An automatic control system and method for oil and gas field mining and production based on artificial intelligence

Abstract

The present application discloses an automatic control system and method for oil and gas field exploitation and production based on artificial intelligence, which relates to the technical field of oil and gas field exploitation and production. The disclosed automatic control system for oil and gas field exploitation and production based on artificial intelligence includes a data acquisition module, an equipment control module, and a cloud-edge collaboration module. The data acquisition module acquires initial data in real time, and generates high-quality data input after preprocessing by the edge computing node. The equipment control module generates preliminary equipment control instructions based on these data, and the cloud-edge collaboration module further utilizes the digital twin model and the cloud simulation platform to simulate the execution effect of the equipment control instructions and optimize them, thereby forming a multi-level and multi-dimensional control strategy, which not only improves the operating efficiency and stability of the equipment, but also significantly reduces energy consumption and equipment loss, thereby improving the equipment control effect in the oil and gas field exploitation and production process.

Description

Technical Field

[0001] The present application relates to the technical field of oil and gas field exploitation and production, and in particular to an automatic control system and method for oil and gas field exploitation and production based on artificial intelligence. Background Art

[0002] Existing technologies for controlling oil and gas field equipment typically rely on manual experience or parameter adjustments through fixed thresholds or simple feedback mechanisms. These typically employ PID control algorithms, rule bases based on expert experience, or static data models. These methods are somewhat effective when equipment is operating in a stable state. However, the complex and ever-changing oil and gas field production environment, coupled with the real-time and heterogeneous nature of multi-source data (such as pressure, temperature, and flow), makes it difficult for traditional control systems to achieve effective control results. Summary of the Invention

[0003] The main purpose of this application is to provide an automatic control system and method for oil and gas field exploitation and production processes based on artificial intelligence, aiming to solve the problem of poor equipment control effect in oil and gas field exploitation and production processes in the existing technology.

[0004] To achieve the above objectives, this application proposes an automatic control system for oil and gas field exploitation and production based on artificial intelligence, the system comprising:

[0005] A data acquisition module, comprising a multi-source sensor array and an edge computing node, for collecting initial data from oil and gas field exploitation and production in real time based on the multi-source sensor array, and processing the initial data based on the edge computing node to generate pre-processed data;

[0006] An equipment control module, connected to the data acquisition module, is used to obtain the pre-processed data and generate corresponding equipment control instructions to adjust the working status and production parameters of corresponding equipment during oil and gas field exploitation and production;

[0007] The cloud-edge collaboration module is connected to the data acquisition module and the device control module, and is used to build a digital twin model of the device based on the preprocessed data, simulate the execution effect of the device control instructions in the cloud based on the digital twin model, and optimize the device control instructions.

[0008] In one embodiment, the system further comprises:

[0009] A security risk control module, connected to the data acquisition module and the cloud-edge collaboration module, is used to compare the pre-processed data with a preset security threshold, so as to generate corresponding security control instructions when the pre-processed data exceeds the preset security threshold;

[0010] The cloud-edge collaboration module is also used to verify the validity of the security control instructions through the digital twin model, and when the security control instructions are confirmed to be valid, output the security control instructions to the device control module.

[0011] In one embodiment, the system further comprises:

[0012] An equipment lifecycle management module, connected to the data acquisition module and the cloud-edge collaboration module, is used to evaluate the health status of the equipment based on the preprocessed data and generate at least one maintenance management strategy based on the health status to improve the operating efficiency and lifespan of the equipment;

[0013] The cloud-edge collaboration module is also used to simulate the performance changes of the equipment under multiple maintenance management strategies through the digital twin model to select the best maintenance management strategy.

[0014] To achieve the above objectives, the present application further provides an artificial intelligence-based automatic control method for oil and gas field exploitation and production processes, wherein the method is based on the artificial intelligence-based automatic control system for oil and gas field exploitation and production processes, and comprises:

[0015] Collecting initial data during oil and gas field exploitation and production in real time based on the multi-source sensor array, and processing the initial data based on the edge computing node to generate pre-processed data;

[0016] Acquire the pre-processed data and generate corresponding equipment control instructions to adjust the working status and production parameters of corresponding equipment during oil and gas field exploitation and production;

[0017] A digital twin model of the device is constructed according to the preprocessed data, and the execution effect of the device control instruction is simulated in the cloud based on the digital twin model, and the device control instruction is optimized.

[0018] In one embodiment, the step of processing the initial data based on the edge computing node to generate pre-processed data includes:

[0019] Cleaning, denoising, and formatting the initial data based on edge computing nodes to eliminate outliers and redundant information in the data;

[0020] Feature extraction and dimensionality reduction are performed on the initial data after cleaning, denoising and formatting, and target features are extracted to obtain preprocessed data.

[0021] In one embodiment, the step of acquiring the pre-processed data and generating corresponding equipment control instructions to adjust the working state and production parameters of corresponding equipment during oil and gas field exploitation and production includes:

[0022] Acquire the preprocessed data and perform multi-dimensional analysis based on a preset expert knowledge base to generate a corresponding equipment status assessment matrix;

[0023] Matching a corresponding algorithm set according to the device status evaluation matrix;

[0024] The output results of multiple control algorithms in the algorithm set are integrated through a dynamic weight allocation mechanism to generate a device control instruction parameter combination;

[0025] Verify the correlation between the device control instruction parameter combination and the current operating state of the device, and establish a corresponding dynamic control model;

[0026] Generate corresponding device control instructions based on the dynamic control model.

[0027] In one embodiment, the step of constructing a digital twin model of the device based on the preprocessed data includes:

[0028] Constructing a real-time data synchronization channel between the physical space and the virtual space of the digital twin model;

[0029] Establish a three-dimensional simulation model that includes the equipment's geometric structure, material properties, and motion constraints;

[0030] Based on the real-time data synchronization channel, the operating parameters and historical maintenance data in the preprocessed data are imported into the three-dimensional simulation model to construct a digital twin model of oil and gas field mining and production equipment.

[0031] In one embodiment, the step of simulating the execution effect of the device control instruction in the cloud based on the digital twin model includes:

[0032] Deploy a joint simulation platform on the cloud that includes a fluid mechanics simulation engine and a mechanical dynamics simulation engine;

[0033] Converting the device control instructions into corresponding simulation parameters and inputting them into the digital twin model;

[0034] Simulating the evolution of the operating state of the device within a preset time period through the joint simulation platform;

[0035] Based on the simulation results of the operation state evolution process, the output improvement rate, energy consumption change rate and equipment loss rate of the equipment are determined to evaluate the execution effect of the equipment control instructions.

