Petroleum collection simulation system based on digital twin technology and processing method thereof

By employing multi-dimensional dynamic data acquisition, data preprocessing, machine learning modeling, and adaptive updating of twin models, the problem of increasing parameter deviations in existing simulation systems has been solved, achieving accuracy and real-time performance in oil extraction simulation systems and improving extraction efficiency and safety.

CN120974934AActive Publication Date: 2025-11-18NANTONG CHENGDA METAL EQUIP MFG CO LTD
View PDF 8 Cites 0 Cited by

Patent Information

Application Number
CN202511483271.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-10-17
Publication Date
2025-11-18
Estimated Expiration
2045-10-17

AI Technical Summary

Technical Problem

Existing oil extraction simulation systems are difficult to adjust in real time with actual data after the initial data is imported, resulting in increased parameter deviations, simulation results that do not match reality, and an inability to provide accurate decision-making basis.

Method used

A multi-dimensional dynamic data acquisition module is used to collect formation and equipment data in real time. Combined with a data preprocessing and change pattern modeling module, preprocessing and machine learning modeling are performed. A twin model parameter adaptive update module is used for real-time comparison and hierarchical incremental update. A simulation result feedback optimization module is used for deviation analysis and optimization.

Benefits of technology

It achieves accurate matching between the twin model and the actual scenario, provides reliable suggestions for optimizing the collection scheme, and improves the safety and efficiency of oil collection.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120974934A_ABST
    Figure CN120974934A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of petroleum collection, in particular to a petroleum collection simulation system based on a digital twin technology and a processing method thereof, and the system comprises a multi-dimensional dynamic data collection module which is used for collecting stratum characteristic dynamic data and equipment state characteristic data in real time, the acquisition frequency is dynamically adjusted according to the fluctuation amplitude of the acquired data; and the data preprocessing and change rule modeling module is connected with the multi-dimensional dynamic data acquisition module, and is used for preprocessing the stratum characteristic dynamic data and the equipment state characteristic data acquired by the multi-dimensional dynamic data acquisition module, and performing fusion modeling based on an oil exploitation physical mechanism and machine learning. Through cooperative work of multiple modules, the problems that an existing petroleum acquisition simulation system only initially imports data, follow-up twinborn model parameters are difficult to adjust along with real-time change of actual data, deviation is increased, and a simulation result does not conform to reality are effectively solved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of oil extraction technology, specifically to an oil extraction simulation system and processing method based on digital twin technology. Background Technology

[0002] As a vital energy resource, oil extraction involves multiple complex stages, including wellhead equipment operation, formation fluid flow, and pipeline transportation. Due to the harsh environment and variable operating conditions at oil extraction sites, directly optimizing extraction plans through on-site testing is not only costly but also carries potential safety risks. Therefore, the industry has begun to introduce digital twin technology to build oil extraction simulation systems. This technology collects various data from actual oil extraction scenarios, such as wellhead pressure, fluid flow rate, and equipment temperature, and constructs a highly accurate twin model in virtual space. This model simulates the extraction process under different operating conditions, providing support for personnel to optimize extraction parameters and predict equipment failures, thereby reducing on-site testing costs and improving the safety and efficiency of the extraction process. Currently, during oil extraction, formation characteristics gradually change over time. For example, formation porosity may subtly alter due to fluid extraction, and the operating parameters of wellhead equipment may fluctuate dynamically due to wear and aging. These changes directly affect the actual state of the extraction process. However, most existing simulation systems only import actual data in the initial modeling stage. Subsequent parameter updates to the twin model are difficult to automatically adjust based on the real-time changes in the actual extracted data. This leads to a gradual increase in the deviation between the parameters in the twin model and the parameters in the actual extraction scenario. Consequently, the simulation results output by the system do not match the actual situation, failing to provide accurate decision-making basis for personnel. This can result in unreasonable adjustments to the extraction plan, affecting oil extraction efficiency and even causing equipment malfunctions. Summary of the Invention

[0003] To address the shortcomings of existing technologies, this invention provides an oil extraction simulation system and its processing method based on digital twin technology. This solves the problems in existing technologies where the twin model only imports initial data and subsequent parameters are difficult to adjust in real time with actual data, leading to increased parameter deviations, simulation results that do not match reality, and the inability to provide accurate decision-making basis.

