Intelligent well drilling system for oil field

Abstract

The invention provides an oil field intelligent drilling system, and relates to the technical field of oil field drilling. The system comprises a data monitoring module, a trajectory prediction module, a control optimization module and an execution control module. Environment parameters and operation parameters are collected; a theoretical well track is generated, well drilling track prediction offset data is obtained, and a final prediction track is obtained; and a control parameter adjustment scheme is generated, and the bit pressure-rotating speed matching curve is adjusted. According to the invention, through cooperative operation of multiple modules, comprehensive monitoring and intelligent management of an oil field drilling process are realized. The efficiency is improved, meanwhile, the abrasion and failure rate of equipment are reduced, and the service life of the equipment is prolonged. By improving the drilling precision and efficiency, reducing accidents and reducing the equipment failure frequency, the drilling cost is optimized on the whole, the drilling period is effectively shortened, the single well yield is improved, and therefore remarkable benefits are brought in economy.

Description

Technical Field

[0001] The present invention relates to the technical field of oilfield drilling, and particularly to an intelligent oilfield drilling system. Background Art

[0002] Oilfield drilling technology has undergone years of development, gradually transitioning from initial mechanical drilling to the modern intelligent drilling stage. Current drilling methods mostly rely on numerical control technology and automated equipment, but there are still many limitations in actual applications. Generally speaking, traditional drilling processes include well design, bit selection, drilling implementation, wellbore stability, etc. In each link, operators often rely on experience and historical data for decision-making, resulting in insufficient utilization of information. In the oilfield drilling industry, with the gradual depletion of resources and the increasing difficulty of exploitation, traditional drilling methods face many challenges. These challenges are mainly reflected in aspects such as drilling accuracy, efficiency, safety, and real-time adaptability. Traditional drilling technologies usually rely on experience and manual judgment, leading to frequent deviations and steering errors in dynamic environments, thus causing a decrease in drilling efficiency and an increase in safety hazards.

[0003] Currently, many drilling systems rely on static models and qualitative analysis, resulting in a large deviation between the theoretical trajectory of the wellbore and the actual situation. Due to the lack of real-time monitoring and prediction capabilities, traditional drilling is prone to downhole deviation. This not only takes time and effort but may also cause equipment damage and safety accidents, increasing additional maintenance and downtime costs. Summary of the Invention

[0004] The present invention provides an intelligent oilfield drilling system to solve the defects existing in the prior art.

[0005] The present invention provides an intelligent oilfield drilling system, comprising: A data monitoring module for real-time collecting downhole environment parameters of the oilfield and equipment operation parameters for remotely controlling drilling.

[0006] A trajectory prediction module for constructing a physical model based on finite element analysis of drill string dynamics to generate a theoretical wellbore trajectory, constructing a prediction model based on a long short-term memory network to obtain drilling trajectory prediction offset data, and combining the theoretical wellbore trajectory and the prediction offset data to obtain a final predicted trajectory.

[0007] A control optimization module for generating a control parameter adjustment scheme according to the final predicted trajectory.

[0008] An execution control module for executing the control parameter adjustment scheme by using a PID control algorithm and adjusting the drill pressure - rotation speed matching curve by real-time correcting the steering tool face angle.

[0009] An intelligent oilfield drilling system provided by the present invention, the data monitoring module includes a multi-source sensing unit, a data transmission unit, and a data preprocessing unit. The multi-source sensing unit is used to collect downhole environment parameters in real time and synchronously obtain the equipment operation parameters of drilling. The downhole environment parameters include temperature, pressure, gamma value, and formation resistivity. The equipment operation parameters include weight on bit, rotary speed, torque, and mud displacement. The data transmission unit is used to be compatible with a variety of communication media to realize the data transmission between the downhole environment parameters and the equipment operation parameters. The data preprocessing unit is used to preprocess the downhole environment parameters and the equipment operation parameters to obtain preprocessed environment data and preprocessed operation data.

[0010] For an intelligent oilfield drilling system provided by the present invention, the process of preprocessing the downhole environment parameters and the equipment operation parameters includes: Perform filtering processing on the downhole environment parameters and the equipment operation parameters to suppress high-frequency noise.

