Method and device for determining injection displacement in deepwater surface conduit well building process

Abstract

The invention relates to the technical field of deepwater oil and gas field development, and particularly discloses a method and device for determining the injection displacement in the deepwater surface layer guide pipe well building process, and the method comprises the steps that drilling parameters and stratum property parameters of deepwater surface layer injection well building operation are obtained; the incidence relation between the injection displacement and the mechanical drilling speed is determined on the basis of the drilling parameters and the stratum property parameters through the target BP neural network; establishing a surface bearing capacity recovery curve according to the drilling parameters and the stratum property parameters by adopting a target long-short-term memory neural network; the surface bearing capacity recovery curve is a curve of the surface bearing capacity changing along with time; and a particle swarm optimization algorithm is used for calculating the target injection displacement based on the incidence relation between the injection displacement and the mechanical drilling speed and the surface bearing capacity recovery curve. According to the scheme, the optimal injection displacement of deepwater surface conduit well building can be determined, the method has reference significance for deepwater surface well building design, and the safety and efficiency of deepwater surface efficient well drilling are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of deepwater oil and gas field development, and in particular to a method and device for determining the jet displacement during deepwater surface conduit well construction. Background Art

[0002] Deepwater oil and gas is a vital energy source, receiving increasing attention worldwide. The development and exploitation of deepwater oil and gas fields not only offers significant economic benefits but also possesses vast resource potential. With the continuous advancement of marine engineering operations and offshore oil and gas exploration and development technologies, the development of deepwater oil and gas resources has become increasingly feasible and economical. These technological developments have not only improved exploration and development efficiency, reduced greenhouse gas emissions, but also lowered investment costs, making the development of deepwater oil and gas resources even more attractive.

[0003] The surface conductor is the first subsea tubing string in deepwater oil and gas development. It provides a stable starting point for subsequent drilling operations and ensures the flow of drilling fluids, drill tools, production and testing tools, and oil and gas. Surface conductor installation is a crucial process in deepwater drilling. Drilling cementing, suction piles, and jetting are commonly used to install surface conductors. Drilling cementing requires complex drilling tools and a long operation cycle, and the cement used is highly contaminated by the seabed. The suction pile method is subject to soil properties and topographical conditions, and its stability is severely affected by strong seawater currents. The jetting method, on the other hand, uses no cement after the surface conductor is installed, relying instead on the bearing capacity provided by the backfill and compaction of the surrounding seabed soil for stability. The jetting method allows for rapid wellhead establishment and generates minimal waste. Its adaptability to a variety of seabed soil types, environmental friendliness, and high operational efficiency make it a popular method for deepwater surface well construction.

[0004] The principle of the surface pipe installation technology is to use water jet to destroy the soil to form a wellbore. After the surface pipe is installed, it relies on the soil to provide bearing capacity to maintain stability. If the hydraulic parameters of the jet are too small, the diameter of the wellbore formed by the jet is smaller than the diameter of the surface pipe, which seriously affects the efficiency of the jet operation. Figure 1 If the hydraulic parameters of the jet are too large, the diameter of the wellbore formed by the jet is significantly larger than the diameter of the surface pipe, and a long waiting time is required to ensure that the soil can stably support the pipe. Figure 1As shown in (b) in the figure. After the catheter is installed, it needs to be left to wait, and the waiting time is mostly obtained based on experience, without specific guidance methods. Too long a waiting time increases project costs, and too long a dynamic positioning of the platform wastes fuel and produces greenhouse gases. Too short a waiting time will cause the catheter to sink, resulting in operation failure. Therefore, accurately controlling the drilling jet displacement is the key to achieving safe, efficient and environmentally friendly drilling. However, the existing technology cannot accurately control the drilling jet displacement to balance the efficient catheter installation operation and the shortest safe waiting time in the deepwater surface well construction process.

[0005] To address the above issues, no effective solutions have been proposed so far. Summary of the Invention

[0006] The embodiments of this specification provide a method and apparatus for determining the jetting rate during deepwater surface well construction, to address the problem in the prior art of being unable to accurately control the drilling jetting rate to balance efficient catheter installation operations and the shortest safe waiting time during deepwater surface well construction.

[0007] The present invention provides a method for determining the injection rate during deepwater surface conductor well construction, comprising:

[0008] Obtain drilling parameters and formation property parameters for deepwater surface jet well construction operations;

[0009] Determining the correlation between the injection rate and the mechanical penetration rate based on the drilling parameters and the formation property parameters using a target BP neural network;

[0010] A target long short-term memory neural network is used to establish a surface bearing capacity recovery curve according to the drilling parameters and formation property parameters; the surface bearing capacity recovery curve is a curve showing the surface bearing capacity changing with time;

[0011] The target injection displacement is calculated by using a particle swarm optimization algorithm based on the correlation between the injection displacement and the mechanical drilling speed and the surface bearing capacity recovery curve.

[0012] In one embodiment, the target BP neural network is constructed in the following manner:

[0013] Determining the structure of a BP neural network and the number of neurons in an input layer, a hidden layer, and an output layer included in the BP neural network according to the physical relationship between the jet displacement and the mechanical drilling speed;

[0014] Drilling parameters and geological parameters are collected; the BP neural network is trained using the collected data to obtain the target BP neural network; and during the training process, the network weights of the BP neural network are adjusted through iterative optimization so that the correlation between the jet displacement and the mechanical penetration rate output by the BP neural network is close to the actual relationship between the jet displacement and the mechanical penetration rate.

[0015] In one embodiment, the target long short-term memory neural network is constructed in the following manner:

[0016] Determine the structure of the LSTM neural network and the number of LSTM units and layers;

[0017] Jet drilling parameters and waiting time are collected; the long short-term memory neural network is trained using the collected jet drilling parameters and waiting time to obtain a target long short-term memory neural network; during the training process, the network weights of the long short-term memory neural network are adjusted through iterative optimization so that the output of the long short-term memory neural network is close to the actual bearing capacity data.

[0018] In one embodiment, a target long short-term memory neural network is used to establish a surface bearing capacity recovery curve based on the drilling parameters and formation property parameters, including:

[0019] Establishing a surface bearing capacity recovery curve based on the bearing capacity data output by the target long short-term memory neural network;

[0020] According to the surface bearing capacity recovery curve, the static waiting time during the jet drilling waiting period is determined. The static waiting time refers to the shortest waiting time to ensure the safety of deepwater surface jet drilling operations.

[0021] In one embodiment, a target injection rate is calculated using a particle swarm optimization algorithm based on the correlation between the injection rate and the mechanical drilling rate and the surface bearing capacity recovery curve, including:

[0022] Initializing a randomly generated set of particles and calculating the corresponding fitness based on the performance indicators of each set of particles; each particle is used to represent a jet drilling plan, and the jet drilling plan includes a corresponding combination of jet displacement and static waiting time;

[0023] Based on the historical best position and global best position of each particle, its speed and position are updated, the new position of each particle is evaluated, and its fitness is calculated; if the fitness of the current particle is better than its historical best position, the individual best position is updated; if the fitness of the current particle is better than the global best position, the global best position is updated, and the iteration continues until the preset number of iterations is met or the fitness threshold is reached;

[0024] The global optimal position is output, and the optimal solution for the jet drilling installation phase and the static waiting phase is obtained.

