Drilling machine drilling speed prediction network training method and application method

Abstract

The invention provides a drilling machine drilling speed prediction network training method and an application method, and belongs to the technical field of drilling engineering optimization, and the training method comprises the steps: carrying out the data preprocessing of the obtained drilling machine logging data, and obtaining the drilling speed prediction training data; constructing an initial population of the improved dung beetle algorithm according to network parameters of the initial drilling machine drilling speed prediction network, and sequentially performing dung beetle behavior simulation population individual updating, golden sine algorithm population individual updating and dynamic subtraction factor algorithm population individual updating on the initial population to obtain a filial generation population; and determining population individual fitness according to the offspring population, performing population screening on the offspring population according to the population individual fitness to obtain an optimized population, and performing iterative optimization to obtain a well drilling machine drilling speed prediction network which is completely trained. Through the dung beetle algorithm improved by the golden sine algorithm and the dynamic subtraction factor, the calculation amount in the training process can be reduced, the training accuracy is improved, and the training efficiency and accuracy are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of drilling engineering optimization, and in particular to a drilling machinery drilling speed prediction network training method and an application method. Background Art

[0002] Drilling engineering is a key link in the process of oil and gas exploration and development, involving complex operations from the surface to the target formation. With the increasing depletion of oil and gas resources and the increasing difficulty of extraction, improving drilling efficiency, reducing drilling costs, and ensuring the safety of drilling operations have become the main concerns of the industry. In drilling engineering, the mechanical drilling speed refers to the depth drilled by the drill bit per unit time, which is an important indicator for measuring drilling efficiency. The optimization of mechanical drilling speed can not only improve drilling efficiency, but also effectively reduce drilling costs and reduce well control risks, thereby ensuring the safety of the entire drilling operation.

[0003] With the development of machine learning technology, data-driven models are gradually applied to the prediction of mechanical drilling speed. Machine learning can automatically identify complex nonlinear relationships and make predictions by learning a large amount of historical data. However, due to the complex influencing factors of mechanical drilling speed, such as drill bit diameter, drill bit weight, drill bit wear, rotation speed, drilling depth, formation strength, hydraulic parameters and many other factors, the mechanical drilling speed will be affected. Considering multiple types of influencing factors in the prediction process can improve the accuracy of the model to a certain extent, but at the same time, the amount of data in the training process will become very large and the calculation will be complicated. Existing training methods such as stochastic gradient descent methods are often difficult to find suitable network parameters efficiently and accurately when faced with huge amounts of data and complex calculations.

[0004] Therefore, there are technical problems in the existing technology, such as the huge amount of data and complex calculations during model training, and the difficulty of existing methods in finding suitable network parameters efficiently and accurately, which need to be improved. Summary of the invention

[0005] In view of this, it is necessary to provide a drilling machinery drilling speed prediction network training method and application method to solve the technical problems existing in the prior art that the training process has a large amount of data, complex calculations, and the existing methods are difficult to find suitable network parameters efficiently and accurately.

[0006] In order to solve the above technical problems, on the one hand, the present invention provides a drilling machinery drilling speed prediction network training method, comprising: Performing data preprocessing on the acquired drilling machinery logging data to obtain drilling speed prediction training data; The drilling speed prediction training data is input into the initial drilling machinery drilling speed prediction network, and the initial population of the improved dung beetle algorithm is constructed according to the network parameters of the initial drilling machinery drilling speed prediction network. The initial population is updated in turn by dung beetle behavior simulation population individual update, golden sine algorithm population individual update and dynamic subtraction factor algorithm population individual update to obtain the offspring population, and the population individual fitness is determined according to the offspring population. The offspring population is screened according to the population individual fitness to obtain the optimized population, and the iterative optimization is performed to obtain a fully trained drilling machinery drilling speed prediction network.

[0007] In a possible implementation, data preprocessing is performed on the acquired drilling machinery logging data to obtain drilling speed prediction training data, including: The acquired drilling machinery logging data is sequentially subjected to outlier removal, mixed sampling, data smoothing and data normalization to obtain the drilling speed prediction training data.