[0036] In one embodiment, the step of optimizing the device control instruction includes:

[0037] Constructing a multi-dimensional evaluation matrix including production improvement rate, energy consumption change rate and equipment loss rate based on the execution effect of the equipment control instructions;

[0038] Iteratively optimizing the equipment control instruction parameter combination using a preset multi-objective optimization algorithm; the optimization objective function corresponding to the multi-objective optimization algorithm includes three dynamic weight coefficients, which respectively correspond to the optimization weights of the output improvement rate, the energy consumption change rate, and the equipment loss rate;

[0039] Dynamically adjusting the dynamic weight coefficients through a fuzzy logic controller to generate a Pareto optimal solution set;

[0040] Based on the real-time operating status of the equipment, select the optimization instruction parameter combination with the highest matching degree in the Pareto optimal solution set as the optimized equipment control instruction parameter combination;

[0041] The optimized device control instruction parameter combination is input into the digital twin model for three-stage verification:

[0042] The first stage verifies whether the transient response characteristics of the equipment meet the preset safety margin;

[0043] The second phase verifies whether the equipment's continuous operation stability reaches the expected threshold;

[0044] The third stage verifies whether the comprehensive energy efficiency index is better than the original control instruction;

[0045] When all three stages of verification are passed, the optimized device control instructions are output to the device control module.

[0046] In one embodiment, the method further comprises:

[0047] comparing the pre-processed data with a preset safety threshold;

[0048] When the pre-processed data exceeds a preset safety threshold, dynamically adjusting the sampling frequency of the multi-source sensor array to obtain verification data;

[0049] Reconstructing a local digital twin model based on the review data to perform fault tracing analysis to obtain a fault tracing report;

[0050] Generate corresponding safety control instructions based on the review data and fault tracing report;

[0051] When the security control instruction is confirmed to be valid, the security control instruction is output to the device control module.

[0052] The automatic control system for oil and gas field exploitation and production processes based on artificial intelligence proposed in this application includes a data acquisition module, an equipment control module and a cloud-edge collaboration module, wherein the data acquisition module includes a multi-source sensor array and an edge computing node, which is used to collect the initial data in the oil and gas field exploitation and production process in real time based on the multi-source sensor array, and process the initial data based on the edge computing node to generate pre-processed data; the equipment control module is connected to the data acquisition module, and is used to obtain the pre-processed data and generate corresponding equipment control instructions to adjust the working status and production parameters of the corresponding equipment in the oil and gas field exploitation and production process; the cloud-edge collaboration module is connected to the data acquisition module and the equipment control module, and is used to construct a digital twin model of the equipment according to the pre-processed data, simulate the execution effect of the equipment control instructions in the cloud based on the digital twin model, and optimize the equipment control instructions. In this way, the initial data is acquired in real time by the data acquisition module, and is pre-processed by the edge computing node to generate high-quality data input. The equipment control module generates preliminary equipment control instructions based on these data, while the cloud-edge collaboration module further uses the digital twin model and cloud simulation platform to simulate the execution effect of the equipment control instructions and optimize them, thereby forming a multi-level and multi-dimensional control strategy. It not only improves the operating efficiency and stability of the equipment, but also significantly reduces energy consumption and equipment loss, thereby improving the equipment control effect during oil and gas field exploitation and production. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0054] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0055] Figure 1 A schematic diagram of the structure of an embodiment of an automatic control system for oil and gas field exploitation and production based on artificial intelligence provided by this application;

[0056] Figure 2 A schematic diagram of the structure of another embodiment of the automatic control system for oil and gas field exploitation and production based on artificial intelligence provided by the present application;

[0057] Figure 3 A schematic structural diagram of another embodiment of the automatic control system for oil and gas field exploitation and production based on artificial intelligence provided by the present application;

[0058] Figure 4 A flow chart illustrating an embodiment of an automatic control method for oil and gas field exploitation and production based on artificial intelligence in this application;

[0059] Figure 5 for Figure 4 Detailed flowchart of step S100 in an embodiment;

[0060] Figure 6 for Figure 4 Detailed flow chart of step S200 in an embodiment;

[0061] Figure 7 for Figure 4 Detailed flowchart of step S300 in an embodiment;

[0062] Figure 8 for Figure 4 A detailed flowchart of another embodiment of step S300;

[0063] Figure 9 for Figure 4 A detailed flow chart of another embodiment of step S300;

[0064] Figure 10 A flow chart illustrating another embodiment of the automatic control method for oil and gas field exploitation and production based on artificial intelligence in this application.

[0065] Description of Figure Numbers:

[0066] 10. Data acquisition module; 20. Device control module; 30. Cloud-edge collaboration module; 40. Security risk control module; 50. Device lifecycle management module.

[0067] The purpose, features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0068] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0069] Existing technologies for controlling oil and gas field equipment typically rely on manual experience or parameter adjustments through fixed thresholds or simple feedback mechanisms. These typically employ PID control algorithms, rule bases based on expert experience, or static data models. These methods are somewhat effective when equipment is operating in a stable state. However, the complex and ever-changing oil and gas field production environment, coupled with the real-time and heterogeneous nature of multi-source data (such as pressure, temperature, and flow), makes it difficult for traditional control systems to achieve effective control results.

[0070] Based on this, the embodiment of the present application provides an automatic control system for oil and gas field exploitation and production process based on artificial intelligence, referring to Figure 1 The system includes a data acquisition module 10, a device control module 20, and a cloud-edge collaboration module 30. The data acquisition module 10 includes a multi-source sensor array and an edge computing node, and is configured to collect initial data from oil and gas field exploitation and production in real time based on the multi-source sensor array, and process the initial data based on the edge computing node to generate preprocessed data.

[0071] In this embodiment, the multi-source sensor array includes but is not limited to pressure sensors, temperature sensors, flow sensors, vibration sensors, etc., which are used to monitor various parameters in the oil and gas field exploitation and production process in real time. Edge computing nodes are small data centers or computing nodes deployed at the edge of the network close to the data source, with real-time data processing and analysis capabilities. By cleaning, denoising, and formatting the initial data through edge computing nodes, outliers and redundant information in the data can be effectively eliminated, thereby improving data quality. Furthermore, edge computing nodes can also perform feature extraction and dimensionality reduction on the processed data, extracting key features related to the target, and providing more accurate data support for subsequent equipment control.