[0004] To achieve the above objectives, the present invention provides the following technical solution: an oil extraction simulation system based on digital twin technology, comprising: A multi-dimensional dynamic data acquisition module is used to acquire dynamic data on formation characteristics and equipment status features in real time, and dynamically adjust the acquisition frequency according to the fluctuation range of the acquired data. The data preprocessing and change pattern modeling module is connected to the multi-dimensional dynamic data acquisition module. It is used to preprocess the dynamic data of formation characteristics and the data of equipment status characteristics acquired by the multi-dimensional dynamic data acquisition module, and to dynamically train the formation characteristic change model and the equipment parameter fluctuation model based on the fusion model of oil extraction physical mechanism and machine learning, so as to predict the future short-term change trend of the dynamic data of formation characteristics and the data of equipment status characteristics. The twin model parameter adaptive update module, connected to the data preprocessing and change pattern modeling module, is used to prioritize twin model parameters according to the influence weight of parameters on the acquisition process, and adopt a hierarchical incremental update mechanism according to the parameter priority to compare the deviation between the real-time data acquired by the multi-dimensional dynamic data acquisition module and the current parameters of the twin model in real time. When the deviation exceeds the preset accuracy threshold, the parameter correction amount is calculated based on the constructed formation characteristic change model and the equipment parameter fluctuation model, and the corresponding formation characteristic parameters and equipment status parameters in the twin model are incrementally updated. The simulation result feedback optimization module is connected to the twin model parameter adaptive update module and the multi-dimensional dynamic data acquisition module. It is used to build a deviation analysis engine, compare the simulated output of the twin model with the actual field data collected in the same period, decompose and attribute the deviation, and based on the deviation attribution results, reverse-optimize the acquisition frequency of the multi-dimensional dynamic data acquisition module or the model parameters of the data preprocessing and change law modeling module, and provide parameter sensitivity analysis function.

[0005] Furthermore, the multi-dimensional dynamic data acquisition module includes: Formation sensing unit is used to deploy in oil well formations to collect dynamic data of at least one formation characteristic, including formation porosity, permeability, and formation pressure gradient, in real time. Microelectromechanical sensing modules are used to be deployed on key components of wellhead equipment to collect in real time at least one of the equipment status characteristics data, such as equipment vibration frequency, component temperature fluctuation, and sealing performance degradation. An adaptive acquisition controller is also provided, which is connected to the formation sensing unit and the microelectromechanical sensing module. It is used to dynamically adjust the acquisition frequency of the formation sensing unit and the microelectromechanical sensing module according to the fluctuation range of the formation characteristic dynamic data or the equipment status characteristic data. When the parameter fluctuation is less than a preset stability threshold, the acquisition frequency is reduced. When the parameter fluctuation exceeds the preset stability threshold, the acquisition frequency is automatically increased.

[0006] Furthermore, the data preprocessing and change pattern modeling module includes: The multi-source data preprocessing unit is used to remove interference values ​​from the dynamic data of formation characteristics and the data of equipment status characteristics through an outlier identification algorithm, and to fill occasional acquisition gaps through a data completion algorithm. The system also includes a dynamic law modeling unit connected to the multi-source data preprocessing unit. This unit takes the preprocessed dynamic data of formation characteristics and equipment status characteristics as input, constructs a basic law framework based on the physical mechanism of oil extraction, and introduces a time-series prediction algorithm to dynamically train the formation characteristic change model and the equipment parameter fluctuation model.

[0007] Furthermore, the formation characteristic change model trained by the dynamic law modeling unit is a formation porosity change curve model with mining volume, and the equipment parameter fluctuation model is an equipment vibration frequency decay model with operating time.