[0008] In a possible implementation, an initial population of the improved dung beetle algorithm is constructed according to the network parameters of the initial drilling mechanical drilling speed prediction network, including: The network parameters of the initial drilling mechanical drilling speed prediction network are used as the optimization target of the improved dung beetle algorithm, and the initial population search space range is determined according to the network parameters. Based on the Sobol sequence method, the initial population of the improved dung beetle algorithm is generated in the initial population search space.

[0009] In a possible implementation, the behavior of dung beetles simulates the update of individual populations, including: The initial population is subjected to rolling behavior simulation, dancing behavior simulation, foraging behavior simulation, stealing behavior simulation and breeding behavior simulation in sequence to update the population individuals of the initial population.

[0010] In a possible implementation, updating the positions of individuals in the initial population based on the golden sine algorithm includes: Based on the golden section number, the search space of the initial population is reduced to obtain the optimized search space; Based on the sine function, the individuals in the initial population are updated by performing individual optimization in the optimization search space.

[0011] In a possible implementation, the dynamic subtraction factor algorithm population individual update includes: Randomly generate a first random vector and a second random vector within a population individual optimization range of the initial population; Subtracting the first random vector from the second random vector and then multiplying the resultant vector by a preset dynamic coefficient to obtain a dynamic subtraction factor; Update the population individuals of the initial population according to the dynamic subtraction factor.

[0012] In a possible implementation, determining the fitness of individuals in a population according to the offspring population includes: Substituting the offspring population into the initial drilling machinery drilling speed prediction network to obtain the optimized drilling machinery drilling speed prediction network; The mechanical drilling speed is predicted according to the optimized drilling mechanical drilling speed prediction network to obtain the predicted mechanical drilling speed; The predicted mechanical drilling speed is denormalized and the average relative error is calculated with the drilling speed prediction training data to obtain the individual fitness of the offspring population.

[0013] On the other hand, the present invention also provides a network application method for predicting drilling speed of drilling machinery, comprising: Obtain the drilling mechanical logging data to be predicted; Input drilling machinery logging data into the well-trained drilling machinery drilling speed prediction network to obtain the mechanical drilling speed prediction result; The fully trained drilling machinery drilling speed prediction network is determined according to the above-mentioned drilling machinery drilling speed prediction network training method.

[0014] On the other hand, the present invention also provides an electronic device, including a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the above-mentioned drilling machinery drilling speed prediction network training method and / or the above-mentioned drilling machinery drilling speed prediction network application method are implemented.

[0015] On the other hand, the present invention also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the above-mentioned drilling machinery drilling speed prediction network training method and / or the above-mentioned drilling machinery drilling speed prediction network application method are implemented.

[0016] The beneficial effects of the present invention are as follows: the drilling machinery drilling speed prediction network training method provided by the embodiment of the present invention first pre-processes the acquired drilling machinery logging data to obtain drilling speed prediction training data; then inputs the drilling speed prediction training data into the initial drilling machinery drilling speed prediction network, constructs the initial population of the improved dung beetle algorithm according to the network parameters of the initial drilling machinery drilling speed prediction network, sequentially updates the dung beetle behavior simulation population individual update, the golden sine algorithm population individual update and the dynamic subtraction factor algorithm population individual update to the initial population to obtain the offspring population, determines the population individual fitness according to the offspring population, screens the offspring population according to the population individual fitness to obtain the optimized population, and iterates and optimizes to obtain the fully trained drilling machinery drilling speed prediction network. The present invention trains the drilling machinery drilling speed prediction network through the dung beetle algorithm, which can reduce the amount of calculation in the training process and improve the efficiency of model training. The dung beetle algorithm is improved through the golden sine algorithm and the dynamic subtraction factor, which can improve the accuracy of training and improve the efficiency and accuracy of the drilling machinery drilling speed prediction network training. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] 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 those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0018] Figure 1 A schematic flow chart of an embodiment of a drilling machinery drilling speed prediction network training method provided by the present invention; Figure 2 A schematic diagram of a process for constructing an initial population of an improved dung beetle algorithm according to an embodiment of the present invention; Figure 3 Schematic diagram of the process of the golden sine algorithm according to an embodiment of the present invention; Figure 4 Schematic diagram of the flow of a dynamic subtraction factor algorithm according to an embodiment of the present invention; Figure 5 A schematic diagram of a process for calculating the fitness of individuals in a population according to an embodiment of the present invention; Figure 6 A schematic flow chart of an embodiment of a network application method for predicting drilling speed of drilling machinery provided by the present invention; Figure 7 A schematic structural diagram of an embodiment of an electronic device provided by the present invention. DETAILED DESCRIPTION

[0019] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions 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 those skilled in the art without creative work are within the scope of protection of the present invention.

