012] The number of input layer nodes, the number of hidden layers, the number of hidden layer nodes, the number of output layer nodes and the training function of the BP neural network are set to construct the gamma value prediction model, wherein the number of output layer nodes is at least 3, corresponding to the upper gamma value, the lower gamma value and the gamma value at the drill bit, respectively.

[0013] Optionally, the number of nodes in the hidden layer is determined by the following formula:

[0014] h = int(a*I+b*U+c);

[0015] Among them, I is the number of nodes in the input layer, U is the number of nodes in the output layer, h is the number of nodes in the hidden layer, int means rounding, and a, b, c are empirical adjustment constants.

[0016] Optionally, the gamma value prediction model is trained in the following manner:

[0017] Initialize the weights and biases randomly, train the training set according to the initial weights, and obtain predicted values; compare the predicted values ​​with the actual gamma value, the lower gamma value, and the gamma value at the drill bit, calculate the output error of each node, and adjust the weights and thresholds;

[0018] Repeating the training with the training set and performing error verification of the predicted value with the verification set, and completing the training when the error is within a preset range;

[0019] The generalization ability of the model and the prediction accuracy of gamma are evaluated through the test set, and the optimal training model is finalized.

[0020] Optionally, calculating the output error of each node includes:

[0021] According to the input parameters at time t, the predicted upper gamma value, lower gamma value and gamma value at the drill bit at time t+1 are obtained;

[0022] The error values ​​of the upper gamma value, the lower gamma value and the gamma value at the drill bit are determined according to the target values ​​of the upper gamma value, the lower gamma value and the gamma value at the drill bit and the predicted upper gamma value, the lower gamma value and the gamma value at the drill bit.

[0023] Optionally, the adjusting weights and thresholds includes:

[0024] Starting from the output layer, the output error of each layer of neurons is calculated layer by layer. Through the error function, the weights and thresholds of each layer are adjusted and corrected, so that the final output of the network gradually approaches the expected value.

[0025] Optionally, adjusting the drilling direction of the drill bit based on the upper gamma value, the lower gamma value and the gamma value at the drill bit to control the wellbore trajectory includes:

[0026] When the upper gamma value is greater than the lower gamma value and the lower gamma value is greater than the gamma value at the drill bit, adjusting the drill bit toward the reservoir;

[0027] When the upper gamma value is smaller than the lower gamma value and the lower gamma value is smaller than the gamma value at the drill bit, the drill bit is adjusted toward the upper reservoir direction.

[0028] According to another aspect of the present invention, a wellbore trajectory control device based on a BP neural network is provided, comprising:

[0029] A parameter acquisition unit, used to acquire at least one standard characteristic parameter from a target well to be drilled and a well within a preset range of the target well to be drilled, wherein the standard characteristic parameter represents a parameter that affects a gamma value;

[0030] A gamma value output unit, used for inputting the characteristic parameters into a pre-trained gamma value prediction model based on a BP neural network to obtain an upper gamma value above the drill bit, a lower gamma value below the drill bit, and a gamma value at the drill bit;

[0031] A trajectory control unit is used to adjust the drilling direction of the drill bit based on the upper gamma value, the lower gamma value and the gamma value at the drill bit to control the wellbore trajectory.

[0032] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0033] At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the wellbore trajectory control method based on the BP neural network described in any embodiment of the present invention.

[0034] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the wellbore trajectory control method based on BP neural network described in any embodiment of the present invention when executed.

[0035] According to another aspect of the present invention, a computer program product is provided. The computer program product includes a computer program. When the computer program is executed by a processor, the wellbore trajectory control method based on the BP neural network described in any embodiment of the present invention is implemented.

[0036] The technical solution of the embodiment of the present invention includes: obtaining at least one standard characteristic parameter from the target well to be drilled and the wells within the preset range of the target well to be drilled, wherein the standard characteristic parameter represents the parameter that affects the gamma value; inputting the characteristic parameter into a pre-trained gamma value prediction model based on a BP neural network to obtain an upper gamma value above the drill bit, a lower gamma value below the drill bit, and a gamma value at the drill bit; adjusting the drilling direction of the drill bit based on the upper gamma value, the lower gamma value, and the gamma value at the drill bit to control the borehole trajectory. The technical solution of the embodiment of the present invention proposes a method for predicting drill bit gamma parameters and controlling borehole trajectory based on BP neural network technology, which predicts the drill bit gamma intensity, obtains changes in formation lithology, and guides the drilling tool to adjust the drilling trajectory in real time to improve the quality of the drilling borehole.