[0008] Furthermore, the twin model parameter adaptive update module divides the twin model parameters into core influence layer parameters and auxiliary influence layer parameters; The twin model parameter adaptive update module includes: The core parameter update unit is used to adopt a real-time comparison and instant update mode. When the deviation between the collected real-time data and the current core influence layer parameters of the twin model exceeds a preset accuracy threshold, the parameter correction amount is calculated based on the formation characteristic change model and the equipment parameter fluctuation model, and the corresponding core influence layer parameters in the twin model are incrementally updated. The auxiliary parameter update unit is used to perform batch fine-tuning of the corresponding auxiliary influence layer parameters in the twin model by summarizing the deviation of the collected data according to a preset periodic verification batch optimization mode, and combining the long-term trend prediction results of the formation characteristic change model and the equipment parameter fluctuation model.

[0009] Furthermore, the twin model parameter adaptive update module also includes: The validity verification unit is connected to the core parameter update unit and the auxiliary parameter update unit. After each parameter update, it compares the updated twin model simulation output with the actual field data collected at the same time through virtual reality comparison verification. If the deviation is less than a set threshold, the update is confirmed to be valid.

[0010] Furthermore, when the update validity verification unit determines that the deviation exceeds the standard, a backtracking mechanism is automatically triggered, and the parameters of the corresponding change pattern model in the data preprocessing and change pattern modeling module are recalibrated by the twin model parameter adaptive update module.

[0011] Furthermore, the simulation result feedback optimization module includes: The deviation analysis engine is used to break down the deviation between the simulated output of the twin model and the actual collected data, locate the source of the deviation as insufficient collection frequency of the multi-dimensional dynamic data acquisition module or poor adaptability of the data preprocessing and change law modeling module to specific working conditions, and generate a deviation attribution report. And a closed-loop optimization strategy unit, connected to the deviation analysis engine, used to adjust the acquisition frequency of the corresponding parameters in the multi-dimensional dynamic data acquisition module, or update the training samples of the machine learning algorithm in the data preprocessing and change law modeling module, or adjust the correlation coefficient of the physical mechanism model based on the deviation attribution results.

[0012] Furthermore, the simulation result feedback optimization module also includes: The decision support enhancement unit, connected to the closed-loop optimization strategy unit, is used to simulate the impact of different acquisition parameters on acquisition efficiency or equipment lifespan based on the updated twin model, and output quantitative optimization suggestions.

[0013] This invention also provides an oil extraction simulation method based on digital twin technology, applied to the oil extraction simulation system based on digital twin technology described in any one of the above, comprising: Real-time acquisition of dynamic data on formation characteristics and equipment status features, and dynamic adjustment of the acquisition frequency based on the fluctuation range of the acquired data; The collected dynamic data of formation characteristics and the data of equipment status characteristics are preprocessed, and the formation characteristic change model and the equipment parameter fluctuation model are dynamically trained based on the fusion model of oil extraction physical mechanism and machine learning, so as to predict the future short-term change trend of the dynamic data of formation characteristics and the data of equipment status characteristics. The twin model parameters are prioritized according to the weight of their influence on the acquisition process. A hierarchical incremental update mechanism is adopted based on the parameter priorities to compare the deviation between the acquired real-time data and the current parameters of the twin model in real time. When the deviation exceeds the preset accuracy threshold, the parameter correction amount is calculated based on the constructed formation characteristic change model and the equipment parameter fluctuation model, and the corresponding formation characteristic parameters and equipment status parameters in the twin model are incrementally updated. A deviation analysis engine is constructed to compare the simulated output of the twin model with the actual field data collected during the same period, and to decompose and attribute the deviation. Based on the bias attribution results, the data acquisition frequency or model parameters are optimized in reverse, and parameter sensitivity analysis is provided.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention effectively solves the problems of existing oil extraction simulation systems, such as only initially importing data, difficulty in adjusting twin model parameters in real time to adapt to actual data changes, leading to increased deviations and inconsistencies between simulation results and reality. Specifically, the multi-dimensional dynamic data acquisition module dynamically adjusts the acquisition frequency based on data fluctuations, ensuring dense and accurate data acquisition during periods of high volatility while avoiding redundant acquisition during stable periods. The data preprocessing and variation pattern modeling module removes outliers, fills in data gaps, and integrates oil extraction physics mechanisms with machine learning modeling to predict short-term data trends. The twin model parameter adaptive update module updates parameters hierarchically according to their influence weights, with real-time comparison and updates of core parameters, periodic batch fine-tuning of auxiliary parameters, and model calibration through validity verification and backtracking mechanisms to ensure parameters closely match reality. The simulation result feedback optimization module decomposes deviation attribution and back-optimizes acquisition frequency and model parameters, providing parameter sensitivity analysis and quantitative decision-making suggestions. Ultimately, this ensures the twin model accurately reflects the actual scenario, reduces simulation deviations, provides a reliable basis for optimizing acquisition schemes, and improves the safety and efficiency of oil extraction. Attached Figure Description