[0020] In the description of the embodiments of the present invention, unless otherwise specified, "multiple" means two or more than two. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" may mean: A exists alone, A and B exist at the same time, and B exists alone.

[0021] The descriptions of "first" and "second" in the embodiments of the present invention are only used for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the technical features defined as "first" and "second" may explicitly or implicitly include at least one of the features.

[0022] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0023] The present invention provides a drilling machinery drilling speed prediction network training method, application method electronic device and medium, which are respectively described below.

[0024] Figure 1 A flow chart of an embodiment of a drilling machinery drilling speed prediction network training method provided by the present invention is shown as follows: Figure 1 As shown, the drilling machinery drilling speed prediction network training method includes: S101, performing data preprocessing on the acquired drilling machinery logging data to obtain drilling speed prediction training data; S102, input the drilling speed prediction training data into the initial drilling machinery drilling speed prediction network, construct the initial population of the improved dung beetle algorithm according to the network parameters of the initial drilling machinery drilling speed prediction network, perform individual updates of the dung beetle behavior simulation population, the golden sine algorithm population and the dynamic subtraction factor algorithm population in turn on the initial population to obtain the offspring population, determine the individual fitness of the population according to the offspring population, screen the offspring population according to the individual fitness of the population to obtain the optimized population, and iterate the optimization to obtain a fully trained drilling machinery drilling speed prediction network.

[0025] Specifically, the embodiment constructs an initial drilling machinery drilling speed prediction network based on a bidirectional long short-term memory network. The long short-term memory network is a special type of recurrent neural network. It introduces a gating mechanism to solve the gradient vanishing or gradient exploding problems that are prone to occur in traditional RNNs when processing long sequences. The bidirectional long short-term memory network is based on the long short-term memory network, and captures the bidirectional contextual information in the sequence data by running two long short-term memory network layers, a forward long short-term memory network layer and a backward long short-term memory network layer, at the same time. The forward long short-term memory network layer processes the data from front to back in the natural order of the sequence, while the backward long short-term memory network layer processes the data from back to front in the reverse order of the sequence. In this way, the output of each time step contains information from both the past and the future, so that the contextual relationship in the sequence can be captured more comprehensively.

[0026] At the same time, the embodiment also integrates the self-attention mechanism on the basis of the bidirectional long short-term memory network to further improve the model performance. In the prediction of drilling speed of drilling machinery, the input received by the neural network is a lot of vectors of different sizes, and there are certain connections between the vectors. However, in actual training, the relationship between these inputs cannot be fully learned, resulting in poor model performance. The self-attention mechanism can solve this problem and pay attention to the input features and the relationship between them.

[0027] In addition, in view of the large amount of data, the bidirectional long short-term memory network has many hyperparameters, so the embodiment chooses to use the dung beetle optimization algorithm with strong search ability and fast convergence speed to automatically find the optimal parameters. In the optimization process, due to the complexity of the model and the large amount of parameters, if a lot of dung beetle particles are selected to search many times, the optimization time will be greatly increased, so only a relatively small number of particles and iterations can be selected. Under this premise, it is necessary to ensure that the initial population is evenly distributed, the global exploration and local development capabilities are strong, and it is not easy to fall into the local optimal solution. Therefore, considering these problems, the present invention has made a series of improvements to the dung beetle optimization algorithm, including Sobol sequence initialization population position, fusion golden sine algorithm and fusion dynamic subtraction factor.