[0037] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0039] Figure 1 It is a flow chart of a wellbore trajectory control method based on BP neural network provided in the first embodiment of the present invention;

[0040] Figure 2 It is a schematic diagram of building a gamma value prediction model based on a BP neural network applicable to the first embodiment of the present invention;

[0041] Figure 3 This is a schematic diagram of BP neural network output error calculation and weight correction applicable to the first embodiment of the present invention;

[0042] Figure 4 This is a schematic diagram of a wellbore trajectory control based on the predicted gamma value at the drill bit applicable to the first embodiment of the present invention;

[0043] Figure 5 It is a structural schematic diagram of a wellbore trajectory control device based on a BP neural network provided in the second embodiment of the present invention;

[0044] Figure 6 It is a structural schematic diagram of an electronic device for implementing the wellbore trajectory control method based on BP neural network according to an embodiment of the present invention. DETAILED DESCRIPTION

[0045] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0046] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0047] Embodiment 1

[0048] Figure 1 This is a flow chart of a borehole trajectory control method based on a BP neural network provided in the first embodiment of the present invention. This embodiment can be applied to the situation of adjusting the drilling trajectory of a drilling tool and improving the quality of a drilling borehole. The method can be executed by a borehole trajectory control device based on a BP neural network. The borehole trajectory control device based on a BP neural network can be implemented in the form of hardware and / or software. The borehole trajectory control device based on a BP neural network can be configured in an electronic device. Figure 1 As shown, the method includes:

[0049] S110. Acquire at least one standard characteristic parameter from a target well to be drilled and a well within a preset range of the target well to be drilled, wherein the standard characteristic parameter represents a parameter that affects a gamma value.

[0050] The training, verification and test samples of this drill bit gamma parameter prediction model are mainly derived from well logging and data from neighboring wells around the well to be drilled. Its data types include drilling engineering parameters and geological parameters. Engineering parameters include: drilling fluid performance parameters, such as drilling fluid density, drilling fluid viscosity, drilling fluid water loss rate, drilling fluid pH value, etc.; wellbore pressure parameters, such as formation pore pressure, formation fracture pressure, bottom hole pressure, casing shoe pressure, etc.; well depth structure parameters, such as well depth, well diameter, casing size and type, drill bit type and size, etc.; drilling operation parameters, such as temperature, drilling pressure, torque, displacement, rotary table speed, pump flow and return flow, etc.; drilling tool parameters, such as tool size, zero length, etc. Geological parameters, including resistivity, upper gamma, lower gamma, geomagnetic field strength, underground temperature gradient, rock elastic modulus, Poisson's ratio, etc.

[0051] S120, inputting the characteristic parameters into a pre-trained gamma value prediction model based on a BP neural network to obtain an upper gamma value above the drill bit, a lower gamma value below the drill bit, and a gamma value at the drill bit.

[0052] It is necessary to use the logging gamma parameters as the target labels and the key drilling engineering / geological parameters as the input, and output the upper and lower gamma values ​​through BP neural network learning to complete the identification of reservoir characteristics at the drill bit. In addition, the drilling design, exploration well and seismic data are combined to analyze the decision-making trajectory adjustment parameters and guide the completion of trajectory adjustment.

[0053] In the embodiment of the present invention, the gamma value prediction model is constructed in the following manner:

[0054] Acquire at least one sample standard characteristic parameter combination and construct a test set, a training set and a validation set, wherein the sample standard characteristic parameter combination includes a sample standard characteristic parameter and a corresponding upper gamma value, a lower gamma value and a gamma value at a drill bit;

[0055] The number of input layer nodes, the number of hidden layers, the number of hidden layer nodes, the number of output layer nodes and the training function of the BP neural network are set to construct the gamma value prediction model, wherein the number of output layer nodes is at least 3, corresponding to the upper gamma value, the lower gamma value and the gamma value at the drill bit, respectively.