[0025] In one embodiment, after calculating the target injection displacement by using a particle swarm optimization algorithm based on the correlation between the injection displacement and the mechanical drilling speed and the surface bearing capacity recovery curve, the method further includes:

[0026] Determining a corresponding target mechanical drilling speed based on the target injection displacement and the correlation between the injection displacement and the mechanical drilling speed;

[0027] Based on the target injection displacement, the target mechanical penetration rate and the surface bearing capacity recovery curve, the shortest well construction time and the real-time bearing capacity of the casing during the waiting period are calculated.

[0028] The present invention also provides a device for determining the jet displacement during deepwater surface conduit well construction, comprising:

[0029] An acquisition module is used to obtain drilling parameters and formation property parameters for deepwater surface jet well construction operations;

[0030] a determination module, configured to determine, using a target BP neural network, a correlation between the injection displacement and the mechanical penetration rate based on the drilling parameters and the formation property parameters;

[0031] An establishment module is used to establish a surface bearing capacity recovery curve based on the drilling parameters and formation property parameters using a target long short-term memory neural network; the surface bearing capacity recovery curve is a curve showing the surface bearing capacity changing over time;

[0032] The calculation module is used to calculate the target injection displacement based on the correlation between the injection displacement and the mechanical drilling speed and the surface bearing capacity recovery curve by using a particle swarm optimization algorithm.

[0033] An embodiment of this specification also provides a computer device, including a processor and a memory for storing processor-executable instructions. When the processor executes the instructions, the steps of the method for determining the injection displacement in the deepwater surface conductor well construction process described in any of the above embodiments are implemented.

[0034] An embodiment of this specification also provides a computer-readable storage medium having computer instructions stored thereon, which, when executed, implement the steps of the method for determining the injection rate in the deepwater surface conductor well construction process described in any of the above embodiments.

[0035] The embodiments of this specification also provide a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of the method for determining the injection rate in the deepwater surface conductor well construction process described in any of the above embodiments.

[0036] In an embodiment of the present specification, a method for determining the jetting rate in a deepwater surface conductor well construction process is provided. A target BP neural network is used to determine the correlation between the jetting rate and the mechanical penetration rate based on the drilling parameters and formation property parameters. A target long-short-term memory neural network is used to establish a surface bearing capacity recovery curve based on the drilling parameters and formation property parameters. The surface bearing capacity recovery curve is a curve showing the change of surface bearing capacity over time. A particle swarm optimization algorithm is used to calculate the target jetting rate based on the correlation between the jetting rate and the mechanical penetration rate and the surface bearing capacity recovery curve. In the above scheme, during the deepwater surface conductor well construction process, the BP neural network is used to optimize the surface conductor jet displacement to control the drilling speed, the long-short-term memory neural network is used to predict the surface conductor bearing capacity recovery with a confidence interval, and the particle swarm optimization algorithm is used to optimize the displacement to obtain the optimal deepwater surface conductor well construction flow rate. This scheme is verified through actual drilling cases, effectively balancing efficient conductor installation operations and the shortest safe waiting time during the deepwater surface well construction process. This provides a new method for deepwater surface conductor well construction, which is traditionally based on experience. It has reference significance for deepwater surface well construction design and improves the safety and efficiency of high-efficiency deepwater surface drilling. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The accompanying drawings described herein are used to provide a further understanding of this specification, constitute a part of this specification, and do not constitute a limitation of this specification. In the accompanying drawings:

[0038] Figure 1 A schematic diagram showing a scenario of installing a surface conduit by the jetting method;

[0039] Figure 2 A flow chart of a method for determining the injection rate in a deepwater surface conductor well construction process according to an embodiment of the present specification is shown;

[0040] Figure 3 A flow chart showing the determination of the upper and lower limits of the injection displacement based on the principles of fluid mechanics and soil mechanics in one embodiment of this specification is shown;

[0041] Figure 4 A schematic diagram of a BP neural network in an embodiment of this specification is shown;

[0042] Figure 5 A diagram showing the calculation principle of the particle swarm optimization algorithm in an embodiment of this specification is shown;

[0043] Figure 6 The overall flow chart of the method for determining the injection displacement during the deepwater surface conduit well construction process in one embodiment of the present specification is shown;

[0044] Figure 7 A schematic diagram of a device for determining the jet displacement during deepwater surface conduit well construction in an embodiment of this specification is shown;

[0045] Figure 8 A schematic diagram of a computer device in an embodiment of the present specification is shown. DETAILED DESCRIPTION

[0046] The principles and spirit of this specification will be described below with reference to several exemplary embodiments. It should be understood that these embodiments are provided solely to enable those skilled in the art to better understand and implement this specification, and are not intended to limit the scope of this specification in any way. Rather, these embodiments are provided to make this specification more thorough and complete, and to fully convey the scope of this disclosure to those skilled in the art.

[0047] Those skilled in the art will appreciate that the embodiments of this specification may be implemented as a system, device, method, or computer program product. Therefore, the disclosure herein may be implemented in the following forms: entirely in hardware, entirely in software (including firmware, resident software, microcode, etc.), or a combination of hardware and software.

[0048] The embodiments of this specification provide a method for determining the injection rate during the deepwater surface conductor well construction process. Figure 2 A flow chart of a method for determining the jet displacement in a deepwater surface conduit well construction process in an embodiment of this specification is shown. Although this specification provides method operation steps or device structures as shown in the following embodiments or drawings, more or fewer operation steps or module units may be included in the method or device based on routine or no creative work. In steps or structures where there is no necessary causal relationship logically, the execution order of these steps or the module structure of the device is not limited to the execution order or module structure described in the embodiments of this specification and shown in the drawings. When the method or module structure is applied to an actual device or terminal product, it can be connected according to the method or module structure shown in the embodiments or drawings for sequential execution or parallel execution (for example, in a parallel processor or multi-threaded processing environment, or even a distributed processing environment).

[0049] Specifically, if Figure 2 As shown, a method for determining the injection rate during deepwater surface conductor well construction provided by an embodiment of this specification may include the following steps:

[0050] Step S201: Acquire drilling parameters and formation property parameters for deepwater surface jet well construction operations.

[0051] The surface conduit is the first layer of tubing in deepwater oil and gas development. It provides a stable starting point for subsequent drilling operations and ensures the passage of drilling fluid, drilling tools, production and testing tools, and oil and gas. The installation of the surface conduit is a critical step in deepwater drilling. In this embodiment, the surface conduit is installed using a jetting method.

[0052] In the deepwater surface jet well construction operation to be studied, drilling parameters and formation property parameters can be obtained. Drilling parameters can include at least one of the following: weight on bit, drill bit extension, jet flow rate, drill bit water hole, etc. Formation property parameters can include at least one of the following: shear strength, recovery capacity, friction properties, and fracture pressure, etc.

[0053] Step S202: using the target BP neural network, determining the correlation between the jet displacement and the mechanical drilling rate based on the drilling parameters and the formation property parameters.