[0011] Adopt a synchronization technology based on a drilling depth counter to perform spatio-temporal alignment on the downhole environment parameters and the equipment operation parameters to generate a spatio-temporally synchronized multi-dimensional data set.

[0012] Adopt wavelet transform and principal component analysis to perform feature extraction and dimensionality reduction processing on the multi-dimensional data set to obtain a feature vector matrix.

[0013] Perform standardization processing on each parameter column in the feature vector matrix by using Z-score standardization to obtain preprocessed environment data and preprocessed operation data.

[0014] For an intelligent oilfield drilling system provided by the present invention, the trajectory prediction module includes a reference trajectory generation unit, an offset prediction unit, and a risk assessment unit. The reference trajectory generation unit is used to construct a three-dimensional drill string dynamics model based on ABAQUS, input the preprocessed operation data, and output a theoretical wellbore trajectory. The offset prediction unit is used to construct a prediction model based on a long short-term memory network, input the preprocessed environment data and real-time trajectory data, and output the predicted offset data of the drilling trajectory. The predicted offset data includes a predicted offset probability and a predicted offset distance. The risk assessment unit is used to adopt a fusion algorithm to combine the theoretical wellbore trajectory and the predicted offset data to obtain a final predicted trajectory, and use Monte Carlo simulation to quantify the offset risk level of the final predicted trajectory.

[0015] For an intelligent oilfield drilling system provided by the present invention, the process of outputting the theoretical wellbore trajectory includes: Define geometric parameters, and the geometric parameters include drill string length, diameter, and material properties.

[0016] According to the geometric parameters, use ABAQUS to establish a physical model of the three-dimensional drill string and match geometric parameters for each part in the physical model.

[0017] Apply boundary conditions and loads to the physical model. The boundary conditions include the fixed conditions and contact conditions of the drill string, and the loads include the weight on bit and torque.

[0018] Select a non-linear solver in the physical model and set the solution parameters according to the time step and solution accuracy.

[0019] Start the solver to simulate the dynamic behavior of the drill string and obtain the theoretical wellbore trajectory.

[0020] According to an intelligent oilfield drilling system provided by the present invention, the process of constructing a prediction model based on a long short-term memory network includes: Collect historical data during the oilfield drilling process. The historical data includes historical environmental data, historical trajectory data, and historical offset data.

[0021] Normalize the historical data and set the time step to construct the normalized data into a time series format.

[0022] Construct a basic long short-term memory network model, define the long short-term memory network structure, including an input layer, an LSTM layer, and an output layer.

[0023] Use the environmental data at the current moment and the trajectory data at the current moment in the historical data as inputs, and the offset data at the next moment as the output to train the basic model. Retain the model parameters that meet the preset accuracy to obtain the prediction model.

[0024] According to an intelligent oilfield drilling system provided by the present invention, the control optimization module includes an offset data analysis unit and a control strategy generation unit. The offset data analysis unit is used to analyze the final predicted trajectory by using statistical analysis methods to obtain the offset pattern and offset trend of the drilling trajectory. The control strategy generation unit is used to generate a control parameter adjustment plan according to the offset pattern and offset trend, and optimize the control parameter adjustment plan by using the particle swarm algorithm.

[0025] According to an intelligent oilfield drilling system provided by the present invention, the process of analyzing the final predicted trajectory by using statistical analysis methods includes: Collect historical trajectories, which include historical reference trajectories and historical offsets.

[0026] Process the historical trajectories by using data smoothing techniques to remove invalid values and noise.

[0027] Statistically analyze the overall characteristics of the historical offsets. The overall characteristics include the mean, standard deviation, maximum value, and minimum value.

[0028] Draw a frequency distribution diagram of the historical offsets according to the overall characteristics.

[0029] The historical offsets are classified using hierarchical clustering to obtain offset patterns, which include normal offsets, abnormal offsets, and periodic offsets.

[0030] Based on the frequency distribution diagram, an offset trend model based on ARIMA is constructed, and the trend of the offset changing over time is obtained through time series analysis of the offset.