[0056] The gamma parameter prediction model at the drill bit adopts BP neural network, and the topological structure includes input layer, multiple hidden layers and output layer. The number of input layer nodes is determined according to the number of input variables. The type of neural network input data requires various parameter combinations to complete model training, and after completing the performance comparison of prediction error, model complexity, etc., the optimal data combination is determined. The number of nodes in the output layer of the neural network is 3, which are the upper gamma value, the lower gamma value and the gamma prediction value at the drill bit, which are used for proofreading and error calculation of the training model, as well as training and optimization of the model. The connection status of the nodes between the number layers of output variables is reflected by weights to ensure that the model can learn and store the mapping relationship between a large number of engineering input parameters and gamma prediction outputs.

[0057] In the embodiment of the present invention, the number of hidden layer nodes is determined by the following formula:

[0058] h = int(a*I+b*U+c);

[0059] Among them, I is the number of nodes in the input layer, U is the number of nodes in the output layer, h is the number of nodes in the hidden layer, int means rounding, and a, b, c are empirical adjustment constants.

[0060] In the embodiment of the present invention, the gamma value prediction model is trained in the following manner:

[0061] Initialize the weights and biases randomly, train the training set according to the initial weights, and obtain predicted values; compare the predicted values ​​with the actual gamma value, the lower gamma value, and the gamma value at the drill bit, calculate the output error of each node, and adjust the weights and thresholds;

[0062] Repeating the training with the training set and performing error verification of the predicted value with the verification set, and completing the training when the error is within a preset range;

[0063] The generalization ability of the model and the prediction accuracy of gamma are evaluated through the test set, and the optimal training model is finalized.

[0064] The test set, training set, and validation set all come from well logging and data from nearby wells to be drilled. Their respective functions are as follows:

[0065] Training set: used to train the model so that the model can learn the features and patterns in the data, thereby adjusting the parameters of the model and improving the model's ability to predict unknown data.

[0066] Validation set: used to conduct a preliminary evaluation of the model's performance during model training, help adjust the model's hyperparameters, and select a suitable model. The results of the validation set can be used to guide model improvement and optimization.

[0067] Test set: used to evaluate the final performance of the model after model training is completed. The test set is independent of the training set. The performance on the test set can be used to test the generalization ability of the model on new data and determine whether the model can make accurate predictions, rather than just performing well on the training set.

[0068] These three data sets work together to develop and optimize the model. In practical applications, the data is usually divided into training and test sets, and then the model is trained on the training set, the model is selected and adjusted on the validation set, and the final performance of the model is evaluated on the test set. This ensures that the model has good accuracy and reliability in practical applications.

[0069] The gamma parameter prediction model of this system completes the training of the BP neural network by adjusting the two parameters of network weight and bias. The training process of the gamma prediction model is divided into two parts: before the model training network, the weights and bias are randomly initialized, the model data is forward transmitted, weighted summation is performed, and the gamma prediction value at the drill bit is calculated; during model training, the output prediction value is compared with the actual value, the error between each node output and the expected output is calculated, the weight and bias terms are adjusted, and the error is back-propagated layer by layer starting from the output layer to complete the above weight updates. Repeat the training set for multiple times until the gamma prediction value at the drill bit reaches the preset allowable error range, and the training is completed.

[0070] This system completes the optimization of engineering / geological parameter combinations based on model training results. Different models are used to train different sample set combinations. Based on the principles of minimum prediction target error and minimum training model complexity, redundant input parameters are eliminated to improve the geological feature characterization ability of the training model input data, and the optimal sample set combination and the best training model for model training, verification and testing are selected. In case the best training model is not unique, multi-model fusion or multi-model prediction result voting or averaging is used to improve the accuracy and stability of the neural network model gamma prediction.