[0072] In this embodiment, the device control module 20 is connected to the data acquisition module 10 and is used to acquire the preprocessed data and generate corresponding device control instructions to adjust the operating status and production parameters of the corresponding equipment during oil and gas field extraction and production. The device control module 20 can construct an equipment status assessment model based on the preprocessed data, use this model to perform real-time assessment of the equipment's operating status, and generate corresponding device control instructions based on the assessment results. Device control instructions may include, but are not limited to, adjusting equipment operating parameters, starting or stopping equipment operation, etc., to meet the actual needs of oil and gas field extraction and production. Furthermore, the device control module 20 can dynamically adjust the control strategy based on the equipment's real-time operating status to ensure stable operation and efficient production.

[0073] In this embodiment, the cloud-edge collaboration module 30 is connected to the data acquisition module 10 and the device control module 20 to construct a digital twin model of the device based on the preprocessed data. Based on the digital twin model, the cloud-edge collaboration module 30 simulates the execution effects of the device control instructions in the cloud and optimizes the device control instructions. Specifically, the cloud-edge collaboration module 30 first uses a deep learning algorithm to deeply mine the preprocessed data to identify underlying patterns and laws of device operation. Subsequently, based on these patterns, the cloud-edge collaboration module 30 constructs a high-precision digital twin model in the cloud that accurately reflects the device's geometry, material properties, and motion constraints. After constructing the digital twin model, the cloud-edge collaboration module 30 converts the device control instructions into corresponding simulation parameters and inputs them into the digital twin model. By simulating the evolution of the device's operating state over a preset time period, the cloud-edge collaboration module 30 can predict the actual execution effects of the device control instructions, including indicators such as the device's yield increase rate, energy consumption change rate, and device loss rate. Based on these prediction results, the cloud-edge collaboration module 30 uses an optimization algorithm to iteratively optimize the device control instructions to find the optimal control strategy. Optimized equipment control instructions can not only effectively improve the operating efficiency and stability of the equipment, but also significantly reduce energy consumption and equipment loss, thereby maximizing the economic benefits of oil and gas field exploration and production.

[0074] In one possible implementation, reference Figure 2 The system also includes a security risk control module 40, which is connected to the data acquisition module 10 and the cloud-edge collaboration module 30, and is used to compare the pre-processed data with a preset security threshold to generate corresponding security control instructions when the pre-processed data exceeds the preset security threshold.

[0075] In this embodiment, the preset safety threshold refers to the safety boundary value set for various key parameters and equipment status during the oil and gas field exploitation and production process. These safety thresholds are determined based on industry standards, historical data analysis, and expert experience, and are intended to ensure that the equipment operates within a safe operating range and prevent safety accidents or equipment damage caused by abnormal parameter fluctuations. Specifically, the safety thresholds include pressure upper limits, temperature ranges, flow limits, etc. When any parameter in the pre-processed data exceeds the preset safety threshold, the safety risk control module 40 will immediately trigger an alarm and generate corresponding safety control instructions to take emergency measures to adjust the equipment status to a safe range. These safety control instructions correspond to different pre-processed data, including shutdown instructions, parameter adjustment instructions, or starting backup equipment, etc., to reduce safety risks and ensure the smooth progress of oil and gas field exploitation and production processes.

[0076] In this embodiment, the cloud-edge collaboration module 30 is also used to verify the validity of the security control instruction through the digital twin model, and when the security control instruction is confirmed to be valid, output the security control instruction to the device control module 20.

[0077] In this embodiment, the cloud-edge collaboration module 30 fully utilizes the simulation capabilities of the digital twin model when verifying the effectiveness of the security control instructions. Specifically, the cloud-edge collaboration module 30 converts the security control instructions into corresponding simulation parameters and inputs them into the digital twin model to simulate the evolution of the operating state of the device after taking security control measures. By comparing the simulation results with the expected security state, the cloud-edge collaboration module 30 can evaluate the effectiveness of the security control instructions. After confirming that the security control instructions can effectively adjust the device state to a safe range, the cloud-edge collaboration module 30 outputs the instructions to the device control module 20, and the device control module 20 executes the corresponding security control measures.

[0078] In one possible implementation, reference Figure 3 The system also includes an equipment lifecycle management module 50, which is connected to the data acquisition module 10 and the cloud-edge collaboration module 30, and is used to evaluate the health status of the equipment based on the preprocessed data, and generate at least one maintenance management strategy based on the health status to improve the operating efficiency and life of the equipment.

[0079] In this embodiment, preprocessed data includes, but is not limited to, equipment vibration frequency, temperature fluctuations, and pressure stability, directly reflecting equipment wear, potential failures, and performance degradation. The equipment lifecycle management module 50 uses a deep learning algorithm to conduct in-depth analysis of this preprocessed data to identify key indicators and trends in equipment health. Furthermore, based on this preprocessed data, the equipment lifecycle management module 50 can construct an equipment health assessment model to assess the equipment's health in real time and predict its remaining useful life.

[0080] In this embodiment, after obtaining the equipment health status assessment results, the equipment lifecycle management module 50 further generates corresponding maintenance management strategies based on the assessment results, including but not limited to regular maintenance plans, preventive maintenance measures, and emergency repair tasks. This ensures that the equipment operates in optimal conditions, extends its service life, and reduces the risk of production interruptions due to equipment failure. Furthermore, the equipment lifecycle management module 50 can dynamically adjust the maintenance management strategies based on the equipment's real-time operating status and maintenance history to meet the actual needs of oil and gas field extraction and production.

[0081] In this embodiment, the cloud-edge collaboration module 30 is also used to simulate the performance changes of the equipment under multiple maintenance management strategies through the digital twin model to select the best maintenance management strategy.

[0082] In this embodiment, after receiving the multiple maintenance management strategies generated by the equipment lifecycle management module 50, the cloud-edge collaboration module 30 uses the constructed digital twin model to perform simulation analysis. Specifically, the cloud-edge collaboration module 30 converts different maintenance management strategies into corresponding simulation parameters, and inputs them into the digital twin model to simulate the performance change process of the equipment under different maintenance strategies. By comparing the simulation results, the cloud-edge collaboration module 30 can evaluate the impact of various maintenance management strategies on equipment performance and select the best maintenance strategy. This step not only improves the scientificity and effectiveness of maintenance management, but also further ensures the smooth operation of oil and gas field exploitation and production processes.