[0015] Figure 1 This is a system structure diagram of the present invention; Figure 2 This is a flowchart of the method of the present invention. Detailed Implementation

[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] Please see Figure 1 This invention provides an oil extraction simulation system based on digital twin technology, comprising: The multi-dimensional dynamic data acquisition module is used to collect dynamic data on formation characteristics and equipment status features in real time, and dynamically adjusts the acquisition frequency according to the fluctuation range of the collected data. The data preprocessing and change pattern modeling module is connected to the multi-dimensional dynamic data acquisition module. It is used to preprocess the dynamic data of formation characteristics and equipment status characteristics acquired by the multi-dimensional dynamic data acquisition module, and to dynamically train the formation characteristic change model and equipment parameter fluctuation model based on the fusion model of oil extraction physical mechanism and machine learning, so as to predict the future short-term change trend of formation characteristic dynamic data and equipment status characteristics data. The twin model parameter adaptive update module, connected to the data preprocessing and change pattern modeling module, is used to prioritize twin model parameters according to the weight of their influence on the acquisition process. Based on the parameter priority, a hierarchical incremental update mechanism is adopted to compare the deviation between the real-time data acquired by the multi-dimensional dynamic data acquisition module and the current parameters of the twin model in real time. When the deviation exceeds the preset accuracy threshold, the parameter correction amount is calculated based on the constructed formation characteristic change model and equipment parameter fluctuation model, and the corresponding formation characteristic parameters and equipment status parameters in the twin model are incrementally updated. The simulation result feedback optimization module, connected to the twin model parameter adaptive update module and the multi-dimensional dynamic data acquisition module, is used to build a deviation analysis engine. It compares the simulated output of the twin model with the actual field data collected during the same period, decomposes and attributes the deviation, and based on the deviation attribution results, it back-optimizes the acquisition frequency of the multi-dimensional dynamic data acquisition module or the model parameters of the data preprocessing and change law modeling module, and provides parameter sensitivity analysis function.

[0018] Specifically, this implementation addresses the problem in the background technology that "existing simulation systems only import initial data, and subsequent parameter updates are difficult to adjust to real-time changes, leading to increased deviations." It achieves dynamic adaptation and accurate simulation through the collaborative operation of four modules, as detailed below: When deploying the multi-dimensional dynamic data acquisition module, sensors are first placed at different formation depths in the oil well, such as the middle and bottom of the oil layer, to collect real-time dynamic data on formation characteristics, such as formation porosity and permeability. Simultaneously, sensor components are installed on key components such as the wellhead pumping unit crank and wellhead valves to collect equipment status characteristic data, such as equipment vibration frequency and component temperature. This module dynamically adjusts the acquisition frequency using a preset fluctuation judgment formula: Let the real-time fluctuation amplitude of a certain parameter be... , ,in The parameter value at the current time. The parameter value is from the previous acquisition time; the preset stability threshold is... .when At that time, the sampling frequency was reduced from the initial 1 time / 2 minutes to 1 time / 5 minutes; when At that time, the sampling frequency is automatically increased to once every 30 seconds to ensure more intensive data collection when fluctuations are severe, providing an accurate data foundation for subsequent modeling.