[0031] According to an intelligent oilfield drilling system provided by the present invention, the process of optimizing the control parameter adjustment scheme using the particle swarm algorithm includes: Define that each particle represents a possible combination of control parameters, and the control parameters include weight on bit, rotary speed, and angle.

[0032] Randomly generate the positions and velocities of a group of particles. The position represents the combination of control parameters, and the velocity represents the adjustment direction of the control parameter combination.

[0033] For each particle, perform fitness evaluation according to the objective function, and the objective function represents the cost function of minimizing the offset.

[0034] Update the velocity and position of each particle within the preset range of control parameters until the preset fitness value is met.

[0035] Output the positions and velocities of the particles that meet the preset fitness value, and use the corresponding control parameter combination and adjustment direction as the control parameter adjustment scheme.

[0036] According to an intelligent oilfield drilling system provided by the present invention, the execution control module includes a scheme execution unit, a feedback detection unit, and a decision-making adjustment unit. The scheme execution unit is used to execute the control parameter adjustment scheme using the PID control algorithm and adjust the device operation parameters in real time. The feedback monitoring unit is used to collect the adjusted device operation data in real time. The decision-making adjustment unit is used to dynamically adjust the control parameter adjustment scheme according to the device operation data and the real-time downhole environment data.

[0037] An intelligent oilfield drilling system provided by the present invention realizes full - range monitoring of the drilling process by collecting downhole environmental parameters and equipment operation status in real time. It can not only obtain key data immediately, but also identify potential risks in a timely manner, such as sudden changes in downhole pressure, abnormal temperature, etc., thus greatly reducing the risk of accidents. Combining finite - element analysis of drill - string dynamics and long - short - term memory network models can accurately predict the deviation of the drilling trajectory. This enables operators to make responses in advance, ensuring that the drilling trajectory is always within a safe and predetermined range, thereby reducing drilling accidents caused by deviation and improving overall safety. By combining the theoretical wellbore trajectory with real - time predicted deviation data, the final predicted trajectory can be generated. A more accurate control - parameter adjustment scheme can be generated. The scientific and data - based optimization method can significantly increase the drilling speed and reduce the drilling cost. Through real - time adjustment of intelligent algorithms, the system can quickly and effectively respond to different situations, ensuring the stability and smoothness of the drilling process. It not only improves efficiency, but also reduces equipment wear and failure rates, and extends the service life of equipment. By improving drilling accuracy and efficiency, reducing accident occurrences and equipment failure frequencies, the overall drilling cost is optimized, the drilling cycle is effectively shortened, the single - well production is increased, thus bringing significant economic benefits. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0039] Figure 1 It is a schematic structural diagram of an intelligent oilfield drilling system provided by an embodiment of the present invention; Figure 2 It is a flow chart for obtaining the theoretical wellbore trajectory in an embodiment of the present invention; Figure 3 It is a flow chart for statistical analysis of the final predicted trajectory in an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0040] To make the objectives, technical solutions, and advantages of the present invention clearer, the following will clearly and completely describe the technical solutions in the present invention with reference to the drawings in the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Based on the embodiments in the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present invention.

[0041] The following will be combined with Figures 1 - 3Describe an intelligent oilfield drilling system of the present invention.

[0042] Figure 1 It is a schematic structural diagram of an intelligent oilfield drilling system provided by an embodiment of the present invention.

[0043] As Figure 1 shown, an intelligent oilfield drilling system provided by an embodiment of the present invention includes a data monitoring module, a trajectory prediction module, a control optimization module, and an execution control module.

[0044] The data monitoring module is used to collect in real time the downhole environment parameters of the oilfield and the equipment operation parameters for remotely controlling the drilling.

[0045] The data monitoring module includes a multi-source sensing unit, a data transmission unit, and a data preprocessing unit. The multi-source sensing unit is used to collect in real time the downhole environment parameters and synchronously obtain the equipment operation parameters of the drilling. The downhole environment parameters include temperature, pressure, gamma value, and formation resistivity. The equipment operation parameters include weight on bit, rotary speed, torque, and mud flow rate. The data transmission unit is used to be compatible with a variety of communication media to realize the data transmission between the downhole environment parameters and the equipment operation parameters. The data preprocessing unit is used to preprocess the downhole environment parameters and the equipment operation parameters to obtain preprocessed environment data and preprocessed operation data.