[0083] In this embodiment, an automatic control system for oil and gas field exploitation and production based on artificial intelligence includes a data acquisition module 10, an equipment control module 20, and a cloud-edge collaboration module 30, wherein the data acquisition module 10 includes a multi-source sensor array and an edge computing node, and is used to collect initial data in the oil and gas field exploitation and production process in real time based on the multi-source sensor array, and process the initial data based on the edge computing node to generate preprocessed data; the equipment control module 20 is connected to the data acquisition module 10, and is used to obtain the preprocessed data and generate corresponding equipment control instructions to adjust the working state and production parameters of the corresponding equipment in the oil and gas field exploitation and production process; the cloud-edge collaboration module 30 is connected to the data acquisition module 10 and the equipment control module 20, and is used to build a digital twin model of the equipment based on the preprocessed data, simulate the execution effect of the equipment control instruction in the cloud based on the digital twin model, and optimize the equipment control instruction. In this way, the initial data is acquired in real time by the data acquisition module 10, and is preprocessed by the edge computing node to generate high-quality data input. The equipment control module 20 generates preliminary equipment control instructions based on these data, and the cloud-edge collaboration module 30 further uses the digital twin model and cloud simulation platform to simulate the execution effect of the equipment control instructions and optimize them, thereby forming a multi-level and multi-dimensional control strategy, which not only improves the operating efficiency and stability of the equipment, but also significantly reduces energy consumption and equipment loss, thereby improving the equipment control effect during oil and gas field exploitation and production.

[0084] It should be noted that the above examples are only used to understand the present application and do not constitute a limitation on the automatic control system for oil and gas field exploitation and production processes based on artificial intelligence in the present application. More simple transformations based on this technical concept are all within the scope of protection of the present application.

[0085] The present application also provides an automatic control method for oil and gas field exploitation and production process based on artificial intelligence, wherein the method is based on the automatic control system for oil and gas field exploitation and production process based on artificial intelligence, with reference to Figure 4 , the method includes steps S100 to S300, wherein:

[0086] Step S100 , collecting initial data in the oil and gas field exploitation and production process in real time based on the multi-source sensor array, and processing the initial data based on the edge computing node to generate preprocessed data.

[0087] In this embodiment, the multi-source sensor array includes but is not limited to pressure sensors, temperature sensors, flow sensors, vibration sensors, etc., which are used to monitor various parameters in the oil and gas field exploitation and production process in real time. Edge computing nodes are small data centers or computing nodes deployed at the edge of the network close to the data source, with real-time data processing and analysis capabilities. By cleaning, denoising, and formatting the initial data through edge computing nodes, outliers and redundant information in the data can be effectively eliminated, thereby improving data quality. Furthermore, edge computing nodes can also perform feature extraction and dimensionality reduction on the processed data, extracting key features related to the target, and providing more accurate data support for subsequent equipment control.

[0088] In one possible implementation, reference Figure 5 The step of processing the initial data based on the edge computing node to generate pre-processed data includes steps S110 to S120, wherein:

[0089] Step S110: Cleaning, denoising, and formatting the initial data based on the edge computing node to eliminate outliers and redundant information in the data.

[0090] In this embodiment, data cleaning utilizes a sliding window mechanism to perform real-time verification of sensor data. Specifically, dynamic threshold intervals can be set to detect sudden changes in pressure and temperature parameters, and the interquartile range method can be used to identify discrete anomalies in flow parameters. Wavelet transform denoising is performed on time series data collected by vibration sensors to eliminate high-frequency noise interference. Formatting primarily targets multi-source heterogeneous data, converting monitored values of different dimensions, such as pressure (MPa), temperature (°C), and flow (m³ / h), into standardized floating-point numbers. Timestamp alignment technology is used to achieve millisecond-level synchronization of multi-channel data.

[0091] Step S120 , performing feature extraction and dimensionality reduction processing on the initial data after cleaning, denoising and formatting, and extracting target features to obtain preprocessed data.

[0092] In this embodiment, the target features can also be extracted by using a sliding window statistical method to calculate the time domain mean, variance and peak value, and extracting the fundamental frequency component and the energy proportion of the harmonics in the vibration spectrum through fast Fourier transform. The dimensionality reduction process is combined with principal component analysis (PCA) and t-distributed stochastic neighbor embedding (t-SNE) algorithm to project the high-dimensional sensor data into a three-dimensional feature space, retaining the key discriminant information of the equipment operating status. For the periodic fluctuation characteristics of the flow data, the dynamic time warping (DTW) algorithm is used to align the time series patterns under different working conditions to construct a multi-dimensional feature matrix with spatiotemporal correlation. The final generated pre-processed data includes the equipment status feature vector, the working condition label matrix and the data quality assessment index. After being standardized and unified in dimension, it is transmitted to the equipment control module 20 via the industrial bus protocol.

[0093] In one optimized implementation, a wavelet packet energy entropy calculation module is added to the feature extraction process for multiphase flow pattern identification in oil and gas wells. By performing multi-layer wavelet packet decomposition on the pressure pulsation signal and calculating the energy distribution entropy of each sub-band, a dedicated dataset containing multidimensional flow pattern feature vectors is constructed. This feature set can be cross-modally fused with the visual features of the wellhead camera to improve the accuracy of gas-liquid two-phase flow pattern identification.

[0094] In this embodiment, preprocessed data is encapsulated into structured data packets with timestamps via the OPC UA protocol. Each packet contains the device ID, feature vector, hash value of the original data, and the digital signature of the edge node, ensuring data integrity and traceability during transmission. In this embodiment, feature extraction and dimensionality reduction are performed on the initial data after cleaning, denoising, and formatting. This reduces the dimensionality of the preprocessed data compared to the initial sensor data. While retaining key operational characteristics of the device, it effectively reduces the computational load of subsequent model inference and reduces the latency in generating device control commands, thus meeting the real-time requirements of oil and gas field production scenarios.

[0095] Step S200: acquiring the pre-processed data and generating corresponding equipment control instructions to adjust the working state and production parameters of corresponding equipment during the oil and gas field exploitation and production process.

[0096] In this embodiment, the device control module 20 can construct a device status assessment model based on preprocessed data. This model can be used to assess the device's operating status in real time and generate corresponding device control instructions based on the assessment results. These device control instructions may include, but are not limited to, adjusting device operating parameters, starting or stopping device operation, and other tasks to meet the actual needs of oil and gas field extraction and production. Furthermore, the device control module 20 can dynamically adjust control strategies based on the device's real-time operating status to ensure stable operation and efficient production.

[0097] In one possible implementation, reference Figure 6 , the step S200 includes steps S210 to S250, wherein:

[0098] Step S210: Acquire the pre-processed data and perform multi-dimensional analysis based on a preset expert knowledge base to generate a corresponding equipment status evaluation matrix.