[0054] During the jetting installation phase of surface well construction, drilling parameters and geological conditions, among other factors, can impact the efficiency of catheter installation. BP (Back Propagation) neural networks, due to their ability to integrate the influence of multiple factors, offer advantages in calculating the relationship between flow rate and ROP. Therefore, after obtaining the drilling parameters and formation property parameters for deepwater surface jetting operations, a target BP neural network can be used to determine the correlation between flow rate and ROP based on these parameters. The target BP neural network can be a pre-trained BP neural network.

[0055] In some embodiments of the present specification, the target BP neural network is constructed in the following manner: according to the physical relationship between the jet displacement and the mechanical drilling speed, the structure of the BP neural network and the number of neurons in the input layer, hidden layer and output layer contained in the BP neural network are determined; drilling parameters and geological parameters are collected; the BP neural network is trained with the collected data to obtain the target BP neural network; during the training process, the network weights of the BP neural network are adjusted through iterative optimization so that the correlation between the jet displacement output by the BP neural network and the mechanical drilling speed is close to the actual relationship between the jet displacement and the mechanical drilling speed.

[0056] Specifically, during drilling operations, drilling fluid flow rate and ROP (Rate of Permeance) interact and are interdependent. Typically, the correlation between drilling fluid flow rate and ROP is positive and nonlinear, as various geological conditions and drilling parameters influence this relationship. Formation properties can include seabed hardness, undrained shear strength, porosity, and permeability. Drilling parameters include drilling fluid properties, drill bit type and size, and drilling process parameters. Drilling fluid properties include density, viscosity, and flow rate. Drilling process parameters include weight on bit, rotational speed, and injection angle.

[0057] In the design of deepwater surface jetting, a BP neural network-based model was developed to predict the relationship between drilling rate and ROP in jet drilling. Drilling and geological parameters were collected, and based on the relationship between drilling rate and ROP, the BP neural network architecture was designed, including the number of neurons in the input, hidden, and output layers. The design of these layers took into account the complexity of the input data and the nonlinear characteristics of the output relationship. The BP neural network was trained using the collected data.

[0058] Specifically, the number of neurons in the input layer of the BP neural network usually corresponds to the number of input features. The BP neural network mainly studies the influence of engineering operation parameters and geological parameters on the mechanical drilling speed during drilling. The drilling parameters are expressed as the difficulty of drilling into and destroying the formation, and the formation parameters are expressed as whether the formation is difficult to drill. In this embodiment, the input features may include formation parameters (such as rock hardness and porosity), drilling fluid parameters (such as density and viscosity), drill bit parameters (such as type and size), and other process parameters (such as bit pressure and displacement). Each input feature corresponds to an input neuron, which transmits data to the hidden layer.

[0059] There's no fixed rule for the number of neurons in a hidden layer, and this number is typically determined through experimentation. There can be one or more hidden layers, and the number of neurons in each hidden layer can vary. Hidden layers are responsible for extracting features and patterns from the input data, processing the data through nonlinear activation functions (such as Sigmoid, ReLU, or Tanh) to capture the complex relationship between input and output. In theory, more hidden layers and neurons can capture more complex patterns, but this can also lead to overfitting, so experimentation is necessary to determine the optimal number.

[0060] The output of the BP neural network is the mechanical penetration rate. Based on the input of different jet displacements, the BP neural network obtains the corresponding mechanical penetration rate, thereby determining the correlation between displacement and penetration rate. The drilling time during the jet drilling installation phase can be determined based on the penetration rate. The output layer converts the data processed by the hidden layer into the final prediction result. For regression problems (such as predicting mechanical penetration rate), the output layer typically uses a linear activation function. This method can design a BP neural network that can handle the complexity of the input data while capturing the nonlinear characteristics of the output relationship. This process requires a certain amount of experience and multiple experiments to achieve the optimal network structure.

[0061] Through iterative optimization, the network weights were adjusted to ensure that the network output closely matched the relationship between actual drilling displacement and mechanical penetration rate. The trained BP neural network was then tested using test data to verify the network's predictive performance. This helped assess the model's reliability and accuracy. Finally, during the network optimization phase, the network structure and training parameters were adjusted based on the test results to further improve the network's predictive accuracy and better adapt it to the needs of actual drilling operations.

[0062] Step S203: Using a target long short-term memory neural network, a surface bearing capacity recovery curve is established according to the drilling parameters and formation property parameters; the surface bearing capacity recovery curve is a curve showing the change of surface bearing capacity over time.

[0063] A target long short-term memory (LSTM) neural network can be used to establish a surface bearing capacity recovery curve based on the drilling parameters and formation property parameters. This curve shows the change in surface bearing capacity over time. The LSTM model can account for the impact of time series and accurately calculate the bearing capacity recovery during the static waiting process. It can also determine the shortest waiting time to avoid excessive waiting that wastes energy and emits excessive greenhouse gases, and excessive waiting that can lead to operational safety accidents.

[0064] The BP neural network can be used to calculate the drilling time through the mechanical drilling speed, and the LSTM neural network can be used to calculate the refined waiting time, thereby calculating the time of the entire jet drilling installation stage and the static waiting stage.

[0065] In some embodiments of the present specification, the target long short-term memory neural network is constructed by: determining the structure of the long short-term memory neural network and the number and layers of long short-term memory units; collecting jet drilling parameters and waiting time; using the collected jet drilling parameters and waiting time to train the long short-term memory neural network to obtain a target long short-term memory neural network; during the training process, adjusting the network weights of the long short-term memory neural network through iterative optimization so that the output of the long short-term memory neural network is close to the actual bearing capacity data.

[0066] To calculate the real-time bearing capacity during the jet drilling wait period, an LSTM method is used to establish a surface bearing capacity recovery curve with confidence intervals, thereby calculating the static waiting time during the jet drilling wait period. After collecting jet drilling parameters and waiting time, the LSTM network structure is designed, including the input layer, LSTM layer, and output layer. The number of LSTM units and layers in the LSTM layer needs to be adjusted according to the specific problem to accommodate the complexity of the data. The LSTM network is trained using engineering data, and the network weights are adjusted through iterative optimization to ensure that the network output is close to the actual bearing capacity data. Based on the LSTM network's prediction results, confidence intervals are established for the surface bearing capacity recovery curve over time to reflect the changing trend of bearing capacity.

[0067] Specifically, the number of LSTM units typically corresponds to the number of input features. LSTMs primarily study bearing capacity recovery during the waiting period, which is primarily influenced by geological parameters. Only the operational parameter during drilling, drilling rate, affects recovery during the waiting period, manifesting as excessive formation damage and slower recovery. The number of layers typically needs to be adjusted through experimentation to find the optimal balance between performance and computing resources. LSTM design must consider the temporal dependencies of the data. For example, the impact of drilling rate on soil recovery (backfill) may require consideration of information at different time scales. Determine the input feature dimensions and set the number of LSTM units. Determine the number of LSTM layers, each layer can have a different number of units. Determine the output layer structure, typically one or more neurons, to predict the impact of drilling rate on soil recovery. Train the LSTM model using time series data (drilling rate data) as input. Apply an appropriate loss function and optimize the model parameters through backpropagation and gradient descent. Use a validation set to evaluate model performance. Adjust the number of LSTM units and layers based on performance metrics such as prediction accuracy and overfitting.