[0046] Temperature sensors are installed at various key positions of the drill string to be able to monitor the downhole temperature changes in real time. A high-pressure-resistant voltage sensor is selected to ensure accurate measurement of the downhole pressure in a high-pressure environment. A gamma-ray detector is used to monitor the radioactive components of the formation to provide support for the formation characteristic analysis. An adaptive resistivity sensor is used to measure the resistivity of the formation through the flow of current to evaluate the fluid saturation.

[0047] A strain gauge sensor is adopted and connected between the drill bit and the drill string to report in real time the applied weight on bit. Through a rotary encoder, the rotary speed of the drilling rig is monitored in real time to ensure consistency with the current drilling plan. A DC torque sensor is applied to record the torque feedback of the drill string to evaluate the energy efficiency and wear during the drilling process. A flowmeter is used to measure the flow rate of the mud to ensure stable mud circulation and prevent blowout.

[0048] All sensors convert analog signals into digital signals through an A / D converter, and an industrial controller is used to synchronously collect various parameters to realize the integration of time-sensitive data.

[0049] The process of preprocessing the downhole environment parameters and the equipment operation parameters includes: An algorithm is adopted to filter the noise of the collected data, identify and remove the measured values that exceed the reasonable range or have anomalies, such as data fluctuations caused by sensor failures or environmental interferences.

[0050] Calibrate the outputs of various sensors to ensure that the data signals meet engineering standards and convert them into corresponding engineering units.

[0051] Adopt a synchronization technology based on the drilling depth counter to align the downhole environment parameters and equipment operation parameters in space and time, and generate a multi-dimensional dataset synchronized in space and time.

[0052] Use wavelet transform and principal component analysis to extract features and reduce the dimension of the multi-dimensional dataset to obtain a feature vector matrix.

[0053] Use Z-score standardization to standardize each parameter column in the feature vector matrix to obtain preprocessed environment data and preprocessed operation data.

[0054] The trajectory prediction module is used to construct a physical model based on the finite element analysis of drill string dynamics to generate a theoretical wellbore trajectory, construct a prediction model based on the long short-term memory network to obtain the predicted offset data of the drilling trajectory, and combine the theoretical wellbore trajectory and the predicted offset data to obtain the final predicted trajectory.

[0055] The trajectory prediction module includes a reference trajectory generation unit, an offset prediction unit, and a risk assessment unit. The reference trajectory generation unit is used to construct a three-dimensional drill string dynamics model based on ABAQUS, input the preprocessed operation data, and output the theoretical wellbore trajectory. The offset prediction unit is used to construct a prediction model based on the long short-term memory network, input the preprocessed environment data and real-time trajectory data, and output the predicted offset data of the drilling trajectory. The predicted offset data includes the predicted offset probability and the predicted offset distance. The risk assessment unit is used to combine the theoretical wellbore trajectory and the predicted offset data using a fusion algorithm to obtain the final predicted trajectory, and use Monte Carlo simulation to quantify the offset risk level of the final predicted trajectory.

[0056] Figure 2 It is a flowchart for obtaining the theoretical wellbore trajectory in an embodiment of the present invention.

[0057] As Figure 2 shown, the process of outputting the theoretical wellbore trajectory includes: Define geometric parameters, which include drill string length, diameter, and material properties.

[0058] According to the geometric parameters, use ABAQUS to establish a physical model of the three-dimensional drill string and match geometric parameters to each part of the physical model. Use ABAQUS software to perform geometric modeling of the three-dimensional model, create the geometric shape of the drill string, including structures such as the drill bit, drill string, and wellbore wall. Input the material properties of the drill string and wellbore wall, including density, elastic modulus, Poisson's ratio, etc., so as to ensure that the model can truly reflect the mechanical properties of the materials.

[0059] Apply boundary conditions and loads to the physical model. The boundary conditions include the fixed conditions and contact conditions of the drill string, and the loads include the weight on bit and torque.

[0060] Select a non-linear solver in the physical model and set the solution parameters according to the time step and solution accuracy.