[0028] Compared with the prior art, the drilling speed prediction network training method provided by the embodiment of the present invention first pre-processes the acquired drilling speed logging data to obtain the drilling speed prediction training data; then inputs the drilling speed prediction training data into the initial drilling speed prediction network, constructs the initial population of the improved dung beetle algorithm according to the network parameters of the initial drilling speed prediction network, sequentially updates the dung beetle behavior simulation population, the golden sine algorithm population and the dynamic subtraction factor algorithm population to obtain the offspring population, determines the population individual fitness according to the offspring population, screens the offspring population according to the population individual fitness to obtain the optimized population, and iterates and optimizes to obtain the fully trained drilling speed prediction network. The present invention trains the drilling speed prediction network of the drilling machinery through the dung beetle algorithm, which can reduce the amount of calculation in the training process and improve the efficiency of model training. The dung beetle algorithm is improved through the golden sine algorithm and the dynamic subtraction factor, which can improve the accuracy of training and improve the efficiency and accuracy of the drilling speed prediction network training of the drilling machinery.

[0029] In some embodiments of the present invention, data preprocessing is performed on the acquired drilling machinery logging data to obtain drilling speed prediction training data, including: The acquired drilling machinery logging data is sequentially subjected to outlier removal, mixed sampling, data smoothing and data normalization to obtain the drilling speed prediction training data.

[0030] Specifically, in the process of establishing training data, the embodiment first obtains a large amount of drilling machinery logging data from the oil and gas field drilling construction site, and organizes it as the original data.

[0031] Then, in order to ensure the effectiveness of the training data in the training model process, the original data needs to be preprocessed to construct the drilling speed prediction training data. The preprocessing process includes: first, the interquartile range method is used to detect and remove outliers in the original data. Then, undersampling and oversampling are used to mix the data to ensure the uniformity and authenticity of the data distribution. The Daubechies 4 wavelet transform function is then used to smooth the data, and the standard deviation is used as the threshold. The nonlinear threshold processing method effectively removes the high-frequency noise while retaining the main signal characteristics. At the same time, considering the dimensional differences of different types of data, all data are normalized by the maximum and minimum normalization algorithm. Finally, the processed data is divided into a training set and a test set to obtain the drilling speed prediction training data, in which the relevant data contained in each feature attribute is used as the input variable X; the label to be predicted is the mechanical drilling speed, which is used as the output variable Y.

[0032] The maximum and minimum normalization formula can be expressed as:

[0033] in, is the normalized data, represents the original sample data before normalization, Represents the minimum value of the original data, Indicates the maximum value of the original data.

[0034] In some embodiments of the present invention, Figure 2 FIG. 1 is a flow chart of constructing an initial population of an improved dung beetle algorithm according to an embodiment of the present invention. Figure 2 As shown in FIG. 1 , the initial population of the improved dung beetle algorithm is constructed according to the network parameters of the initial drilling mechanical drilling speed prediction network, including: S201, taking the network parameters of the initial drilling mechanical drilling speed prediction network as the optimization target of the improved dung beetle algorithm, and determining the initial population search space range according to the network parameters; S202. Generate an initial population of the improved dung beetle algorithm in the initial population search space based on the Sobol sequence method.

[0035] Specifically, in the process of constructing the initial population, the embodiment uses the network parameters of the initial drilling mechanics drilling speed prediction network as the optimization target of the improved dung beetle algorithm, including the hidden layer, initial learning rate, batch size, number of network stacks, L1 regularization coefficient and Dropout coefficient in the drilling mechanics drilling speed prediction network. In addition, in order to make the initial population have a better distribution effect, the embodiment uses the Sobol sequence method to construct the initial population. The Sobol sequence is a low-variance deterministic sequence. Compared with the traditional random number generator, the Sobol sequence has a better uniform distribution property and can provide a faster convergence speed at the same number of points.

[0036] In some embodiments of the present invention, the behavior of dung beetles simulates the renewal of population individuals, including: The initial population is subjected to rolling behavior simulation, dancing behavior simulation, foraging behavior simulation, stealing behavior simulation and breeding behavior simulation in sequence to update the population individuals of the initial population.

[0037] Specifically, the dung beetle algorithm is a swarm intelligence optimization algorithm inspired by nature, which is inspired by the behavior and ecological habits of dung beetles. The algorithm has the advantages of strong search ability and fast convergence speed, and can solve complex problems. In the dung beetle algorithm, the population individual update is carried out through the simulation of dung beetle behavior, including rolling behavior simulation, dancing behavior simulation, foraging behavior simulation, stealing behavior simulation and reproduction behavior simulation.