[0019] The data preprocessing and variation pattern modeling module first preprocesses the collected data: it uses a 3σ outlier detection algorithm to remove interfering values, the formula being... ,in For a single data entry, This is the mean of the dataset containing this parameter. The standard deviation is used; data meeting this condition are considered outliers and removed. For occasional data gaps, linear interpolation is used to fill in the gaps, with the formula being: ,in For vacant moments Data, , They are respectively Adjacent acquisition times , The data is preprocessed, and a basic framework is constructed based on the physical mechanisms of oil extraction, such as Darcy's law. Simultaneously, an LSTM time-series prediction algorithm is introduced for fusion modeling, training a formation characteristic variation model and a device parameter fluctuation model. Taking the formation characteristic variation model (formation porosity variation model) as an example, the formula is... ,in Let be the formation porosity at time t. Initial porosity, This is the porosity attenuation coefficient, expressed in % / m³. The oil extraction rate at time τ is expressed in m³ / h. The cumulative mining volume from time 0 to time t is expressed in m³. The product of the cumulative mining volume and the cumulative mining volume is expressed in % units. The dimensions are consistent with %, so normalization is not required. This model can predict the trend of formation porosity changes over the next 12 hours, providing a basis for parameter updates.

[0020] The twin model parameter adaptive update module first prioritizes parameters: based on the analytic hierarchy process (AHP), it calculates the weight of each parameter's influence on the acquisition process. For example, formation pressure and wellhead flow rate, with a weight exceeding 0.6, are classified as core influence parameters; ambient temperature and pipeline humidity, with a weight below 0.3, are classified as auxiliary influence parameters. A hierarchical incremental update mechanism is then employed: for core influence parameters, the deviation is calculated in real time. ,in To collect data in real time, The current parameter values ​​of the twin model; preset accuracy threshold. ,when At that time, the correction amount is calculated based on the formation characteristic change model and the equipment parameter fluctuation model. , For the reasonable values ​​of the parameters predicted by the model, the core parameters are updated incrementally in real time. For the auxiliary influence layer parameters, the deviations are summarized on a 1-hour cycle, and combined with the model's long-term trend prediction results, such as the environmental temperature change trend in the next 24 hours, batch fine-tuning is performed to avoid frequent updates that consume system resources.

[0021] The simulation results feedback optimization module constructs a deviation analysis engine: it compares the mining volume and equipment energy consumption data simulated by the twin model with the actual data collected during the same period, and uses the deviation decomposition formula... Disassembly deviation, among which For the total deviation, Due to data collection bias, To model biases in the model, we can use this to pinpoint the sources of those biases. For example, if... If the percentage exceeds 60%, it is determined that the collection frequency is insufficient; if If the percentage exceeds 70%, the model is deemed to have poor adaptability to high sand content conditions. Backward optimization based on attribution results: When the data collection frequency is insufficient, the stability threshold of the corresponding parameters is adjusted. The threshold was lowered from 0.5% to 0.3%, triggering a higher frequency of data collection. When model adaptability was poor, training samples under high sand content conditions were added, such as 200 sets of historical data with sand content exceeding 15%. A parameter sensitivity analysis function was also provided, which calculates the rate of change in simulation results when parameters change by 1%, thus identifying key optimization directions.

[0022] This implementation method uses dynamic data acquisition, fusion modeling, hierarchical updates, and closed-loop optimization to ensure that the parameters of the twin model remain consistent with the actual scene, significantly reducing the deviation between simulation results and reality, and providing accurate support for the optimization of the data acquisition scheme.