[0061] Start the solver to simulate the dynamic behavior of the drill string and obtain the theoretical wellbore trajectory. Run the ABAQUS finite element simulation, using the explicit dynamic analysis or implicit static analysis method, to calculate the motion characteristics and deformation of the drill string during drilling. Obtain the deformation and stress field distribution of the drill string through numerical simulation, and then calculate the corresponding theoretical wellbore trajectory. After the model operation, output the coordinate data of the theoretical wellbore trajectory, including information such as the wellbore position and inclination angle.

[0062] The process of constructing a prediction model based on the long short-term memory network includes: Collect historical data of the oilfield drilling process. The historical data includes historical environmental data, historical trajectory data, and historical offset data.

[0063] Normalize the historical data and set the time step to construct the normalized data into a time series format.

[0064] Construct the basic long short-term memory network model, define the long short-term memory network structure, including the input layer, LSTM layer, and output layer. The input layer sets the input nodes according to the number of features of the collected data. The LSTM layer adds multiple LSTM units to increase the learning ability of the model and process the time series information. The output layer is set as a fully connected layer to output the predicted offset probability and predicted offset distance.

[0065] Use the environmental data at the current moment and the trajectory data at the current moment in the historical data as the input, and the offset data at the next moment as the output to train the basic model. Retain the model parameters that meet the preset accuracy to obtain the prediction model.

[0066] Train the model through historical data, use the mean squared error as the loss function, and apply the Adam optimization algorithm to adjust the model weights. Evaluate the model performance and adjust the training process through the early stopping method to avoid overfitting.

[0067] Process the input new data with the trained prediction model, output the predicted offset probability and predicted offset distance, and quantify the potential offset of the drilling trajectory.

[0068] The risk assessment unit uses a fusion algorithm to combine the theoretical wellbore trajectory and the predicted offset data to obtain the final predicted trajectory and quantify its offset risk level.

[0069] The theoretical wellbore trajectory is combined with the offset prediction data through fusion algorithms such as the weighted average method or the method based on Bayesian inference to generate the final predicted trajectory. By setting weights, the relative importance of the theoretical trajectory and the prediction result is adjusted to better adapt to the current drilling situation.

[0070] The Monte Carlo simulation method is used to conduct a risk assessment on the final predicted trajectory. By randomly sampling the offset probability, the possible offset situations are repeatedly simulated; According to the simulation results, the probability distribution of deviation from the predetermined trajectory is statistically obtained to evaluate the risk level. The risk level can be quantified into different levels such as low, medium, and high, and a risk report is generated.

[0071] The control optimization module is used to generate a control parameter adjustment scheme according to the final predicted trajectory.

[0072] The control optimization module includes an offset data analysis unit and a control strategy generation unit. The offset data analysis unit is used to analyze the final predicted trajectory by using statistical analysis methods to obtain the offset pattern and offset trend of the drilling trajectory. The control strategy generation unit is used to generate a control parameter adjustment scheme according to the offset pattern and offset trend, and optimize the control parameter adjustment scheme by using the particle swarm algorithm.

[0073] Figure 3 It is the flowchart of the statistical analysis of the final predicted trajectory in the embodiment of the present invention.

[0074] As Figure 3 shown, the process of analyzing the final predicted trajectory by using statistical analysis methods includes: Collect historical trajectories, where the historical trajectories include historical reference trajectories and historical offsets.

[0075] The data smoothing technology is used to process the historical trajectories to remove invalid values and noise.

[0076] Statistically analyze the overall characteristics of the historical offsets, where the overall characteristics include the mean, standard deviation, maximum value, and minimum value.

[0077] Draw a frequency distribution diagram of the historical offsets according to the overall characteristics.

[0078] Use the hierarchical clustering method to classify the historical offsets to obtain the offset patterns, where the offset patterns include normal offsets, abnormal offsets, and periodic offsets.

[0079] According to the frequency distribution diagram, construct an offset trend model based on ARIMA, and analyze the trend of the offset changing with time through time series to obtain the offset trend.