[0068] In some embodiments of this specification, a target long-short-term memory neural network is used to establish a surface bearing capacity recovery curve based on the drilling parameters and formation property parameters, including: establishing the surface bearing capacity recovery curve based on the bearing capacity data output by the target long-short-term memory neural network; determining a static waiting time during the jet drilling waiting period based on the surface bearing capacity recovery curve, wherein the static waiting time is the minimum waiting time to ensure the safety of deepwater surface jet drilling operations. Determining the minimum waiting time during the jet drilling waiting period based on the bearing capacity recovery curve ensures that the bearing capacity of the casing is safe and that excessive waiting time is not wasted while positioning the work platform, thereby ensuring the safety and efficiency of deepwater surface well construction operations.

[0069] Step S204 , using a particle swarm optimization algorithm, based on the correlation between the injection displacement and the mechanical drilling speed and the surface bearing capacity recovery curve, calculate the target injection displacement.

[0070] After determining the correlation between jet displacement and ROP, as well as the surface bearing capacity recovery curve, a particle swarm optimization algorithm can be used to calculate the target jet displacement based on this correlation and the surface bearing capacity recovery curve. The particle swarm optimization (PSO) algorithm optimizes the coupled effects of displacement on the jet drilling installation and standby phases, calculating the optimal solution for both the installation and standby phases. The optimal solution includes the target jet displacement, target ROP, target drilling time, target standby time, and target surface conductor bearing capacity for the jet drilling installation phase.

[0071] In some embodiments of the present specification, a particle swarm optimization algorithm is used to calculate the target injection displacement based on the correlation between the injection displacement and the mechanical drilling speed and the surface bearing capacity recovery curve, including: initializing a randomly generated group of particles, and calculating the corresponding fitness based on the performance indicators of each group of particles; each particle is used to represent a jet drilling plan, and the jet drilling plan includes a combination of a corresponding injection displacement and a static waiting time; according to the historical best position and the global best position of each particle, its speed and position are updated, the new position of each particle is evaluated, and its fitness is calculated; if the fitness of the current particle is better than its historical best position, the individual best position is updated; if the fitness of the current particle is better than the global best position, the global best position is updated, and it is continuously iterated until a preset number of iterations is met or a fitness threshold is reached; the global best position is output, and the optimal solution for the jet drilling installation stage and the static waiting stage is obtained.

[0072] In this example, to improve the overall efficiency and safety of jet drilling, a particle swarm optimization (PSO) algorithm was used to optimize the flow rate combinations during the jet drilling installation and standby phases. Starting from a random solution, the PSO algorithm iteratively searches for the optimal solution, uses fitness to evaluate the quality of the solution, and then pursues the currently found optimal value to find the global optimum. The optimal solution for the entire jet drilling installation and standby phases includes the optimal solutions for the following parameters: jet flow rate and mechanical penetration rate during drilling, drilling time, standby time, and surface conductor load capacity.

[0073] Based on the goal of jet drilling, a fitness function is defined to evaluate the quality of each particle. This fitness function takes into account factors such as jet efficiency, safety, and environmental impact to ensure that the optimization results meet actual needs.

[0074] Based on each particle's historical best position and the global best position, its velocity and position are updated, each particle's new position is evaluated, and its fitness is calculated. If the current particle's fitness is better than its historical best position, the individual best position is updated; if the current particle's fitness is better than the global best position, the global best position is updated. This process iterates until the preset number of iterations or fitness threshold is met.

[0075] Finally, the global optimal position is output, which is the optimal solution for the entire jet drilling installation phase and the static waiting phase. This optimal solution can guide actual jet drilling operations and improve efficiency and safety.

[0076] In the above-described embodiment, during deepwater surface conductor well construction, a BP neural network is used to optimize the surface conductor jet flow rate to control the drilling speed, a long short-term memory neural network is used to predict the surface conductor bearing capacity recovery with a confidence interval, and a particle swarm optimization algorithm is used to optimize the flow rate to obtain the optimal deepwater surface conductor well construction flow rate. This embodiment is verified through actual drilling cases. The embodiment effectively balances efficient conductor installation operations and minimizes safe waiting time during deepwater surface well construction. This provides a new method for deepwater surface conductor well construction, which relies mainly on experience. The method has reference significance for deepwater surface well construction design and improves the safety and efficiency of high-efficiency deepwater surface drilling.

[0077] In some embodiments of the present specification, after calculating the target injection displacement based on the correlation between the injection displacement and the mechanical drilling rate and the surface bearing capacity recovery curve using a particle swarm optimization algorithm, the further embodiment includes: determining the corresponding target mechanical drilling rate based on the target injection displacement and the correlation between the injection displacement and the mechanical drilling rate; and calculating the real-time bearing capacity of the casing during the shortest well construction time and the waiting period based on the target injection displacement, the target mechanical drilling rate, and the surface bearing capacity recovery curve.

[0078] During the results analysis and decision support phase, the optimized jet flow rate and ROP are used to calculate the minimum well construction time and the real-time casing load capacity during the waiting period. These results are fed back into the actual production process to guide drilling operations, thereby improving efficiency and reducing costs.

[0079] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments. For details, please refer to the description of the aforementioned related processing embodiments, and no further description is given here.

[0080] The foregoing description of this specification describes specific embodiments. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in an order different from that described in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order shown or the sequential order to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0081] The above method is described below with reference to a specific embodiment. However, it should be noted that this specific embodiment is only for better illustrating this specification and does not constitute an improper limitation to this specification.

[0082] This specific embodiment provides a method for determining the jet flow rate during deepwater surface conductor well construction. Deepwater surface conductor well construction operations include the surface conductor installation phase and the post-installation waiting phase. The jet flow rate is a crucial factor in the surface conductor installation process. This jet flow rate not only determines the surface conductor's load-bearing capacity but is also directly related to the jetting speed. If the jet flow rate does not reach the minimum groundbreaking rate, the conductor's lowering speed will be very slow. Once the jet flow rate exceeds this minimum, increasing the flow rate can significantly improve the jetting efficiency. However, as the flow rate increases, the water jet's interference with the soil surrounding the conductor increases, further reducing the surface conductor's load-bearing capacity. Prolonged waiting times after installation can even lead to sinking accidents. Therefore, precisely controlling the jet flow rate during jet drilling requires considering both the hydraulic groundbreaking capacity and the conductor's load-bearing capacity, which is crucial for improving the efficiency and safety of surface well construction.

[0083] The shallow seabed soil is soft. To ensure jetting efficiency and prevent wellbore tilt and clay bit jamming, water jetting is primarily used for groundbreaking, supplemented by mechanical drilling. To ensure the efficiency of the jetting guide tube, the minimum flow rate required to achieve rock breaking capacity is set as the lower limit of the jetting flow rate during design. During well construction using the jetting method to install the surface guide tube, its impact on the soil can be viewed as the expansion of a cylindrical hole in infinite soil. The wellbore radius formed by the hydraulic jet corresponds to the initial radius of the cylindrical hole, while the radius of the surface guide tube is the radius after the hole is expanded.