[0099] In this embodiment, the preset expert knowledge base can be a knowledge system constructed based on professional knowledge and historical experience in the field of oil and gas field exploration and production, and includes equipment failure modes, common problems and their solutions, safe ranges for equipment operating parameters, and optimization suggestions. The equipment status assessment matrix quantitatively evaluates the operating status of the equipment by integrating key features in the preprocessed data and the results of the equipment status assessment model, combined with the multi-dimensional information in the preset expert knowledge base. Specifically, the equipment status assessment matrix can include multiple dimensions such as equipment health index, failure probability prediction, and performance degradation trend to comprehensively reflect the current status and future trends of the equipment.

[0100] Step S220: Matching a corresponding algorithm set according to the device status evaluation matrix.

[0101] In this embodiment, the algorithm set includes multiple algorithms for device control and analysis. When matching the algorithm set, the system selects the algorithm most appropriate for the current device state based on various dimensions in the device state assessment matrix, such as device health index, failure probability prediction, and performance degradation trend, to generate corresponding device control instructions. This step ensures the targeted and effective device control instructions, further improving the effectiveness of device control during oil and gas field extraction and production. Furthermore, an algorithm set mapping table is constructed to facilitate algorithm set matching. This mapping table associates different state intervals in the device state assessment matrix with pre-set algorithm sets. When the results of the device state assessment matrix fall within a certain state interval, the corresponding algorithm set is automatically matched, allowing for the subsequent generation of targeted device control instructions. The algorithm set includes a variety of control algorithms and optimization strategies, such as PID control algorithms, fuzzy control algorithms, and genetic algorithms, to meet the control requirements of different device states and operating scenarios. By accurately matching the algorithm set, the device control module 20 can more accurately generate device control instructions, ensuring that equipment during oil and gas field extraction and production is always in optimal operating condition.

[0102] Step S230 , fusing the output results of the multiple control algorithms in the algorithm set through a dynamic weight allocation mechanism to generate a device control instruction parameter combination.

[0103] In this embodiment, a dynamic weight allocation mechanism assigns corresponding weights to each control algorithm based on different dimensions in the device status assessment matrix, such as device health index, failure probability prediction, and performance degradation trend. The size of the weight reflects the importance and influence of each algorithm in the current device state. The output results of multiple control algorithms are fused through weighted summation or weighted averaging to generate a set of optimal device control instruction parameter combinations. In this embodiment, dynamic weight allocation and algorithm fusion calculations enhance the complementarity and synergy between different algorithms, ensuring the accuracy and reliability of device control instructions.

[0104] Step S240 , verifying the correlation between the device control instruction parameter combination and the current operating state of the device, and establishing a corresponding dynamic control model.

[0105] In this embodiment, correlation verification can utilize a Monte Carlo random sampling method to generate 100,000 simulation samples within the feasible domain of device operating parameters. By calculating the Pearson correlation coefficient and cosine similarity between command parameter combinations and real-time device operating data, optimized parameter combinations with correlation coefficients greater than a preset threshold (e.g., 0.95) are selected. The dynamic control model can utilize a long-short-term memory network architecture, using historical device operating sequences as a training set. A nonlinear mapping relationship between control parameters and device performance indicators is established, and network weight parameters are continuously updated through an online learning mechanism.

[0106] Step S250: Generate corresponding device control instructions based on the dynamic control model.

[0107] In this embodiment, the generated device control instructions contain the device ID code, execution time window, parameter adjustment range, and safety constraints, and are structured and encapsulated using the OPC UA protocol. The pre-verification process invokes the virtual debugging function of the digital twin model through the cloud-edge collaboration module 30. The dynamic response curve of the device after the control instruction is executed is simulated in a 3D visualization interface, focusing on verifying key indicators such as whether pressure fluctuations exceed safety thresholds and whether temperature gradient changes meet material tolerance standards. When the simulation results meet the preset constraints, the control instructions are transmitted to the PLC control system through the Industrial Internet of Things gateway. If a potential risk is detected, the control rollback mechanism is triggered, automatically calling the historical optimization solution under similar operating conditions from the backup control strategy library.

[0108] Step S300: construct a digital twin model of the device according to the preprocessed data, simulate the execution effect of the device control instruction in the cloud based on the digital twin model, and optimize the device control instruction.

[0109] In this embodiment, the cloud-edge collaboration module 30 first uses a deep learning algorithm to deeply mine the preprocessed data to identify the underlying laws and patterns of device operation. Subsequently, based on these laws and patterns, the cloud-edge collaboration module 30 constructs a high-precision digital twin model in the cloud that accurately reflects the device's geometric structure, material properties, and motion constraints. After constructing the digital twin model, the cloud-edge collaboration module 30 converts the device control instructions into corresponding simulation parameters and inputs them into the digital twin model. By simulating the evolution of the device's operating state within a preset time period, the cloud-edge collaboration module 30 can predict the actual execution effect of the device control instructions, including indicators such as the device's output improvement rate, energy consumption change rate, and equipment loss rate. Based on these prediction results, the cloud-edge collaboration module 30 uses an optimization algorithm to iteratively optimize the device control instructions to seek the optimal control strategy. The optimized device control instructions can not only effectively improve the operating efficiency and stability of the equipment, but also significantly reduce energy consumption and equipment loss, thereby maximizing the economic benefits of oil and gas field extraction and production.

[0110] In one possible implementation, reference Figure 7 , the step of constructing the digital twin model of the device according to the preprocessed data includes steps S310A to S330A, wherein:

[0111] Step S310A: construct a real-time data synchronization channel between the physical space and the virtual space of the digital twin model.

[0112] In this embodiment, a real-time data synchronization channel between the physical and virtual spaces is implemented based on the OPC UA protocol, ensuring seamless integration between real-time device operational data and the digital twin model. Through this synchronization channel, various sensor data, operational status information, and fault alarms from physical devices can be transmitted to the cloud in real time, providing accurate and comprehensive data support for the digital twin model. Simultaneously, simulation results and optimization instructions from the digital twin model can also be promptly fed back to the physical devices through this channel, enabling precise execution of control instructions and closed-loop optimization.

[0113] Step S320A: Establish a three-dimensional simulation model including the device geometry, material properties, and motion constraints.