[0080] Determine the control parameters to be adjusted, such as weight on bit, rotary speed, torque, and mud displacement, according to the offset pattern and offset trend. Set preliminary control parameter adjustment rules. For example, if the prediction model shows that too high weight on bit may cause offset, then the weight on bit needs to be reduced.

[0081] Based on the analysis results, formulate a control parameter adjustment plan. Each adjustment should be quantitatively analyzed according to the influence of different offset patterns, such as reflecting the offset trend observed over a certain period of time into the adjustment strategy. If an increase in weight on bit causes offset, then reduce the weight on bit. If too low rotary speed is related to frequent offset, then moderately increase the rotary speed.

[0082] The process of optimizing the control parameter adjustment plan using the particle swarm algorithm includes: Define that each particle represents a possible combination of control parameters, and the control parameters include weight on bit, rotary speed, and angle.

[0083] Randomly generate the positions and velocities of a group of particles. The position represents the combination of control parameters, and the velocity represents the adjustment direction of the combination of control parameters.

[0084] For each particle, conduct fitness evaluation according to the objective function, and the objective function represents the cost function for minimizing the offset.

[0085] Update the velocity and position of each particle within the preset range of control parameters until the preset fitness value is satisfied.

[0086] Output the positions and velocities of the particles that satisfy the preset fitness value, and use the corresponding combination of control parameters and adjustment direction as the control parameter adjustment plan.

[0087] The execution control module is used to execute the control parameter adjustment plan using the PID control algorithm, and adjust the weight on bit - rotary speed matching curve by correcting the orientation tool face angle in real time.

[0088] The execution control module includes a plan execution unit, a feedback detection unit, and a decision - making adjustment unit. The plan execution unit is used to execute the control parameter adjustment plan using the PID control algorithm and adjust the equipment operation parameters in real time. The feedback monitoring unit is used to collect the adjusted equipment operation data in real time. The decision - making adjustment unit is used to dynamically adjust the control parameter adjustment plan according to the equipment operation data and real - time downhole environment data.

[0089] The PID control algorithm consists of three parts: proportional, integral, and derivative, which are used for the immediate response of control parameters, the correction of the system's past errors, and the prediction of future change trends, respectively.

[0090] According to the control parameter adjustment plan provided by the control strategy generation unit, set parameters such as the target weight on bit, rotary speed, and torque.

[0091] During each control cycle, the operation data of equipment such as the weight on bit and rotational speed are collected in real time, and the error between the current output and the set value is calculated. Based on the error value, the PID control signal is calculated.

[0092] The drive control signal of the equipment is adjusted in real time to regulate the weight on bit and rotational speed, ensuring that they can approach the set value in real time. The adjusted control signal is sent to the drilling equipment through the actuator or control interface, and the parameter adjustment is executed.

[0093] Through the multi-source sensing unit, the key operation parameters of the equipment during the drilling process are monitored and captured in real time, such as the current weight on bit, rotational speed, torque, mud displacement, temperature, etc.

[0094] The collected feedback data is sent to the monitoring system in real time through the data transmission unit.

[0095] A real-time monitoring control panel is established to centrally display and record the operation data of each equipment, facilitating the operators to observe, monitor and make real-time analysis.

[0096] The real-time feedback data is stored in the database for subsequent inspection, analysis and data mining. By setting timestamps, it is ensured that all data is traceable.

[0097] Machine learning algorithms are used to evaluate the equipment condition, monitor the feedback data, and conduct real-time analysis in combination with environmental parameters such as temperature and pressure.

[0098] When the monitored equipment parameters (such as the weight on bit and rotational speed) continuously deviate from the design target, the response ability of the evaluation system is evaluated. For example, if the weight on bit is higher than the target value for a long time, the control strategy needs to be re-evaluated and adjusted.

[0099] If the error between the operation data collected in the feedback monitoring unit and the target value continuously exceeds the preset threshold, the decision-making adjustment unit will call the previously set control parameter adjustment plan and make corrections. For example, increase the Kp of the PID control to enhance the system's response ability to the error. The new control parameters are sent to the plan execution unit to adjust the operation parameters of the drilling equipment, forming a closed-loop control process.