[0084] The momentum of the water jet before it acts on the soil is ρQu, and after it acts on the soil, its momentum becomes ρQucosθ. According to the momentum theorem, the force of the water jet on the seabed soil surface is:

[0085] F=ρQu(1-cosθ) (1)

[0086] Where ρ is the density of the jet drilling fluid, in kg / m3 ; Q is the injection displacement, unit is m 3 / s; u is the jet velocity, unit is m / s; θ is the jet return angle, unit is °; F is the total impact force of the jet, unit is N.

[0087] If the water jet angle remains unchanged, the total impact force F of the jet also remains unchanged.

[0088] During the injection process, the momentum of the jet along the way is equal to the total momentum of the jet at the beginning.

[0089]

[0090] Where R0 is the nozzle radius, in meters; u0 is the initial jet velocity at the nozzle position, in meters per second; and J is the fluid momentum, in kg·m / s.

[0091] According to Bernoulli's theorem, the pressure at the nozzle outlet is:

[0092]

[0093] The force within the jet half-width is:

[0094]

[0095] Where p0 is the pressure at the nozzle outlet, unit is pa; F b is the jet force along the section, unit is N; λ is the jet force reduction coefficient.

[0096] The average force per unit area can be obtained from formula (4):

[0097]

[0098] Among them, p b is the force acting on the unit area of ​​soil, unit is Pa; S b is the nozzle effective area, unit m 2 ; α is the jet diffusion angle, unit is °; x is the jet impact range, unit is m.

[0099] During the spraying process, the water jet acts on the soil surface. When the jet force is greater than the critical failure pressure of the soil, the soil will fail:

[0100] p b ≥p cr (6)

[0101] Among them, p cr The critical failure pressure of soil can be approximately equal to the shear strength of the soil, unit: Pa;

[0102] Determine the diameter and extension of the drill bit according to the diameter of the conduit; determine the relative position of the drill bit nozzle and the bottom of the conduit, and the equivalent diameter d of the nozzle according to the structural characteristics of the jet drill bit. ne And the water jet attack range x.

[0103] Minimum injection displacement to meet ground-breaking conditions:

[0104]

[0105] Among them, Q0 is the minimum displacement value of hydraulic earth breaking; d ne is the nozzle equivalent diameter, unit is m; k f is the nozzle flow coefficient; Z is the number of nozzles. u It is the jetting impact force on soil, unit is Pa.

[0106] The relationship between the lowering speed of the catheter and the injection displacement is:

[0107]

[0108] Where v is the pipe installation speed when the displacement is Q, in m / s; Q is the actual injection displacement, in m 3 / min; Q0 is the minimum displacement value of hydraulic earth-breaking, unit is m 3 / min;v max is the fastest speed of the conduit, in m / s; a and b are influencing factors, which depend on the formation; R 2 is a regression factor that depends on the stratum.

[0109] The minimum displacement required to meet rock-breaking conditions depends on multiple factors, including the nozzle's equivalent diameter, drilling fluid density, nozzle structure, jet damage range, and soil shear strength. These parameters can be determined through the design of the drill bit and nozzle structure, as well as soil sampling and analysis.

[0110] While ensuring efficient jet drilling operations, it's important to consider the impact of excessive jet flow on the conductor's bearing capacity, preventing prolonged waiting times and possible sinking accidents. Models for the relationship between jet flow and surface conductor running speed, as well as jet flow and surface conductor bearing capacity, determine the upper limit for jet flow and the window for surface conductor rest time.

[0111] According to the soil data and the size of the surface conduit, the surface conduit ultimate bearing capacity calculation model (9) is used to establish the surface conduit ultimate bearing capacity profile.

[0112]

[0113] Considering the influence of injection disturbance and introducing the surface duct bearing capacity reduction coefficient, the real-time bearing capacity calculation model after the surface duct is injected in place can be expressed as:

[0114] F ut =KF u (10)

[0115] Among them, F ut is the real-time bearing capacity of the surface conduit, in N; F u is the ultimate bearing capacity of the surface conduit, in N; L is the depth of the surface conduit into the mud, in m; d o is the outer diameter of the surface conduit, in m; f u —Friction force per unit area of ​​the contact surface between the surface pipe and the soil, in Pa; A—Cross-sectional area of ​​the lower end of the surface pipe, in m 2 ;q u —Ultimate resistance per unit area at the lower end of the surface conduit, in N / m 2 . F ut is the real-time bearing capacity of the surface conduit, in N; K is the reduction coefficient of the injection bearing capacity of the surface conduit, dimensionless.

[0116] During the jetting process, the water jet disturbs the soil, resulting in a decrease in the surface conduit bearing capacity. Based on the jetting volume and the drill bit extension, the surface conduit real-time bearing capacity calculation model (9) is used to determine the surface conduit real-time bearing capacity and establish the surface conduit real-time bearing capacity profile.

[0117]

[0118] In the vertical downward direction, the conduit is subjected to its own weight as well as the weight of the anti-sinking plate, the low-pressure wellhead, and the lower joint of the running tool. The vertical load borne by the surface conduit when the running tool is released is as shown in formula (12):

[0119] G r =λk b L c +W LPWH +W m +W CADA (12)

[0120] Among them, Q0 is the minimum ejection soil breaking and rock carrying displacement, unit is m 3 / min; Q is the injection displacement, d o is the outer diameter of the surface catheter, f u — Friction force per unit area of ​​the contact surface between the surface pipe and the soil, q u —The ultimate resistance per unit area of ​​the lower end of the surface conduit, t2 is the static waiting time of the surface conduit, unit: h. iis the inner diameter of the surface conduit, in m; l is the amount of drill bit extended during the jet installation of the surface conduit, in m. S is the length from the nozzle to the end of the drill bit, in m; m and n are the area coefficients; λ is the jet force reduction coefficient, G r is the total weight borne by the catheter, in N; K b is the linear weight of the catheter, in N / m; L c is the length of the catheter, in m; W LPWH is the weight of the anti-sinking plate, unit is N; W m is the underwater wellhead weight, unit N; W CADA The CADA weight of the continuous drilling tool, in N.

[0121] The surface conductor rest time from the moment the injection is completed to the moment the delivery tool is released. Combining equations (11) and (12) yields the surface conductor rest time from the moment the injection is completed to the moment the delivery tool is released. Using parameters such as soil data, operating parameters, and surface conductor dimensions, simulations and calculations were performed to obtain real-time bearing capacity recovery curves for the conductor at different times. This safe rest time was determined and used to guide drilling operations in 82 wells on site, without any accidents.

[0122] In the process of jet drilling, the design of jet displacement is a complex process that requires comprehensive consideration of multiple factors, including the size of the pipe, the installation depth, the soil parameters and the drill nozzle parameters. In order to ensure the efficiency of the jet and the safety of the operation, the upper and lower limits of the jet displacement are determined based on the principles of fluid mechanics and soil mechanics, such as Figure 3 The upper limit of the jetting displacement is designed to avoid excessive impact on the bearing capacity of the casing, while the lower limit of the jetting displacement is designed to ensure jetting efficiency. The jetting displacement design during surface well construction ensures that the jet drilling process can effectively break up the soil while ensuring the safety of the casing.