[0038] The rolling ball behavior simulation simulates the behavior of dung beetles rolling dung balls. Each dung beetle individual moves in a given direction and updates its position in the search space. This step achieves global exploration by adjusting the position of the individual to find a better solution. The formula is expressed as:

[0039] in, Indicates the current iteration number, Indicates The iteration The location information of dung beetles. is the deflection coefficient, is the light intensity coefficient, is a natural coefficient with a value of 1 or -1. is the global worst position, To simulate light intensity changes.

[0040] Dance behavior simulation refers to simulating the dance behavior of dung beetles when they encounter obstacles or need to re-determine their direction. The dung beetle will climb to the top of the dung ball, re-determine the direction of movement through a series of rotations and pauses, and then continue to roll the dung ball. This helps avoid local optimal solutions and enhances the algorithm's search ability. The formula is expressed as:

[0041] Foraging behavior simulation simulates the process of dung beetles looking for food. Dung beetles will move in the search space and try to find a better solution. This step combines global exploration and local development strategies to further optimize the quality of the solution.

[0042] The stealing behavior simulation simulates the behavior of dung beetles stealing dung balls from other dung beetles. In a dung beetle population, some dung beetle individuals will try to obtain better solutions from other individuals, thereby enhancing the search ability of the entire population. This behavior can prevent the algorithm from falling into local optimality and improve global search performance. The formula is expressed as:

[0043] in, Indicates A thief The position information of the iteration, is a random vector of normally distributed size, is a constant.

[0044] Reproduction behavior simulation simulates the reproduction process of dung beetles, generating new dung beetle individuals through operations such as selection, crossover, and mutation. These new individuals increase diversity in the population, help explore more search space, and avoid falling into local optimal solutions.

[0045] In some embodiments of the present invention, Figure 3 FIG. 1 is a flow chart of the golden sine algorithm according to an embodiment of the present invention. Figure 3 As shown, the individual positions of the initial population are updated based on the golden sine algorithm, including: S301, reducing the search space of the initial population based on the golden section number to obtain an optimized search space; S302, performing individual optimization in the optimization search space based on the sine function to update the population individuals of the initial population.

[0046] Specifically, in order to ensure that a better parameter combination is optimized and the optimization time is reduced, the embodiment also adds a golden sine algorithm to the dung beetle optimization algorithm. The golden sine algorithm uses a sine function to find the optimal value and uses the golden section number to narrow the search space and improve the search speed. The formula is expressed as follows:

[0047] in, Indicates Individuals of a population The position of the iteration, Indicates The individual in The optimal position of the iteration, for A random number in the range, for A random number in the range, and is the golden ratio coefficient.

[0048] In some embodiments of the present invention, Figure 4 FIG. 1 is a flow chart of a dynamic subtraction factor algorithm according to an embodiment of the present invention. Figure 4 As shown, the dynamic subtraction factor algorithm population individual update includes: S401, randomly generating a first random vector and a second random vector within a population individual optimization range of an initial population; S402, subtracting the first random vector from the second random vector and multiplying the resultant result by a preset dynamic coefficient to obtain a dynamic subtraction factor; S403. Update the population individuals of the initial population according to the dynamic subtraction factor.

[0049] Specifically, in order to prevent falling into the local optimal solution during the optimization process, a dynamic subtraction factor is integrated into the position update of the dung beetle optimization algorithm. That is, each time the position is updated, a subtraction factor is added to the original position update method to obtain a better new position.

[0050] The dynamic subtraction factor algorithm generates two random vectors within the optimization range, multiplies them by the preset dynamic coefficient, and obtains the dynamic subtraction factor. The dynamic subtraction factor is used to update the position of the individual population. The preset dynamic coefficient can dynamically adjust the size of the subtraction factor. As the number of iterations increases, the obtained parameters are closer to the optimal parameters, and the subtraction factor used gradually decreases. The dynamic subtraction factor formula is expressed as follows:

[0051]

[0052]

[0053] in, is a random vector, and Indicates the boundary value of the parameter that needs to be optimized. is a randomly generated vector. The number of vectors is the same as the number of parameters to be optimized. The element values ​​are Random sampling of the interval is obtained. It is the subtraction factor coefficient, which can be customized. is the dynamic coefficient, Indicates the current iteration number, is the maximum number of iterations.