[0114] In this embodiment, the 3D simulation model was constructed using computer-aided design and computer-aided engineering software to ensure high accuracy and realism. This model not only encompasses the device's geometric structure but also details its material properties and motion constraints, fully reflecting its physical characteristics and operating state. When constructing the 3D simulation model, the complexity and diversity of the device were fully considered to ensure that the model accurately simulates its behavior under different operating conditions.

[0115] Step S330A: Import the operating parameters and historical maintenance data in the preprocessed data into the three-dimensional simulation model based on the real-time data synchronization channel to construct a digital twin model of oil and gas field mining and production equipment.

[0116] In this embodiment, the imported operating parameters include, but are not limited to, real-time monitoring data such as pressure, temperature, flow, and vibration. After preprocessing, these data can accurately reflect the actual operating status of the equipment. Historical maintenance data records the equipment's past maintenance records, fault conditions, and their solutions, providing a wealth of historical experience and knowledge for the digital twin model. By importing this data into a three-dimensional simulation model and combining it with deep learning algorithms for training and optimization, a high-precision digital twin model is ultimately constructed. This model can simulate the equipment's operating status in real time, predict future trends, and provide strong support for optimizing equipment control instructions.

[0117] In one possible implementation, reference Figure 8 The step of simulating the execution effect of the device control instruction in the cloud based on the digital twin model includes steps S310B to S340B, wherein:

[0118] S310B deploys a joint simulation platform on the cloud that includes a fluid mechanics simulation engine and a mechanical dynamics simulation engine.

[0119] In this embodiment, the co-simulation platform utilizes a high-performance cloud computing architecture to ensure real-time and accurate simulations. The fluid dynamics simulation engine focuses on simulating the flow and pressure distribution of fluids during oil and gas well extraction and production, while the mechanical dynamics simulation engine simulates the mechanical motion and interactions of equipment. By integrating these two simulation engines, the co-simulation platform can comprehensively simulate equipment operating conditions under complex operating conditions, providing accurate and reliable predictions for optimizing equipment control instructions.

[0120] S320B, converting the device control instructions into corresponding simulation parameters and inputting them into the digital twin model.

[0121] In this embodiment, device control instructions are parameterized and converted into a series of quantifiable simulation parameters. These parameters encompass the device's operating parameters, adjustment strategies, and safety constraints, ensuring accurate simulation of the instructions within the digital twin model. This step crystallizes the abstract concept of device control instructions into numerical inputs recognizable by the digital twin model, providing a foundation for subsequent simulation analysis.

[0122] S330B, simulating the evolution of the operating state of the device within a preset time period through the joint simulation platform.

[0123] In this embodiment, the joint simulation platform, based on a digital twin model, combines fluid dynamics simulation engines with mechanical dynamics simulation engines to dynamically simulate the equipment's operating status over a preset time period. The simulation process not only considers the equipment's geometry, material properties, and motion constraints, but also fully incorporates real-time monitoring data and historical maintenance data to ensure the accuracy and reliability of the simulation results. This step can intuitively demonstrate the execution effect of equipment control instructions in the actual operating environment, including the changing trends of key indicators such as equipment output, energy consumption, and losses, providing a reference for subsequent instruction optimization.

[0124] S340B, determining the output improvement rate, energy consumption change rate and equipment loss rate of the equipment based on the simulation results of the operation state evolution process, so as to evaluate the execution effect of the equipment control instruction.

[0125] In this example, the impact of device control instructions on equipment performance is quantitatively evaluated by comparing and analyzing simulation results with preset target values. Specifically, the yield improvement rate reflects the degree of improvement in equipment production capacity after executing the instruction, the energy consumption change rate reveals the effectiveness of the instruction in optimizing energy consumption, and the equipment wear rate reflects the effect of the instruction on extending equipment life. These evaluation indicators not only provide data support for optimizing device control instructions but also provide a basis for subsequent decision-making.

[0126] In one possible implementation, reference Figure 9 The step of optimizing the device control instruction includes steps S310C to S360C, wherein:

[0127] Step S310C: constructing a multi-dimensional evaluation matrix including output improvement rate, energy consumption change rate and equipment loss rate based on the execution effect of the equipment control instruction.

[0128] In this embodiment, the construction of the multi-dimensional evaluation matrix comprehensively considers the comprehensive performance of the equipment control instructions in terms of output improvement, energy consumption optimization, and equipment loss. By quantitatively analyzing the simulation results of different control instructions, a matrix containing multiple evaluation dimensions is formed, which provides a comprehensive and objective evaluation basis for the optimization of instructions. Specifically, the output improvement rate dimension reflects the contribution of the instruction to improving the production efficiency of the equipment, the energy consumption change rate dimension reveals the effect of the instruction in energy conservation and emission reduction, and the equipment loss rate dimension reflects the positive role of the instruction in extending the service life of the equipment. These evaluation dimensions are interrelated and mutually constrained, and together constitute an indicator system for the optimization of equipment control instructions.

[0129] Step S320C, using a preset multi-objective optimization algorithm to iteratively optimize the equipment control instruction parameter combination; the optimization objective function corresponding to the multi-objective optimization algorithm includes three dynamic weight coefficients, which correspond to the optimization weights of the output improvement rate, the energy consumption change rate and the equipment loss rate respectively.

[0130] In this embodiment, the pre-set multi-objective optimization algorithm dynamically adjusts these three weighting coefficients during an iterative process to find the optimal balance between yield improvement, energy efficiency optimization, and equipment loss. Specifically, the weighting coefficients are adjusted based on historical simulation data and a learning algorithm, gradually approaching the optimal solution through trial and error and feedback. This step ensures that the optimization of equipment control instructions comprehensively considers multiple performance indicators to maximize overall benefits.

[0131] Step S330C: dynamically adjusting the dynamic weight coefficient through a fuzzy logic controller to generate a Pareto optimal solution set.

[0132] In this embodiment, the fuzzy logic controller adjusts the dynamic weight coefficients in real time based on the historical execution results and current operating status of the equipment control instructions. By introducing fuzzy sets and fuzzy rules, the controller can process uncertainty and achieve fine-tuning of the weight coefficients. The Pareto optimal solution set generated in this step includes multiple combinations of equipment control instruction parameters that achieve a good balance between yield improvement rate, energy consumption change rate, and equipment loss rate, providing decision makers with a wide range of options.

[0133] Step S340C: selecting the optimized instruction parameter combination with the highest matching degree from the Pareto optimal solution set based on the real-time operating status of the device as the optimized device control instruction parameter combination.