[0123] During the jet installation phase of surface well construction, various factors, including drilling parameters and geological conditions, can affect the efficiency of pipe installation. The BP neural network has advantages because it can integrate the influence of multiple factors and is particularly advantageous for calculating the relationship between displacement and mechanical penetration rate. The BP neural network can be designed based on the above formulas (1)-(8).

[0124] The LSTM model can take into account the impact of time series, accurately calculate the recovery of carrying capacity during the waiting process, and determine the shortest waiting time to avoid excessive waiting that wastes energy and emits excessive greenhouse gases, and too short waiting that causes operational safety accidents. The LSTMBP neural network can be designed according to the above formulas (9)-(12).

[0125] The particle swarm optimization algorithm can optimize the coupling effect of displacement on the two stages of jet drilling installation and static waiting, and calculate the optimal solution of the entire jet drilling installation stage and static waiting stage.

[0126] In the design of deepwater surface jetting stage, a model based on BP neural network is established to predict the relationship between drilling displacement and mechanical penetration rate of jet drilling. The calculation principle and route are as follows: Figure 4 shown.

[0127] Drilling and geological parameters were collected, and based on the relationship between drilling rate and ROP, the BP neural network architecture was designed, including the number of neurons in the input, hidden, and output layers. The design of these layers took into account the complexity of the input data and the nonlinear characteristics of the output relationship.

[0128] The BP neural network was trained using the collected data. Through iterative optimization, the network weights were adjusted so that the network output closely matched the relationship between actual drilling rate and ROP. The trained BP neural network was then tested using test data to verify the network's predictive performance. This helped assess the model's reliability and accuracy.

[0129] Finally, in the network optimization stage, the network structure or training parameters were adjusted according to the test results to further improve the prediction accuracy of the network and better adapt to the needs of actual drilling operations.

[0130] When calculating the real-time bearing capacity during the jet drilling wait period, the LSTM method is used to establish a surface bearing capacity recovery curve that takes into account confidence intervals, thereby calculating the static waiting time during the jet drilling wait period. The confidence interval refers to the fact that due to fluctuations and operational errors in the operation, establishing a surface bearing capacity recovery curve within a certain error range can provide a safer static waiting time.

[0131] After collecting jet drilling parameters and waiting times, the LSTM network structure is designed, consisting of an input layer, an LSTM layer, and an output layer. The number of LSTM units and the number of layers within the LSTM layer need to be adjusted based on the specific problem to accommodate the complexity of the data. Because parameters vary across wells, inputs can be tailored to the operational characteristics of each well. The LSTM network is trained using engineering data, and through iterative optimization, the network weights are adjusted to ensure that the network output approximates the actual bearing capacity data. Engineering data can include jet drilling weight, displacement, drill bit parameters, rotational speed, engineering operational parameters, and geological parameters. Geological parameters can include soil shear strength, fracture pressure, loss pressure, bulk density, and friction parameters.

[0132] Based on the prediction results of the LSTM network, confidence intervals for the recovery curve of the surface bearing capacity over time are established to reflect the changing trend of bearing capacity. Based on the bearing capacity recovery curve, the minimum waiting time during jet drilling is determined to ensure the safe bearing capacity of the casing and avoid excessive waiting time that wastes the positioning of the work platform, thereby ensuring safe and efficient deepwater surface well construction operations. Specifically, based on the soil's bearing capacity recovery—that is, the maximum weight that the soil can bear—compared with the downward weight of equipment such as the casing's own weight, the underwater wellhead weight, and the CADA weight (known before the operation), the minimum safe operation time can be determined. This ensures operational safety while minimizing waiting time and energy consumption.

[0133] In order to improve the overall efficiency and safety of jet drilling, the embodiment of this specification uses a particle swarm optimization (PSO) algorithm to optimize the displacement combination of the jet drilling installation and static waiting stages. The particle swarm algorithm starts from a random solution, searches for the optimal solution through iteration, uses fitness to evaluate the quality of the solution, and follows the currently searched optimal value to find the global optimum. The mechanical drilling speed increases with the increase in displacement, but if the increase is too large, the waiting time will be too long. Therefore, by considering the combined effects of the drilling period and the waiting period, a suitable displacement is established. The displacement is the main factor that controls and affects the mechanical drilling speed. The calculation principle and route are as follows Figure 5 shown.

[0134] Based on the goal of jet drilling, a fitness function is defined to evaluate the quality of each particle. This fitness function takes into account factors such as jet efficiency, safety, and environmental impact to ensure that the optimization results meet actual needs.

[0135] In one embodiment, the fitness function is:

[0136] f(t)best=t1+t2;

[0137] Among them, f(t)best is the fitness function, which represents the global time, t1 is the time during drilling, and t2 is the waiting time during the waiting period. t1 can be calculated based on the casing running speed in formula (8). Specifically, Where L is the depth of the conduit and v is the speed at which the conduit is lowered. t2 can be calculated using formula (11).

[0138] Based on each particle's historical best position and the global best position, its velocity and position are updated, each particle's new position is evaluated, and its fitness is calculated. If the current particle's fitness is better than its historical best position, the individual best position is updated; if the current particle's fitness is better than the global best position, the global best position is updated. This process iterates until the preset number of iterations or fitness threshold is met.

[0139] Finally, the global optimal position is output, representing the optimal solution for the entire jet drilling installation and standby phases. This optimal solution can guide actual jet drilling operations, improving efficiency and safety. The optimal solution for the entire jet drilling installation and standby phases includes the optimal solutions for the following parameters: jet flow rate and ROP during drilling, drilling time, standby time, and surface casing load capacity.

[0140] In this specific embodiment, three different algorithms, BP neural network, LSTM and particle swarm optimization (PSO), are used to predict the relationship between the drilling displacement and mechanical penetration rate of jet drilling. The three algorithms have similar steps in the data collection and processing stages. They need to collect key parameter data, clean it, process missing values, and normalize or standardize the data to ensure the quality and consistency of the data. In order to provide effective solutions for different data types and analysis requirements, the specific details of the data collection and processing stages are different. The three methods assist and complement each other in prediction, among which the drilling jet displacement is only during drilling, but it will affect the waiting time. Therefore, the BP neural network predicts the drilling displacement and drilling rate, and the LSTM predicts the waiting time and bearing capacity. PSO predicts the global optimal displacement, thereby obtaining the jet displacement, drilling rate, drilling time, static waiting time, and bearing capacity.

[0141] The data collection phase of the BP neural network algorithm involves gathering data on key parameters such as nozzle diameter, drilling fluid density, jet damage range, and soil properties. During the data preprocessing phase, the collected data is cleaned, missing values ​​are addressed, and the data is normalized or standardized to ensure quality and consistency.

[0142] The data collection phase of the LSTM algorithm also involves collecting jet drilling parameter data, primarily time series data and input features that may affect bearing capacity, such as drilling rate and soil properties. This data is cleaned and missing values ​​are processed to prepare for the subsequent use of a proprietary soft-computing system for predicting the bearing capacity recovery of the conductor during the static period of deepwater surface well construction.

[0143] The particle swarm optimization (PSO) algorithm mainly relies on the above two algorithms and randomly generates a group of particles. Each particle represents a jet drilling scheme, which includes different combinations of jet displacement and static waiting time, providing an initial solution for the PSO algorithm.