[0023] In this embodiment, the multi-dimensional dynamic data acquisition module includes: Formation sensing unit is used to deploy in oil well formations to collect dynamic data of at least one formation characteristic, including formation porosity, permeability, and formation pressure gradient, in real time. Microelectromechanical sensing modules are used to be deployed on key components of wellhead equipment to collect in real time at least one of the equipment status characteristics data, such as equipment vibration frequency, component temperature fluctuation, and sealing performance degradation. An adaptive acquisition controller is also included, which is connected to the formation sensing unit and the microelectromechanical sensing module. It is used to dynamically adjust the acquisition frequency of the formation sensing unit and the microelectromechanical sensing module according to the fluctuation range of dynamic data of formation characteristics or equipment status characteristic data. When the parameter fluctuation is less than the preset stability threshold, the acquisition frequency is reduced, and when the parameter fluctuation exceeds the preset stability threshold, the acquisition frequency is automatically increased.

[0024] Specifically, the formation sensing unit uses distributed fiber optic sensors deployed along the outer wall of the well casing, with a deployment range extending, for example, from the wellhead down to 10 meters below the bottom of the oil layer, to collect formation porosity, permeability, and formation pressure gradient data in real time. Formation porosity is calculated by converting the formation acoustic wave propagation velocity detected by the sensors, and permeability is calculated based on the correlation between porosity and formation fluid viscosity.

[0025] The microelectromechanical sensing module uses MEMS vibration sensors and temperature sensors, which are installed on key components such as the crank pin, gearbox, and sealing surface of the wellhead pumping unit. For example, a vibration sensor is installed at the crank pin with a sampling rate of 1000Hz to collect the vibration frequency of the equipment in real time; a temperature sensor is installed at the sealing surface with an accuracy of ±0.1℃ to collect the temperature fluctuation of the component in real time. At the same time, a pressure sensor monitors the attenuation of sealing performance, and the degree of attenuation of sealing performance is reflected by the difference between the leakage pressure of the medium at the seal and the initial sealing pressure.

[0026] The adaptive acquisition controller has a built-in data processing chip that first performs real-time calculations on the acquired dynamic data of formation characteristics and equipment status features to obtain the fluctuation range of each parameter. (Parameter definition is the same as in Implementation 1). Preset stability threshold. For example, the formation porosity stability threshold is set to 0.5%, and the equipment vibration frequency stability threshold is set to 2Hz. When the fluctuation amplitude of a certain parameter... When the formation porosity fluctuation is 0.3%, the controller sends a command to the corresponding sensor to reduce the acquisition frequency; for example, when the formation porosity fluctuation is 0.3%, the acquisition frequency of the formation sensing unit is reduced from 1 time / minute to 1 time / 5 minutes. When the vibration frequency of the device fluctuates at 3Hz, the controller automatically increases the acquisition frequency of the vibration data from the microelectromechanical sensing module from 1 time / 30 seconds to 1 time / 5 seconds to ensure that critical fluctuation data is not lost, while avoiding redundant data acquisition in a stable state and reducing data transmission and storage costs.

[0027] In this embodiment, the data preprocessing and change pattern modeling module includes: The multi-source data preprocessing unit is used to remove interference values ​​from dynamic data of formation characteristics and equipment status characteristic data through outlier identification algorithms, and to fill occasional acquisition gaps through data completion algorithms. It also includes a dynamic law modeling unit, which is connected to a multi-source data preprocessing unit. This unit uses preprocessed dynamic data of formation characteristics and equipment status characteristics as input to construct a basic law framework based on the physical mechanism of oil extraction. It also introduces a time-series prediction algorithm to dynamically train the formation characteristic change model and the equipment parameter fluctuation model.