[0071] In the embodiment of the present invention, calculating the output error of each node includes:

[0072] According to the input parameters at time t, the predicted upper gamma value, lower gamma value and gamma value at the drill bit at time t+1 are obtained;

[0073] The error values ​​of the upper gamma value, the lower gamma value and the gamma value at the drill bit are determined according to the target values ​​of the upper gamma value, the lower gamma value and the gamma value at the drill bit and the predicted upper gamma value, the lower gamma value and the gamma value at the drill bit.

[0074] In the embodiment of the present invention, the adjusting weight and threshold includes:

[0075] Starting from the output layer, the output error of each layer of neurons is calculated layer by layer. Through the error function, the weights and thresholds of each layer are adjusted and corrected, so that the final output of the network gradually approaches the expected value.

[0076] S130, adjusting the drilling direction of the drill bit based on the upper gamma value, the lower gamma value and the gamma value at the drill bit to control the wellbore trajectory.

[0077] In the embodiment of the present invention, step S130 specifically includes:

[0078] When the upper gamma value is greater than the lower gamma value and the lower gamma value is greater than the gamma value at the drill bit, adjusting the drill bit toward the reservoir;

[0079] When the upper gamma value is smaller than the lower gamma value and the lower gamma value is smaller than the gamma value at the drill bit, the drill bit is adjusted toward the upper reservoir direction.

[0080] The scheme of the present invention is explained in detail below through a specific embodiment:

[0081] Figure 2 This is a schematic diagram of building a gamma value prediction model based on a BP neural network applicable to the first embodiment of the present invention. It mainly includes parameter combination and data processing, model building and setting, feature combination and training model finalization, field test and feedback.

[0082] By collecting logging data and information from neighboring wells around the well to be drilled, multi-source engineering / geological parameters are obtained, among which the gamma parameters come from the azimuth gamma instrument while drilling. Before dividing the training set, validation set, and test set, the sample set of each parameter combination needs to be eliminated and normalized for outliers and noise to improve the accuracy of the prediction model.

[0083] Five characteristic parameters that affect the gamma value are selected, namely, drilling fluid density, drilling pressure, rotation speed, torque, and vertical depth, to construct characteristic vectors and establish a drill bit gamma value prediction model based on BP neural network. The model parameters and training functions are set, the number of input nodes is 5, the number of hidden layers is 2, the number of hidden layer nodes is 9, and the number of output nodes is 3 to explore the nonlinear relationship between input and output. Here, considering that the 3-layer BP network can map any nonlinear relationship, but the convergence is poor, the number of hidden layers is increased to 2 layers to improve the convergence efficiency and the training is not easy to fall into the local optimum. The hidden layer nodes are determined according to the empirical formula, h = int (a*I+b*U+c), where I is the number of input layer nodes, U is the number of output layer nodes, h is the number of hidden layer nodes, int means rounding, a, b, c are empirical adjustment constants, usually ranging from 1 to 10, and 1 is taken here; the activation function selects S type:

[0084]

[0085] Figure 3 1 is a schematic diagram of BP neural network output error calculation and weight correction applicable to the first embodiment of the present invention. When calculating the input layer input, the known trajectory parameters X(k)={x(t-k+1)...x(t-2), x(t-1), x(t)} at consecutive k moments are used as model input. The input parameter X(t) at moment t can be expressed as x(t)={DFD(t), WOB(t), Rev(t), TORQ(t), TVD(t)}, where DFD, WOB, Rev, TORQ, and TVD represent drilling fluid density, drilling pressure, rotation speed, torque, and vertical depth, respectively. Calculate the moment t+1. The gamma output of the BP neural network prediction model is Y(t+1), then Y(t+1)=f(X(k))={UGa(t+1), DGa(t+1), PGa(t+1)}, where UGa, DGa, and PGa are the upper gamma, lower gamma, and gamma values ​​at the drill bit of the network prediction output, respectively; the network target output is T(t+1)={TUGa(t+1), TDGa(t+1), TPGa(t+1)}, where TUGa, TDGa, and TPGa are the target values ​​of the upper gamma, lower gamma, and gamma values ​​at the drill bit, respectively, and they are all known parameters of the well or neighboring wells obtained through logging data and other data.