[0144] This solution combines machine learning, deep learning, and intelligent optimization methods to provide a comprehensive data-driven solution for drilling operations. By analyzing and mining large amounts of data, the system can automatically identify key factors and provide real-time decision support. The calculation principle and route are as follows: Figure 6 shown.

[0145] During the data acquisition and processing stage, relevant data such as formation parameters and drilling parameters were collected, and pre-processing operations such as cleaning and normalization were performed on these data to improve the accuracy of subsequent analysis.

[0146] When using a BP neural network to predict jet drilling, the number of neurons in the input layer is designed based on the dimensionality of the input data, and the appropriate number of hidden layers and neurons per layer are selected. The number of neurons in the output layer is determined by the type of relationship to be output. The neural network is trained using the collected data, adjusting weights and bias values ​​to ensure that the model accurately reflects the complex, nonlinear relationship between well depth and drilling rate.

[0147] During the LSTM network's prediction phase, the temporal dependencies in the sequence data are captured. Using a model of the duct load-bearing capacity recovery mechanism, the LSTM network is trained using data and calibrated with actual engineering data, enabling it to predict the recovery of duct load-bearing capacity after injection.

[0148] During the PSO-based optimization phase, a set of randomly generated particles is initialized, and the fitness of each particle is calculated based on its performance metrics. Standard genetic algorithm operations, including selection, crossover, and mutation, are performed, and the particle population is iteratively updated to find the global optimal solution.

[0149] Finally, during the results analysis and decision support phase, the optimized jet flow rate and ROP are used to calculate the minimum well construction time and the real-time casing load capacity during the waiting period. These results are fed back into the actual production process to guide drilling operations, thereby improving efficiency and reducing costs.

[0150] Based on the same inventive concept, the embodiments of this specification also provide a device for determining the jet displacement during deep-water surface conduit well construction, as described in the following embodiments. Since the principle of the device for determining the jet displacement during deep-water surface conduit well construction is similar to that of the method for determining the jet displacement during deep-water surface conduit well construction, the implementation of the device for determining the jet displacement during deep-water surface conduit well construction can refer to the implementation of the method for determining the jet displacement during deep-water surface conduit well construction, and the repeated parts will not be repeated. As used below, the terms "unit" or "module" can refer to a combination of software and / or hardware that implements a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware, is also possible and contemplated. Figure 7 This is a structural block diagram of a device for determining the jet displacement during deepwater surface conduit well construction according to an embodiment of this specification. Figure 7As shown, it includes: an acquisition module 701, a determination module 702, an establishment module 703 and a calculation module 704. The structure is described below.

[0151] The acquisition module 701 is used to obtain drilling parameters and formation property parameters for deepwater surface jet well construction operations.

[0152] The determination module 702 is configured to determine the correlation between the jet displacement and the mechanical penetration rate based on the drilling parameters and the formation property parameters using the target BP neural network.

[0153] The establishment module 703 is used to use the target long short-term memory neural network to establish a surface bearing capacity recovery curve according to the drilling parameters and formation property parameters; the surface bearing capacity recovery curve is a curve showing the change of surface bearing capacity over time.

[0154] The calculation module 704 is used to calculate the target injection displacement based on the correlation between the injection displacement and the mechanical drilling speed and the surface bearing capacity recovery curve by using a particle swarm optimization algorithm.

[0155] In some embodiments of the present specification, the target BP neural network is constructed in the following manner: according to the physical relationship between the jet displacement and the mechanical drilling speed, the structure of the BP neural network and the number of neurons in the input layer, hidden layer and output layer contained in the BP neural network are determined; drilling parameters and geological parameters are collected; the BP neural network is trained with the collected data to obtain the target BP neural network; during the training process, the network weights of the BP neural network are adjusted through iterative optimization so that the correlation between the jet displacement output by the BP neural network and the mechanical drilling speed is close to the actual relationship between the jet displacement and the mechanical drilling speed.

[0156] In some embodiments of the present specification, the target long short-term memory neural network is constructed by: determining the structure of the long short-term memory neural network and the number and layers of long short-term memory units; collecting jet drilling parameters and waiting time; using the collected jet drilling parameters and waiting time to train the long short-term memory neural network to obtain a target long short-term memory neural network; during the training process, adjusting the network weights of the long short-term memory neural network through iterative optimization so that the output of the long short-term memory neural network is close to the actual bearing capacity data.

[0157] In some embodiments of the present specification, the establishment module is specifically used to: establish a surface bearing capacity recovery curve based on the bearing capacity data output by the target long short-term memory neural network; determine the static waiting time during the jet drilling waiting period based on the surface bearing capacity recovery curve, and the static waiting time refers to the static waiting time to ensure the safety of deep-water surface jet drilling operations.

[0158] In some embodiments of the present specification, the calculation module is specifically used to: initialize a randomly generated group of particles, and calculate the corresponding fitness based on the performance indicators of each group of particles; each particle is used to represent a jet drilling plan, and the jet drilling plan includes a corresponding combination of jet displacement and static waiting time; based on the historical best position and global best position of each particle, its speed and position are updated, the new position of each particle is evaluated, and its fitness is calculated; if the fitness of the current particle is better than its historical best position, the individual best position is updated; if the fitness of the current particle is better than the global best position, the global best position is updated, and it is continuously iterated until a preset number of iterations is met or a fitness threshold is reached; the global best position is output, and the optimal solution for the jet drilling installation phase and the static waiting phase is obtained.

[0159] In some embodiments of the present specification, the calculation module is further used to: after calculating the target injection displacement based on the correlation between the injection displacement and the mechanical drilling speed and the surface bearing capacity recovery curve using a particle swarm optimization algorithm, determine the corresponding target mechanical drilling speed according to the target injection displacement and the correlation between the injection displacement and the mechanical drilling speed; and calculate the real-time bearing capacity of the casing during the shortest well construction time and the waiting period based on the target injection displacement, the target mechanical drilling speed and the surface bearing capacity recovery curve.

[0160] This specification also provides a computer device. Figure 8 The diagram shows the structure of a computer device for determining the jetting rate during deepwater surface conduit well construction, according to an embodiment of this specification. Specifically, the computer device may include an input device 81, a processor 82, and a memory 83. The memory 83 is configured to store processor-executable instructions. When the processor 82 executes these instructions, the steps of the method for determining the jetting rate during deepwater surface conduit well construction, as described in any of the aforementioned embodiments, are implemented.

[0161] In this embodiment, the input device can specifically be one of the primary devices for exchanging information between a user and a computer system. The input device can include a keyboard, mouse, camera, scanner, light pen, handwriting input tablet, voice input device, etc.; the input device is used to input raw data and programs for processing these data into the computer. The input device can also receive data transmitted from other modules, units, and devices. The processor can be implemented in any appropriate manner. For example, the processor can take the form of a microprocessor or processor and a computer-readable medium storing computer-readable program code (such as software or firmware) executable by the (micro)processor, logic gates, switches, application-specific integrated circuits (ASICs), programmable logic controllers, and embedded microcontrollers, etc. The memory can specifically be a memory device used to store information in modern information technology. The memory can include multiple levels. In digital systems, anything that can store binary data can be considered a memory device. In integrated circuits, a circuit with storage functionality that does not have a physical form is also called a memory device, such as a RAM or FIFO. In systems, a physical storage device is also called a memory device, such as a memory stick or a TF card.