[0134] In this embodiment, the real-time operating status of the equipment is transmitted to the cloud in real time via a real-time data synchronization channel between the physical and virtual spaces, ensuring data accuracy and timeliness. Within the Pareto optimal solution set, an advanced matching algorithm comprehensively considers the equipment's current status, operating parameters, and historical maintenance data, selecting a set of parameters that best matches the equipment's real-time status from multiple optimized instruction parameter combinations. This set of parameters not only effectively improves equipment operating efficiency and output, but also strikes an optimal balance between energy consumption optimization and equipment loss, ensuring the continued stability and cost-effectiveness of oil and gas field extraction and production processes.

[0135] Step S350C: Input the optimized device control instruction parameter combination into the digital twin model for three-stage verification:

[0136] The first stage verifies whether the transient response characteristics of the equipment meet the preset safety margin;

[0137] The second phase verifies whether the equipment's continuous operation stability reaches the expected threshold;

[0138] The third stage verifies whether the comprehensive energy efficiency index is better than the original control instructions.

[0139] In this embodiment, the above three-stage verification process can ensure the reliability and efficiency of the optimized device control instruction parameter combination in actual application. In the first stage, by simulating the transient response of the equipment under extreme working conditions, it is verified whether it can operate quickly and stably while ensuring safety. The second stage focuses on the stability of the equipment during long-term continuous operation to ensure that the optimized instructions do not cause fluctuations in equipment performance or abnormal wear. The third stage comprehensively evaluates the energy efficiency indicators of the equipment, including energy utilization, emission levels and operating costs, to ensure that the optimized control instructions can bring significant economic and environmental benefits in actual production.

[0140] In step S360C, if all three stages of verification pass, the optimized device control instructions are output to the device control module 20. In this embodiment, the optimized device control instructions undergo rigorous verification and validation before being securely and reliably transmitted to the device control module 20. Upon receiving the instructions, the device control module 20 immediately activates the corresponding execution mechanism, converting the optimized instructions into specific device operation instructions to ensure that the device operates according to the intended method and parameters. This step achieves seamless integration from the digital twin model to the actual device, ensuring that the optimization results can be quickly converted into actual production benefits.

[0141] In one possible implementation, reference Figure 10 The method further includes steps S410 to S450, wherein:

[0142] Step S410: Compare the pre-processed data with a preset safety threshold.

[0143] In this embodiment, pre-processed data is monitored in real time and compared with a preset safety threshold. This step is the basis of security monitoring and can promptly detect potential safety hazards. If the pre-processed data exceeds the safety threshold, the system immediately enters emergency response mode.

[0144] Step S420 : When the pre-processed data exceeds a preset safety threshold, dynamically adjust the sampling frequency of the multi-source sensor array to obtain verification data.

[0145] In this embodiment, the system dynamically adjusts the sampling frequency of the multi-source sensor array to obtain more accurate verification data. This adjustment, based on real-time data analysis, can specifically improve the monitoring accuracy of key parameters and provide accurate data support for subsequent fault tracing.

[0146] Step S430: Reconstruct the local digital twin model based on the review data to perform fault tracing analysis to obtain a fault tracing report.

[0147] In this embodiment, the cloud-edge collaboration module 30 uses the verified data to reconstruct a local digital twin model for fault tracing analysis. This process simulates the device's operating status and the fault process, accurately locating the source of the fault and generating a detailed fault tracing report. This report not only includes the specific location of the fault but also analyzes the cause and impact of the fault, providing an important basis for subsequent repair and optimization.

[0148] Step S440: Generate corresponding safety control instructions based on the review data and the fault tracing report.

[0149] In this embodiment, after confirming the fault tracing report, corresponding safety control instructions are generated based on the reviewed data and the fault tracing report to quickly eliminate safety hazards, prevent further deterioration of the fault, and ensure the safety of equipment and personnel. In this embodiment, the generation of safety control instructions takes into account the actual operating status of the equipment and the type of fault, ensuring the pertinence and effectiveness of the instructions.

[0150] Step S450 : When it is confirmed that the security control instruction is valid, the security control instruction is output to the device control module 20 .

[0151] When the security control instruction is verified to be valid, the system outputs it to the device control module 20. Upon receiving the instruction, the device control module 20 immediately activates the corresponding security execution mechanism to ensure that the device can operate in accordance with the requirements of the security control instruction. This achieves a seamless connection from security monitoring to actual device control, ensuring that the security control instruction can be quickly and accurately converted into actual safety measures.

[0152] In this embodiment, the automatic control method for oil and gas field exploitation and production based on artificial intelligence collects the initial data of the oil and gas field exploitation and production process in real time based on the multi-source sensor array, and processes the initial data based on the edge computing node to generate preprocessed data, and then obtains the preprocessed data and generates corresponding equipment control instructions to adjust the working status and production parameters of the corresponding equipment in the oil and gas field exploitation and production process. Then, a digital twin model of the equipment is constructed according to the preprocessed data, and the execution effect of the equipment control instructions is simulated in the cloud based on the digital twin model, and the equipment control instructions are optimized, thereby forming a multi-level and multi-dimensional control strategy, which not only improves the operating efficiency and stability of the equipment, but also significantly reduces energy consumption and equipment loss, thereby improving the equipment control effect in the oil and gas field exploitation and production process.

[0153] The above description is only part of the embodiments of the present application and does not limit the patent scope of the present application. All equivalent structural transformations made by using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect application in other related technical fields are included in the patent protection scope of the present application.

Claims

1. An automatic control method for oil and gas field exploitation and production based on artificial intelligence, characterized in that: The method includes: Collecting initial data from oil and gas field exploitation and production in real time based on a multi-source sensor array, and processing the initial data based on edge computing nodes to generate pre-processed data; Acquire the pre-processed data and generate corresponding equipment control instructions to adjust the working status and production parameters of corresponding equipment during oil and gas field exploitation and production; Constructing a digital twin model of the device according to the preprocessed data, simulating the execution effect of the device control instruction in the cloud based on the digital twin model, and optimizing the device control instruction; The step of optimizing the device control instruction comprises: Constructing a multi-dimensional evaluation matrix including production improvement rate, energy consumption change rate and equipment loss rate based on the execution effect of the equipment control instructions; Iteratively optimizing the equipment control instruction parameter combination using a preset multi-objective optimization algorithm; the optimization objective function corresponding to the multi-objective optimization algorithm includes three dynamic weight coefficients, which respectively correspond to the optimization weights of the output improvement rate, the energy consumption change rate, and the equipment loss rate; Dynamically adjusting the dynamic weight coefficients through a fuzzy logic controller to generate a Pareto optimal solution set; Based on the real-time operating status of the equipment, select the optimization instruction parameter combination with the highest matching degree in the Pareto optimal solution set as the optimized equipment control instruction parameter combination; The optimized device control instruction parameter combination is input into the digital twin model for three-stage verification: The first stage verifies whether the transient response characteristics of the equipment meet the preset safety margin; The second phase verifies whether the equipment's continuous operation stability reaches the expected threshold; The third stage verifies whether the comprehensive energy efficiency index is better than the original control instruction; When all three stages of verification are passed, the optimized device control instructions are output to the device control module.