[0086] Starting from the output layer, the output errors of neurons in each layer are calculated layer by layer, and the gradient descent method is used to adjust and correct the weights and thresholds of each layer, so that the final output of the network gradually approaches the expected value.

[0087] The error is calculated as follows:

[0088]

[0089] The upper gamma, lower gamma, and predicted gamma output errors at the drill bit are:

[0090]

[0091] The above-mentioned model is used to iteratively train the different combination features of the above five engineering parameters. By comparing the predicted gamma value at the output drill bit under each combination mode with the gamma target value of the training set, the prediction error is calculated, and the characteristic characterization ability of each combination parameter is evaluated to complete the finalization of the optimal parameter combination mode. The optimal parameter combination obtained in this training example includes four characteristic parameters: drilling fluid density, drilling pressure, rotation speed, and vertical depth. At the same time, according to the above-mentioned optimal parameter combination type, the relevant parameters in the combined verification set form the optimal verification set, and the model is further verified to evaluate its generalization ability and the prediction accuracy of gamma, and complete the finalization of the optimal training model.

[0092] The above-mentioned stereotyped model is applied to the field computing platform, and the real-time drilling fluid density, drilling pressure, rotation speed, and vertical depth parameters obtained on site are used to construct a feature vector. The gamma value at the drill bit is predicted through the above-mentioned stereotyped training model, and the prediction results are analyzed. If the error is large, the model parameters are adjusted and the model is retrained.

[0093] Figure 4 This is a schematic diagram of a borehole trajectory control based on the predicted gamma value at the drill bit applicable to the first embodiment of the present invention. The gamma measurement sub is installed between the drill bit and the drill collar. By collaboratively analyzing and predicting the gamma near the drill bit and other logging parameters, the position of the drilling tool in the reservoir can be clarified, and the adjustment and optimization of the drilling trajectory can be guided to ensure that the drill bit travels through high-quality reservoirs. Before drilling, a pre-drilling geological model is built based on the attenuation law of gamma in the formation; a gamma radioactivity intensity distribution map is established using the data of adjacent wells; and the coordinates and vertical depth of the target point are calculated using spatial geometric relationships, which are used as a fine correction map. During the drilling process, the downhole measurement instrument measures the well inclination, azimuth, tool face, upper gamma, lower gamma and other parameters in real time, and transmits them to the ground control system wirelessly or wired. The ground analyzes the downhole measurement instrument parameters, uses the well inclination, azimuth, and tool face parameters to reverse and correct the map, calculate the target layer position, and determine the exact position of the drill tool in the reservoir. At the same time, by predicting the changing trend of the gamma value, it is determined whether the drilling trajectory is about to deviate and whether it exceeds the preset error range. If it exceeds the set threshold, the drilling direction of the drill bit is adjusted to correct the deviation, thereby completing accurate control of the drilling status of the wellbore trajectory.

[0094] When the tool passes through the reservoir, the characteristics and trajectory adjustment strategy based on the upper and lower gammas are as follows: when the drill tool is in a thin sandstone formation, the gamma value fluctuation range is small, and the upper gamma, lower gamma and gamma at the drill bit are basically equal; when the drilling tool passes through the upper reservoir and enters the mudstone, the lower gamma value becomes smaller, and the upper gamma value is greater than the lower gamma value, and the lower gamma value is greater than the gamma value at the drill bit. It is necessary to adjust the drill bit in real time to the downward reservoir direction and return to the reservoir; when the drilling tool passes through the lower reservoir and enters the mudstone, the upper gamma value becomes smaller, and the upper gamma value is less than the lower gamma value, and the lower gamma value is less than the gamma value at the drill bit. It is necessary to adjust the drill bit in real time to the upward reservoir direction and return to the reservoir.

[0095] In order to locate the drilling tool in a complex reservoir and overcome the problem of inaccurate trajectory control, the present invention proposes a drill bit gamma parameter prediction and wellbore trajectory control system based on BP neural network technology. Taking multiple engineering / geological parameters while drilling as input, the BP neural network nonlinear learning predicts the upper and lower gamma values ​​of the drill bit, and combines the drilling design, exploration well and seismic data to calculate the position of the tool in the reservoir, guide the adjustment of drilling parameters, and achieve accurate control of the wellbore trajectory.