[0162] In this embodiment, the specific functions and effects achieved by the computer device can be explained in comparison with other embodiments and will not be repeated here.

[0163] In an embodiment of the present specification, a computer storage medium based on a method for determining the jet displacement during a deep-water surface conduit well construction process is also provided. The computer storage medium stores computer program instructions, which, when executed, implement the steps of the method for determining the jet displacement during a deep-water surface conduit well construction process described in any of the above embodiments.

[0164] In this embodiment, the storage medium includes, but is not limited to, random access memory (RAM), read-only memory (ROM), cache, hard disk drive (HDD), or memory card. The memory can be used to store computer program instructions. The network communication unit can be an interface configured in accordance with the standards specified by the communication protocol for network connection communication.

[0165] In this embodiment, the functions and effects specifically implemented by the program instructions stored in the computer storage medium can be explained in comparison with other embodiments and will not be repeated here.

[0166] Obviously, those skilled in the art should understand that the various modules or steps of the above-mentioned embodiments of this specification can be implemented using a general-purpose computing device, they can be concentrated on a single computing device, or distributed across a network composed of multiple computing devices. Alternatively, they can be implemented using program code executable by the computing device, so that they can be stored in a storage device and executed by the computing device. In some cases, the steps shown or described can be performed in a different order than herein, or they can be made into separate integrated circuit modules, or multiple modules or steps can be made into a single integrated circuit module for implementation. Thus, the embodiments of this specification are not limited to any specific combination of hardware and software.

[0167] It should be understood that the above description is intended to be illustrative and not limiting. Numerous embodiments and applications beyond the examples provided will be readily apparent to those skilled in the art upon reading the above description. Therefore, the scope of this specification should not be determined with reference to the above description, but rather with reference to the preceding claims, along with the full scope of equivalents to which such claims are entitled.

[0168] The above description is merely a preferred embodiment of this specification and is not intended to limit this specification. Those skilled in the art will readily appreciate that various modifications and variations to the embodiments of this specification are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of this specification shall be within the scope of protection of this specification.

Claims

1. A method for determining the jet displacement during deepwater surface conduit well construction, characterized in that: include: Obtain drilling parameters and formation property parameters for deepwater surface jet well construction operations; Determine the correlation between the jet displacement and the mechanical drilling rate based on the drilling parameters and the formation property parameters by using the target BP neural network; A target long short-term memory neural network is used to establish a surface bearing capacity recovery curve according to the drilling parameters and formation property parameters; the surface bearing capacity recovery curve is a curve of surface bearing capacity changing with time; The target injection displacement is calculated by using a particle swarm optimization algorithm based on the correlation between the injection displacement and the mechanical drilling speed and the surface bearing capacity recovery curve.

2. The method for determining the jet displacement in the deepwater surface conduit well construction process according to claim 1, characterized in that: The target BP neural network is constructed in the following way: According to the physical relationship between the jet displacement and the mechanical drilling speed, the structure of the BP neural network and the number of neurons in the input layer, the hidden layer and the output layer included in the BP neural network are determined; Collecting drilling parameters and geological parameters; using the collected data to train the BP neural network to obtain the target BP neural network; During the training process, the network weights of the BP neural network are adjusted through iterative optimization, so that the correlation between the jet displacement and the mechanical drilling speed output by the BP neural network is close to the actual relationship between the jet displacement and the mechanical drilling speed.

3. The method for determining the jet displacement in the deepwater surface conduit well construction process according to claim 1, characterized in that: The target long short-term memory neural network is constructed in the following way: Determine the structure of the LSTM neural network and the number of LSTM units and layers; Collecting jet drilling parameters and waiting time; using the collected jet drilling parameters and waiting time, training the long short-term memory neural network to obtain a target long short-term memory neural network; During the training process, the network weights of the long short-term memory neural network are adjusted through iterative optimization so that the output of the long short-term memory neural network is close to the actual carrying capacity data.

4. The method for determining the jet displacement in the deepwater surface conduit well construction process according to claim 1, characterized in that: Using a target long short-term memory neural network, a surface bearing capacity recovery curve is established according to the drilling parameters and formation property parameters; including: Establishing a surface bearing capacity recovery curve according to the bearing capacity data output by the target long short-term memory neural network; According to the surface bearing capacity recovery curve, the static waiting time during the jet drilling waiting period is determined, and the static waiting time refers to the shortest waiting time to ensure the safety of deep-water surface jet drilling operations.

5. The method for determining the jet displacement in the deepwater surface conduit well construction process according to claim 1, characterized in that: Using a particle swarm optimization algorithm, based on the correlation between the jetting displacement and the mechanical drilling speed and the surface bearing capacity recovery curve, the target jetting displacement is calculated, including: Initialize a group of randomly generated particles, and calculate the corresponding fitness according to the performance index of each group of particles; each particle is used to represent a jet drilling scheme, and the jet drilling scheme includes a combination of a corresponding jet displacement and a static waiting time; According to the historical best position and global best position of each particle, update its speed and position, evaluate the new position of each particle, and calculate its fitness; if the fitness of the current particle is better than its historical best position, update the individual best position; if the fitness of the current particle is better than the global best position, update the global best position, and continue to iterate until the preset number of iterations is met or the fitness threshold is reached; The global optimal position was output, and the optimal solution for the jet drilling installation phase and the static waiting phase was obtained.

6. The method for determining the jet displacement in the deepwater surface conduit well construction process according to claim 1, characterized in that: After calculating the target injection displacement by using a particle swarm optimization algorithm based on the correlation between the injection displacement and the mechanical drilling speed and the surface bearing capacity recovery curve, the method further includes: Determine a corresponding target mechanical drilling speed according to the target injection displacement and the correlation between the injection displacement and the mechanical drilling speed; Based on the target injection displacement, the target mechanical drilling speed and the surface bearing capacity recovery curve, the shortest well construction time and the real-time bearing capacity of the pipe during the waiting period are calculated.

7. A device for determining the jet displacement during deepwater surface conduit well construction, characterized in that: include: An acquisition module, used to acquire drilling parameters and formation property parameters for deepwater surface jet well construction operations; A determination module, for determining the correlation between the jet displacement and the mechanical drilling rate based on the drilling parameters and the formation property parameters by using a target BP neural network; Establishing a module for establishing a surface bearing capacity recovery curve according to the drilling parameters and formation property parameters by using a target long short-term memory neural network; the surface bearing capacity recovery curve is a curve of surface bearing capacity changing with time; The calculation module is used to calculate the target injection displacement based on the correlation between the injection displacement and the mechanical drilling speed and the surface bearing capacity recovery curve by using a particle swarm optimization algorithm.

8. A computer device, characterized in that: The method comprises a processor and a memory for storing processor-executable instructions, wherein the processor implements the steps of the method according to any one of claims 1 to 6 when executing the instructions.

9. A computer-readable storage medium having computer instructions stored thereon, characterized in that: When the instructions are executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images