2. The automatic control method for oil and gas field exploitation and production process based on artificial intelligence according to claim 1, characterized in that: The step of processing the initial data based on the edge computing node to generate pre-processed data includes: Cleaning, denoising, and formatting the initial data based on edge computing nodes to eliminate outliers and redundant information in the data; Feature extraction and dimensionality reduction are performed on the initial data after cleaning, denoising and formatting, and target features are extracted to obtain preprocessed data.

3. The automatic control method for oil and gas field exploitation and production based on artificial intelligence according to claim 1, characterized in that: The step of obtaining the pre-processed data and generating corresponding equipment control instructions to adjust the working state and production parameters of corresponding equipment during oil and gas field exploitation and production includes: Acquire the preprocessed data and perform multi-dimensional analysis based on a preset expert knowledge base to generate a corresponding equipment status assessment matrix; Matching a corresponding algorithm set according to the device status evaluation matrix; The output results of multiple control algorithms in the algorithm set are integrated through a dynamic weight allocation mechanism to generate a device control instruction parameter combination; Verify the correlation between the device control instruction parameter combination and the current operating state of the device, and establish a corresponding dynamic control model; Generate corresponding device control instructions based on the dynamic control model.

4. The automatic control method for oil and gas field exploitation and production process based on artificial intelligence according to claim 1, characterized in that: The step of constructing a digital twin model of the device according to the preprocessed data includes: Constructing a real-time data synchronization channel between the physical space and the virtual space of the digital twin model; Establish a three-dimensional simulation model that includes the equipment's geometric structure, material properties, and motion constraints; Based on the real-time data synchronization channel, the operating parameters and historical maintenance data in the preprocessed data are imported into the three-dimensional simulation model to construct a digital twin model of oil and gas field mining and production equipment.

5. The automatic control method for oil and gas field exploitation and production process based on artificial intelligence according to claim 4, characterized in that: The step of simulating the execution effect of the device control instruction in the cloud based on the digital twin model includes: Deploy a joint simulation platform on the cloud that includes a fluid mechanics simulation engine and a mechanical dynamics simulation engine; Converting the device control instructions into corresponding simulation parameters and inputting them into the digital twin model; Simulating the evolution of the operating state of the device within a preset time period through the joint simulation platform; Based on the simulation results of the operation state evolution process, the output improvement rate, energy consumption change rate and equipment loss rate of the equipment are determined to evaluate the execution effect of the equipment control instructions.

6. The automatic control method for oil and gas field exploitation and production process based on artificial intelligence according to claim 5, characterized in that: The method further comprises: comparing the pre-processed data with a preset safety threshold; When the pre-processed data exceeds a preset safety threshold, dynamically adjusting the sampling frequency of the multi-source sensor array to obtain verification data; Reconstructing a local digital twin model based on the review data to perform fault tracing analysis to obtain a fault tracing report; Generate corresponding safety control instructions based on the review data and fault tracing report; When the security control instruction is confirmed to be valid, the security control instruction is output to the device control module.

7. An automatic control system for oil and gas field exploitation and production based on artificial intelligence, characterized in that: The system comprises: A data acquisition module, comprising a multi-source sensor array and an edge computing node, for collecting initial data from oil and gas field exploitation and production in real time based on the multi-source sensor array, and processing the initial data based on the edge computing node to generate pre-processed data; An equipment control module, connected to the data acquisition module, is used to obtain the pre-processed data and generate corresponding equipment control instructions to adjust the working status and production parameters of corresponding equipment during oil and gas field exploitation and production; A cloud-edge collaboration module, connected to the data acquisition module and the device control module, is used to build a digital twin model of the device based on the preprocessed data, simulate the execution effect of the device control instructions in the cloud based on the digital twin model, and optimize the device control instructions; Optimizing the device control instructions includes: Constructing a multi-dimensional evaluation matrix including production improvement rate, energy consumption change rate and equipment loss rate based on the execution effect of the equipment control instructions; Iteratively optimizing the equipment control instruction parameter combination using a preset multi-objective optimization algorithm; the optimization objective function corresponding to the multi-objective optimization algorithm includes three dynamic weight coefficients, which respectively correspond to the optimization weights of the output improvement rate, the energy consumption change rate, and the equipment loss rate; Dynamically adjusting the dynamic weight coefficients through a fuzzy logic controller to generate a Pareto optimal solution set; Based on the real-time operating status of the equipment, select the optimization instruction parameter combination with the highest matching degree in the Pareto optimal solution set as the optimized equipment control instruction parameter combination; The optimized device control instruction parameter combination is input into the digital twin model for three-stage verification: The first stage verifies whether the transient response characteristics of the equipment meet the preset safety margin; The second phase verifies whether the equipment's continuous operation stability reaches the expected threshold; The third stage verifies whether the comprehensive energy efficiency index is better than the original control instruction; When all three stages of verification are passed, the optimized device control instructions are output to the device control module.

8. The automatic control system for oil and gas field exploitation and production process based on artificial intelligence according to claim 7, characterized in that: The system further comprises: A security risk control module, connected to the data acquisition module and the cloud-edge collaboration module, is used to compare the pre-processed data with a preset security threshold, so as to generate corresponding security control instructions when the pre-processed data exceeds the preset security threshold; The cloud-edge collaboration module is also used to verify the validity of the security control instructions through the digital twin model, and when the security control instructions are confirmed to be valid, output the security control instructions to the device control module.

9. The automatic control system for oil and gas field exploitation and production process based on artificial intelligence according to claim 7, characterized in that: The system further comprises: An equipment lifecycle management module, connected to the data acquisition module and the cloud-edge collaboration module, is used to evaluate the health status of the equipment based on the preprocessed data and generate at least one maintenance management strategy based on the health status to improve the operating efficiency and lifespan of the equipment; The cloud-edge collaboration module is also used to simulate the performance changes of the equipment under multiple maintenance management strategies through the digital twin model to select the optimal maintenance management strategy.

Images