[0028] Specifically, the multi-source data preprocessing unit first identifies outliers: for dynamic formation characteristic data such as formation pressure and equipment status characteristic data such as equipment temperature, a 3σ algorithm is used to remove interfering values ​​(formula same as implementation 1); for example, in a set of wellhead temperature data, the temperature value at a certain moment is 85℃, while the mean of the dataset is... Standard deviation Calculated The data is identified as an outlier and removed. For occasional data gaps, such as a 10-minute data loss due to a temporary power outage of the sensor, linear interpolation is used to fill the gap (formula is the same as in implementation 1). For example, if there is a 10:10 gap between 10:00 (data is 25Hz) and 10:20 (data is 28Hz) in the equipment vibration frequency range, the calculated... Complete the data completion.

[0029] The dynamic law modeling unit takes preprocessed data as input and first constructs a basic law framework based on the physical mechanism of oil extraction; for example, for formation characteristics, it constructs correlations based on Darcy's law, the formula of which is... ,in For fluid flow rate, For penetration rate, For the flow cross-sectional area, For pressure difference, For fluid viscosity, The flow length is used to determine the physical relationship between formation parameters using this law. Then, the ARIMA time-series prediction algorithm is introduced to dynamically train the formation characteristic variation model and the equipment parameter fluctuation model; for example, when training the equipment parameter fluctuation model, preprocessed equipment vibration frequency data is used as input, and the ARIMA model formula is... ,in for The d-th order difference is used to station the data. For constant terms, These are the autoregressive coefficients. The moving average coefficient is... The error is white noise. This model can predict the changes in equipment vibration frequency over the next 6 hours, providing a predictive basis for updating twin model parameters and improving the foresight of parameter updates.

[0030] In this embodiment, the formation characteristic change model trained by the dynamic law modeling unit is a formation porosity change curve model with mining volume, and the equipment parameter fluctuation model is a model of equipment vibration frequency decay with operating time.

[0031] Specifically, when training the formation characteristic variation model (formation porosity variation curve model with extraction volume) using the dynamic law modeling unit, preprocessed formation porosity data and corresponding cumulative extraction volume data are used as samples. An exponential decay model is employed, with the following formula: ,in Formation porosity when the cumulative extraction volume is Q, expressed as a percentage. Formation porosity when the initial cumulative extraction volume is 0, expressed as a percentage. This is the porosity attenuation coefficient, with units of m⁻³. This represents the cumulative mining volume, expressed in m³. In this formula... Since it is a dimensionless quantity, the calculation method is m⁻³×m³=1, which is consistent with the requirements of exponential functions and does not require normalization; for example, the initial porosity of an oil well. attenuation coefficient ⁻³ When the cumulative mining volume is ³ porosity It accurately reflects the decreasing trend of porosity as the mining volume decreases.

[0032] When training the equipment parameter fluctuation model (the model of equipment vibration frequency decaying with operating time), preprocessed equipment vibration frequency data and cumulative equipment operating time data are used as samples, and a linear decay model is constructed, with the following formula: ,in The vibration frequency of the equipment over a cumulative operating time of T, expressed in Hz. The vibration frequency when the initial running time is 0, in Hz. This is the vibration frequency attenuation coefficient, with units of Hz / h. This represents the cumulative operating time of the device, measured in hours (h). In the formula... The dimension is Hz, and the calculation method is Hz / h × h = Hz. The dimensions are consistent with Hz; for example, the initial vibration frequency of a wellhead valve. attenuation coefficient When the cumulative runtime At that time, the vibration frequency It accurately reflects the decay law of vibration frequency with the duration of operation, providing support for the prediction of equipment status.

[0033] In this embodiment, the twin model parameter adaptive update module divides the twin model parameters into core influence layer parameters and auxiliary influence layer parameters; The twin model parameter adaptive update module includes: The core parameter update unit is used to adopt a real-time comparison and instant update mode. When the deviation between the collected real-time data and the current core influence layer parameters of the twin model exceeds the preset accuracy threshold, the parameter correction amount is calculated based on the formation characteristic change model and the equipment parameter fluctuation model, and the corresponding core influence layer parameters in the twin model are incrementally updated. The auxiliary parameter update unit is used to perform batch fine-tuning of the corresponding auxiliary influence layer parameters in the twin model by summarizing the deviation of the collected data according to a preset periodic verification batch optimization mode, and combining the long-term trend prediction results of the formation characteristic change model and the equipment parameter fluctuation model.