[0054] In some embodiments of the present invention, Figure 5FIG. 1 is a flow chart of calculating the fitness of individuals in a population according to an embodiment of the present invention. Figure 5 As shown, the fitness of individuals in the population is determined based on the offspring population, including: S501, substituting the offspring population into the initial drilling machinery drilling speed prediction network to obtain an optimized drilling machinery drilling speed prediction network; S502, predicting the mechanical drilling speed according to the optimized drilling mechanical drilling speed prediction network to obtain a predicted mechanical drilling speed; S503, after denormalizing the predicted mechanical drilling speed, the average relative error is calculated with the drilling speed prediction training data to obtain the individual fitness of the offspring population.

[0055] Specifically, after each round of population update, it is necessary to calculate the fitness value of each individual. To this end, the embodiment substitutes the updated population into the drilling mechanical drilling speed prediction network, predicts the mechanical drilling speed to obtain the predicted mechanical drilling speed, and calculates the average relative error between the predicted mechanical drilling speed and the training data set after denormalization as the individual fitness of the population.

[0056] The anti-normalization formula is expressed as:

[0057] in, is the predicted mechanical drilling speed output by the model, is the predicted drilling speed value obtained after denormalization.

[0058] The average relative error formula is expressed as:

[0059] in, is the actual drilling speed value, The predicted drilling speed value.

[0060] Embodiments use the calculated fitness values ​​of dung beetle individuals to select individuals with better performance and retain them in the population, and eliminate individuals with poorer fitness. In this way, the population gradually converges to the optimal solution.

[0061] In summary, in order to reduce the amount of calculation and improve the training speed, the present invention first performs data preprocessing on the acquired drilling machinery logging data to obtain drilling speed prediction training data; then the drilling speed prediction training data is input into the initial drilling machinery drilling speed prediction network, and the initial population of the improved dung beetle algorithm is constructed according to the network parameters of the initial drilling machinery drilling speed prediction network, and the initial population is sequentially updated by dung beetle behavior simulation population individual update, golden sine algorithm population individual update and dynamic subtraction factor algorithm population individual update to obtain the offspring population, and the population individual fitness is determined according to the offspring population, and the offspring population is screened according to the population individual fitness to obtain the optimized population, and iterative optimization is performed to obtain a fully trained drilling machinery drilling speed prediction network. The present invention trains the drilling machinery drilling speed prediction network through the dung beetle algorithm, which can reduce the amount of calculation in the training process and improve the efficiency of model training. The dung beetle algorithm is improved by the golden sine algorithm and the dynamic subtraction factor, which can improve the accuracy of training and improve the efficiency and accuracy of the drilling machinery drilling speed prediction network training.

[0062] The present invention also provides a network application method for predicting drilling speed of drilling machinery, combining Figure 6 Come and see, Figure 6 A flow chart of an embodiment of a network application method for predicting drilling speed of drilling machinery provided by the present invention is shown as follows: Figure 6 As shown, the network application method for predicting drilling speed of drilling machinery includes: S601, obtaining drilling mechanical logging data to be predicted; S602, inputting the drilling machinery logging data into a well-trained drilling machinery drilling speed prediction network to obtain a mechanical drilling speed prediction result; The fully trained drilling machinery drilling speed prediction network is determined according to the above-mentioned drilling machinery drilling speed prediction network training method.

[0063] In an embodiment of the present invention, first, the drilling machinery logging data to be predicted is effectively acquired, and then the drilling machinery drilling speed is effectively predicted using the above-mentioned fully trained drilling machinery drilling speed prediction network, and the mechanical drilling speed prediction result can be output.

[0064] like Figure 7 As shown, the present invention also provides an electronic device 700. The electronic device 700 includes a processor 701, a memory 702 and a display 703. Figure 7 Only some components of the electronic device 700 are shown, but it should be understood that it is not required to implement all of the components shown, and more or fewer components may be implemented instead.