[0096] Embodiment 2

[0097] Figure 5 Schematic diagram of the structure of a wellbore trajectory control device based on BP neural network provided in the second embodiment of the present invention. Figure 5 As shown, the device comprises:

[0098] A parameter acquisition unit 510 is used to acquire at least one standard characteristic parameter from a target well to be drilled and a well within a preset range of the target well to be drilled, wherein the standard characteristic parameter represents a parameter that affects a gamma value;

[0099] The gamma value output unit 520 is used to input the characteristic parameters into a pre-trained gamma value prediction model based on a BP neural network to obtain an upper gamma value above the drill bit, a lower gamma value below the drill bit, and a gamma value at the drill bit;

[0100] The trajectory control unit 530 is used to adjust the drilling direction of the drill bit based on the upper gamma value, the lower gamma value and the gamma value at the drill bit to control the wellbore trajectory.

[0101] Optionally, the device further includes: a model building training unit 540;

[0102] The model building training unit 540 is used to build a gamma value prediction model in the following manner:

[0103] Acquire at least one sample standard characteristic parameter combination and construct a test set, a training set and a validation set, wherein the sample standard characteristic parameter combination includes a sample standard characteristic parameter and a corresponding upper gamma value, a lower gamma value and a gamma value at a drill bit;

[0104] The number of input layer nodes, the number of hidden layers, the number of hidden layer nodes, the number of output layer nodes and the training function of the BP neural network are set to construct the gamma value prediction model, wherein the number of output layer nodes is at least 3, corresponding to the upper gamma value, the lower gamma value and the gamma value at the drill bit, respectively.

[0105] Optionally, the model building training unit 540 is used to train the gamma value prediction model in the following manner:

[0106] Initialize the weights and biases randomly, train the training set according to the initial weights, and obtain predicted values; compare the predicted values ​​with the actual gamma value, the lower gamma value, and the gamma value at the drill bit, calculate the output error of each node, and adjust the weights and thresholds;

[0107] Repeating the training with the training set and performing error verification of the predicted value with the verification set, and completing the training when the error is within a preset range;

[0108] The generalization ability of the model and the prediction accuracy of gamma are evaluated through the test set, and the optimal training model is finalized.

[0109] Optionally, the gamma value output unit 520, when adjusting the weight and the threshold, specifically performs:

[0110] Starting from the output layer, the output error of each layer of neurons is calculated layer by layer. Through the error function, the weights and thresholds of each layer are adjusted and corrected, so that the final output of the network gradually approaches the expected value.

[0111] Optionally, the trajectory control unit 530 is specifically configured to execute:

[0112] When the upper gamma value is greater than the lower gamma value and the lower gamma value is greater than the gamma value at the drill bit, adjusting the drill bit toward the reservoir;

[0113] When the upper gamma value is smaller than the lower gamma value and the lower gamma value is smaller than the gamma value at the drill bit, the drill bit is adjusted toward the upper reservoir direction.

[0114] The wellbore trajectory control device based on BP neural network provided in the embodiment of the present invention can execute the wellbore trajectory control method based on BP neural network provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0115] Embodiment 3

[0116] Figure 6 A schematic diagram of the structure of an electronic device 10 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0117] like Figure 6As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0118] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0119] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as a wellbore trajectory control method based on a BP neural network.

[0120] In some embodiments, the borehole trajectory control method based on the BP neural network can be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the borehole trajectory control method based on the BP neural network described above can be executed. Alternatively, in other embodiments, the processor 11 can be configured to execute the borehole trajectory control method based on the BP neural network in any other appropriate manner (for example, by means of firmware).

[0121] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0122] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0123] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0124] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0125] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0126] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0127] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0128] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A wellbore trajectory control method based on BP neural network, characterized in that: include: Acquire at least one standard characteristic parameter from a target well to be drilled and a well within a preset range of the target well to be drilled, wherein the standard characteristic parameter represents a parameter that has an impact on the gamma value; Inputting the characteristic parameters into a pre-trained gamma value prediction model based on a BP neural network to obtain an upper gamma value above the drill bit, a lower gamma value below the drill bit, and a gamma value at the drill bit; The drilling direction of the drill bit is adjusted based on the upper gamma value, the lower gamma value and the gamma value at the drill bit to control the wellbore trajectory.