[0100] In summary, this embodiment provides an intelligent oilfield drilling system that realizes comprehensive monitoring of the drilling process by collecting downhole environmental parameters and equipment operating status in real time. It can not only obtain key data immediately but also identify potential risks in a timely manner, such as sudden changes in downhole pressure and abnormal temperatures, thus greatly reducing the risk of accidents. Combining finite element analysis of drill string dynamics and long short-term memory network models can accurately predict the deviation of the drilling trajectory. This enables operators to respond in advance, ensuring that the drilling trajectory always remains within a safe and predetermined range, thereby reducing drilling accidents caused by deviations and improving overall safety. By combining the theoretical wellbore trajectory with real-time predicted deviation data, the final predicted trajectory can be generated. A more accurate control parameter adjustment scheme can be generated. The scientific and data-based optimization method can significantly increase the drilling speed and reduce the drilling cost. Through real-time adjustment of intelligent algorithms, the system can quickly and effectively respond to different situations, ensuring the stability and smoothness of the drilling process. This not only improves efficiency but also reduces equipment wear and failure rates, extending the service life of the equipment. By improving the accuracy and efficiency of drilling, reducing accident occurrences and equipment failure frequencies, the drilling cost is optimized as a whole, the drilling cycle is effectively shortened, and the single-well output is increased, thus bringing significant economic benefits.

[0101] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative effort.

[0102] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence, or the part that contributes to the prior art can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0103] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments or perform equivalent replacements for some of the technical features. However, such modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. An intelligent oilfield drilling system, characterized in that: include: Data monitoring module, used to collect real-time downhole environmental parameters of the oil field and equipment operating parameters for remote control drilling; A trajectory prediction module is used to construct a physical model based on finite element analysis of drill string dynamics to generate a theoretical wellbore trajectory, construct a prediction model based on a long short-term memory network to obtain drilling trajectory prediction offset data, and combine the theoretical wellbore trajectory and the predicted offset data to obtain a final predicted trajectory; A control optimization module, used for generating a control parameter adjustment scheme according to the final predicted trajectory; The execution control module is used to execute the control parameter adjustment scheme by adopting a PID control algorithm, and adjust the drilling pressure-rotation speed matching curve by correcting the steering tool face angle in real time.

2. The intelligent oilfield drilling system according to claim 1, characterized in that: The data monitoring module includes a multi-source sensing unit, a data transmission unit and a data preprocessing unit; the multi-source sensing unit is used to collect downhole environmental parameters in real time and synchronously obtain drilling equipment operating parameters, the downhole environmental parameters include temperature, pressure, gamma value and formation resistivity, and the equipment operating parameters include drilling pressure, rotation speed, torque and mud displacement; the data transmission unit is used to be compatible with multiple communication media to realize data transmission between the downhole environmental parameters and the equipment operating parameters; the data preprocessing unit is used to preprocess the downhole environmental parameters and the equipment operating parameters to obtain preprocessed environmental data and preprocessed operating data.

3. The intelligent oilfield drilling system according to claim 2, characterized in that: The process of preprocessing the downhole environmental parameters and the equipment operating parameters includes: Filtering the downhole environmental parameters and the equipment operating parameters to suppress high-frequency noise; Using a synchronization technology based on a drilling depth counter to perform time-space alignment on the downhole environmental parameters and the equipment operation parameters, and generate a time-space synchronized multidimensional data set; Using wavelet transform and principal component analysis to perform feature extraction and dimensionality reduction processing on the multidimensional data set to obtain a feature vector matrix; Each parameter column in the characteristic vector matrix is ​​standardized by using Z-score standardization to obtain preprocessing environment data and preprocessing operation data.

4. The intelligent oilfield drilling system according to claim 3, characterized in that: The trajectory prediction module includes a reference trajectory generation unit, an offset prediction unit and a risk assessment unit; the reference trajectory generation unit is used to construct a three-dimensional drill string dynamics model based on ABAQUS, input the preprocessed operation data, and output a theoretical wellbore trajectory; the offset prediction unit is used to construct a prediction model based on a long short-term memory network, input the preprocessed environment data and real-time trajectory data, and output predicted offset data of the drilling trajectory, wherein the predicted offset data includes a predicted offset probability and a predicted offset distance; the risk assessment unit is used to use a fusion algorithm to combine the theoretical wellbore trajectory and the predicted offset data to obtain a final predicted trajectory, and use Monte Carlo simulation to quantify the offset risk level of the final predicted trajectory.