[0065] In some embodiments, the processor 701 may be a central processing unit (CPU), a microprocessor or other data processing chip, which is used to run the program code or process data stored in the memory 702, such as the drilling machinery drilling speed prediction network training method and / or the drilling machinery drilling speed prediction network application method in the present invention.

[0066] In some embodiments, the processor 701 may be a single server or a server group. The server group may be centralized or distributed. In some embodiments, the processor 701 may be local or remote. In some embodiments, the processor 701 may be implemented in a cloud platform. In one embodiment, the cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an internal cloud, a multi-cloud, etc., or any combination thereof.

[0067] In some embodiments, the memory 702 may be an internal storage unit of the electronic device 700, such as a hard disk or memory of the electronic device 700. In other embodiments, the memory 702 may also be an external storage device of the electronic device 700, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc., equipped on the electronic device 700.

[0068] Furthermore, the memory 702 may include both an internal storage unit of the electronic device 700 and an external storage device. The memory 702 is used to store application software installed in the electronic device 700 and various data.

[0069] In some embodiments, the display 703 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch device, etc. The display 703 is used to display information of the electronic device 700 and to display a visual user interface. The components 701-703 of the electronic device 700 communicate with each other via a system bus.

[0070] In one embodiment, when the processor 701 executes the drilling machinery penetration rate prediction network training program in the memory 702, the following steps may be implemented: Performing data preprocessing on the acquired drilling machinery logging data to obtain drilling speed prediction training data; The drilling speed prediction training data is input into the initial drilling machinery drilling speed prediction network, and the initial population of the improved dung beetle algorithm is constructed according to the network parameters of the initial drilling machinery drilling speed prediction network. The initial population is updated in turn by dung beetle behavior simulation population individual update, golden sine algorithm population individual update and dynamic subtraction factor algorithm population individual update to obtain the offspring population, and the population individual fitness is determined according to the offspring population. The offspring population is screened according to the population individual fitness to obtain the optimized population, and the iterative optimization is performed to obtain a fully trained drilling machinery drilling speed prediction network.

[0071] In one embodiment, when the processor 701 executes the drilling machinery penetration rate prediction network application in the memory 702, the following steps may be implemented: Obtain the drilling mechanical logging data to be predicted; The drilling machinery logging data is input into the well-trained drilling machinery drilling speed prediction network to obtain the mechanical drilling speed prediction result.

[0072] It should be understood that: when the processor 701 executes the drilling machinery drilling speed prediction network training program and / or the drilling machinery drilling speed prediction network application program in the memory 702, in addition to the above functions, other functions can also be realized. For details, please refer to the description of the corresponding method embodiments above.

[0073] Furthermore, the embodiment of the present invention does not specifically limit the type of the electronic device 700 mentioned, and the electronic device 700 may be a portable electronic device such as a mobile phone, a tablet computer, a personal digital assistant (PDA), a wearable device, a laptop computer, etc. Exemplary embodiments of portable electronic devices include but are not limited to portable electronic devices equipped with IOS, Android, Microsoft or other operating systems. The above-mentioned portable electronic devices may also be other portable electronic devices, such as a laptop computer with a touch-sensitive surface (e.g., a touch panel). It should also be understood that in some other embodiments of the present invention, the electronic device 700 may not be a portable electronic device, but a desktop computer with a touch-sensitive surface (e.g., a touch panel).

[0074] Accordingly, an embodiment of the present application also provides a computer-readable storage medium, which is used to store computer-readable programs or instructions. When the program or instructions are executed by a processor, it can implement the steps or functions in the drilling machinery drilling speed prediction network training method and / or drilling machinery drilling speed prediction network application method provided in the above-mentioned method embodiments.

[0075] Those skilled in the art will appreciate that all or part of the processes of the above-mentioned embodiments can be implemented by instructing related hardware (such as a processor, a controller, etc.) through a computer program, and the computer program can be stored in a computer-readable storage medium, wherein the computer-readable storage medium is a disk, an optical disk, a read-only storage memory, or a random access memory, etc.