2. The method according to claim 1, characterized in that The gamma value prediction model is constructed in the following way: Acquire at least one sample standard characteristic parameter combination and construct a test set, a training set and a validation set, wherein the sample standard characteristic parameter combination includes a sample standard characteristic parameter and a corresponding upper gamma value, a lower gamma value and a gamma value at a drill bit; The number of input layer nodes, the number of hidden layers, the number of hidden layer nodes, the number of output layer nodes and the training function of the BP neural network are set to construct the gamma value prediction model, wherein the number of output layer nodes is at least 3, corresponding to the upper gamma value, the lower gamma value and the gamma value at the drill bit, respectively.

3. The method according to claim 2, characterized in that The number of nodes in the hidden layer is determined by the following formula: h = int(a*I+b*U+c); Among them, I is the number of nodes in the input layer, U is the number of nodes in the output layer, h is the number of nodes in the hidden layer, int means rounding, and a, b, c are empirical adjustment constants.

4. The method according to claim 1, characterized in that: The gamma value prediction model is trained in the following way: Initialize the weights and biases randomly, train the training set according to the initial weights, and obtain predicted values; compare the predicted values ​​with the actual gamma value, the lower gamma value, and the gamma value at the drill bit, calculate the output error of each node, and adjust the weights and thresholds; Repeating the training with the training set and performing error verification of the predicted value with the verification set, and completing the training when the error is within a preset range; The generalization ability of the model and the prediction accuracy of gamma are evaluated through the test set, and the optimal training model is finalized.

5. The method according to claim 4, characterized in that The calculating the output error of each node includes: According to the input parameters at time t, the predicted upper gamma value, lower gamma value and gamma value at the drill bit at time t+1 are obtained; The error values ​​of the upper gamma value, the lower gamma value and the gamma value at the drill bit are determined according to the target values ​​of the upper gamma value, the lower gamma value and the gamma value at the drill bit and the predicted upper gamma value, the lower gamma value and the gamma value at the drill bit.

6. The method according to claim 4, characterized in that The adjusting weights and thresholds includes: Starting from the output layer, the output error of each layer of neurons is calculated layer by layer. Through the error function, the weights and thresholds of each layer are adjusted and corrected, so that the final output of the network gradually approaches the expected value.

7. The method according to claim 1, characterized in that The step of adjusting the drilling direction of the drill bit based on the upper gamma value, the lower gamma value and the gamma value at the drill bit to control the wellbore trajectory comprises: When the upper gamma value is greater than the lower gamma value and the lower gamma value is greater than the gamma value at the drill bit, adjusting the drill bit toward the reservoir; When the upper gamma value is smaller than the lower gamma value and the lower gamma value is smaller than the gamma value at the drill bit, the drill bit is adjusted toward the upper reservoir direction.

8. A wellbore trajectory control device based on BP neural network, characterized in that: include: A parameter acquisition unit, used to acquire at least one standard characteristic parameter from a target well to be drilled and a well within a preset range of the target well to be drilled, wherein the standard characteristic parameter represents a parameter that affects a gamma value; A gamma value output unit, used for inputting the characteristic parameters into a pre-trained gamma value prediction model based on a BP neural network to obtain an upper gamma value above the drill bit, a lower gamma value below the drill bit, and a gamma value at the drill bit; A trajectory control unit is used to adjust the drilling direction of the drill bit based on the upper gamma value, the lower gamma value and the gamma value at the drill bit to control the wellbore trajectory.

9. An electronic device, characterized in that The electronic device comprises: At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the wellbore trajectory control method based on the BP neural network as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the wellbore trajectory control method based on a BP neural network as described in any one of claims 1 to 7 when executed.

11. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the computer program implements the wellbore trajectory control method based on the BP neural network according to any one of claims 1 to 7.

Images