5. The intelligent oilfield drilling system according to claim 1, characterized in that: The process of outputting theoretical wellbore trajectory includes: defining geometric parameters, including drill string length, diameter, and material properties; According to the geometric parameters, a physical model of the three-dimensional drill string is established using ABAQUS, and geometric parameters are matched for each part in the physical model; Applying boundary conditions and loads to the physical model, wherein the boundary conditions include a fixed condition and a contact condition of the drill string, and the loads include a bit pressure and a torque; Selecting a nonlinear solver in the physical model and setting solution parameters according to a time step and solution accuracy; Start the solver to simulate the dynamic behavior of the drill string and obtain the theoretical wellbore trajectory.

6. The intelligent oilfield drilling system according to claim 1, characterized in that: The process of building a prediction model based on LSTM networks includes: Collecting historical data of the oilfield drilling process, wherein the historical data includes historical environmental data, historical trajectory data and historical offset data; Normalizing the historical data, and setting a time step to construct the normalized data into a time series format; Build the basic model of long short-term memory network and define the structure of long short-term memory network, including input layer, LSTM layer and output layer; The current moment environmental data and the current moment trajectory data in the historical data are used as input, and the next moment offset data is used as output, the basic model is trained, and the model parameters that meet the preset accuracy are retained to obtain a prediction model.

7. The intelligent oilfield drilling system according to claim 1, characterized in that: The control optimization module includes an offset data analysis unit and a control strategy generation unit; the offset data analysis unit is used to analyze the final predicted trajectory using a statistical analysis method to obtain the offset mode and offset trend of the drilling trajectory; the control strategy generation unit is used to generate a control parameter adjustment plan according to the offset mode and the offset trend, and use a particle swarm algorithm to optimize the control parameter adjustment plan.

8. The intelligent oilfield drilling system according to claim 1, characterized in that: The process of analyzing the final predicted trajectory using a statistical analysis method includes: Collecting historical trajectories, wherein the historical trajectories include historical reference trajectories and historical offsets; The historical trajectory is processed using data smoothing technology to remove invalid values ​​and noise; Counting overall characteristics of the historical offsets, where the overall characteristics include an average value, a standard deviation, a maximum value, and a minimum value; Draw a frequency distribution diagram of the historical offsets according to the overall characteristics; Using a hierarchical clustering method to classify the historical offsets to obtain an offset pattern, wherein the offset pattern includes a normal offset, an abnormal offset, and a periodic offset; According to the frequency distribution diagram, an ARIMA-based offset trend model is constructed, and the offset trend is obtained by analyzing the trend of the offset changing over time through time series analysis.

9. The intelligent oilfield drilling system according to claim 1, characterized in that: The process of optimizing the control parameter adjustment scheme using the particle swarm algorithm includes: It is defined that each particle represents a possible control parameter combination, wherein the control parameters include drilling pressure, rotation speed and angle; Randomly generate a group of particle positions and velocities, where the positions represent the control parameter combination, and the velocities represent the adjustment direction of the control parameter combination; For each particle, a fitness evaluation is performed according to an objective function, where the objective function represents a cost function that minimizes the offset; Update the speed and position of each particle within the preset control parameter range until the preset fitness value is met; The position and speed of the particle that meets the preset fitness value are output, and the corresponding control parameter combination and adjustment direction are used as the control parameter adjustment scheme.

10. The intelligent oilfield drilling system according to claim 1, characterized in that: The execution control module includes a scheme execution unit, a feedback detection unit and a decision adjustment unit; the scheme execution unit is used to adopt a PID control algorithm to execute the control parameter adjustment scheme and adjust the equipment operating parameters in real time; the feedback monitoring unit is used to collect the adjusted equipment operating data in real time; the decision adjustment unit is used to dynamically adjust the control parameter adjustment scheme according to the equipment operating data and real-time downhole environmental data.

Images