[0076] The drilling machinery drilling speed prediction network training method, application method, electronic device and medium provided by the present invention are introduced in detail above. Specific examples are used in this article to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea; at the same time, for technical personnel in this field, according to the idea of ​​the present invention, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A drilling machine drilling speed prediction network training method, characterized in that: include: Performing data preprocessing on the acquired drilling machinery logging data to obtain drilling speed prediction training data; The drilling speed prediction training data is input into the initial drilling machinery drilling speed prediction network, and the initial population of the improved dung beetle algorithm is constructed according to the network parameters of the initial drilling machinery drilling speed prediction network. The initial population is sequentially updated with dung beetle behavior simulation population individuals, golden sine algorithm population individuals and dynamic subtraction factor algorithm population individuals to obtain a child population, the individual fitness of the population is determined according to the child population, the child population is screened according to the individual fitness of the population to obtain an optimized population, and iterative optimization is performed to obtain a fully trained drilling machinery drilling speed prediction network.

2. The drilling machinery drilling speed prediction network training method according to claim 1 is characterized in that: The method of preprocessing the acquired drilling machinery logging data to obtain drilling speed prediction training data includes: The acquired drilling machinery logging data is sequentially subjected to outlier removal, mixed sampling, data smoothing and data normalization to obtain the drilling speed prediction training data.

3. The drilling machinery drilling speed prediction network training method according to claim 1 is characterized in that: The initial population of the improved dung beetle algorithm is constructed according to the network parameters of the initial drilling mechanical drilling speed prediction network, including: Taking the network parameters of the initial drilling mechanical drilling speed prediction network as the optimization target of the improved dung beetle algorithm, and determining the initial population search space range according to the network parameters; The initial population of the improved dung beetle algorithm is generated within the initial population search space based on the Sobol sequence method.

4. The drilling machinery drilling speed prediction network training method according to claim 1 is characterized in that: The dung beetle behavior simulates the individual renewal of the population, including: The initial population is subjected to rolling behavior simulation, dancing behavior simulation, foraging behavior simulation, stealing behavior simulation and breeding behavior simulation in sequence to update the population individuals of the initial population.

5. The drilling machinery drilling speed prediction network training method according to claim 1 is characterized in that: The updating of the individual positions of the initial population based on the golden sine algorithm includes: The search space of the initial population is reduced based on the golden section number to obtain an optimized search space; The population individuals of the initial population are updated by performing individual optimization in the optimization search space based on a sine function.

6. The drilling machine drilling speed prediction network training method according to claim 1 is characterized in that: The dynamic subtraction factor algorithm population individual update includes: Randomly generate a first random vector and a second random vector within a population individual optimization range of the initial population; Subtracting the first random vector from the second random vector and multiplying the resultant result by a preset dynamic coefficient to obtain a dynamic subtraction factor; The population individuals of the initial population are updated according to the dynamic subtraction factor.

7. The drilling machinery drilling speed prediction network training method according to claim 1 is characterized in that: The step of determining the fitness of individuals in the population according to the offspring population comprises: Substituting the offspring population into the initial drilling machinery drilling speed prediction network to obtain an optimized drilling machinery drilling speed prediction network; Performing mechanical drilling speed prediction according to the optimized drilling mechanical drilling speed prediction network to obtain a predicted mechanical drilling speed; The predicted mechanical drilling speed is denormalized and the average relative error is calculated with the drilling speed prediction training data to obtain the population individual fitness of the offspring population.

8. A network application method for predicting drilling speed of drilling machinery, characterized in that: include: Obtain the drilling mechanical logging data to be predicted; Inputting the drilling machinery logging data into a well-trained drilling machinery drilling speed prediction network to obtain a mechanical drilling speed prediction result; Wherein, the fully trained drilling machinery drilling speed prediction network is determined according to the drilling machinery drilling speed prediction network training method according to any one of claims 1 to 7.

9. An electronic device comprising a processor, a memory, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the drilling machinery drilling speed prediction network training method according to any one of claims 1 to 7 and / or the drilling machinery drilling speed prediction network application method according to claim 8 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the drilling machinery drilling speed prediction network training method according to any one of claims 1 to 7 and / or the drilling machinery drilling speed prediction network application method according to claim 8 are implemented.

Images