[0034] Specifically, the twin model parameter adaptive update module first prioritizes parameters using the analytic hierarchy process (AHP): Five experts in the fields of oil extraction and digital twins are invited to score the influence weight of each parameter, such as formation pressure, permeability, equipment temperature, and ambient humidity. Based on the scoring results, the weight values ​​are calculated, and parameters with a weight ≥ 0.5 are classified as core influence layer parameters, such as formation pressure and wellhead flow rate, while parameters with a weight < 0.5 are classified as auxiliary influence layer parameters, such as ambient humidity and pipeline outer wall temperature.

[0035] The core parameter update unit adopts a real-time comparison and instant update mode: it collects actual data of core influence layer parameters in real time. The current core parameter values ​​of the twin model Calculate the deviation (Equivalent to the first implementation of the formula), preset accuracy threshold .when At that time, the parameter correction amount is calculated based on the formation characteristic change model and the equipment parameter fluctuation model. ,in This refers to the reasonable value of the parameter predicted by the model; for example, the actual value of formation pressure. Current value of the model ,deviation Predicting reasonable pressure values ​​using a formation property variation model Then the correction amount Incremental updates are performed on the formation pressure parameters in the twin model to ensure that the core parameters are accurate in real time.

[0036] The auxiliary parameter update unit adopts a periodic verification batch optimization mode: the preset update cycle is 2 hours, and the actual collected data of the auxiliary influence layer parameters are summarized every 2 hours to calculate the average deviation. ,in This refers to the amount of data collected within the period. This refers to the deviation of a single data point. Combined with the long-term trend prediction results of the equipment parameter fluctuation model, such as the environmental humidity change trend over the next 12 hours, auxiliary parameters are fine-tuned in batches; for example, the average deviation of environmental humidity. If the model predicts that humidity will rise slowly in the future, the environmental humidity parameter in the twin model will be adjusted upward by 0.3% in batches. This ensures the accuracy of the auxiliary parameters while avoiding frequent updates that consume system computing power.

[0037] In this embodiment, the twin model parameter adaptive update module further includes: The update validity verification unit is connected to the core parameter update unit and the auxiliary parameter update unit. After each parameter update, it is used to compare the updated twin model simulation output with the actual field data collected at the same time through virtual reality comparison verification. If the deviation is less than the set threshold, the update is confirmed to be valid.

[0038] Specifically, the validity verification unit is connected to the core parameter update unit and the auxiliary parameter update unit through a data interface. After each parameter update is completed, the virtual-actual comparison and verification process is triggered immediately. The parameter update includes real-time update of core parameters and batch update of auxiliary parameters.

[0039] During verification, first obtain the simulated data output by the updated twin model, such as simulated mining volume and simulated equipment temperature, and then retrieve the actual field data collected during the same period (within 10 minutes after the update). Calculate the average deviation between the two. ,in To verify the amount of data, For the j-th simulated data, This is the j-th actual data point. A preset threshold is set. For example, the set threshold for simulated mining volume is 0.5 m³ / h, and the set threshold for simulated equipment temperature is 0.3℃. If the calculated... If the update is successful, a verification report indicating "update valid" will be generated and stored in the system database. For example, if the average deviation of the simulated mining volume after the update is 0.3 m³ / h < 0.5 m³ / h, the update is deemed valid, ensuring that the updated twin model can accurately reflect the actual scenario and provide a reliable foundation for subsequent simulations.

[0040] In this embodiment, when the error of the update validity verification unit exceeds the standard, the backtracking mechanism is automatically triggered, and the parameters of the corresponding change pattern model in the data preprocessing and change pattern modeling module are recalibrated by the twin model parameter adaptive update module.

[0041] Specifically, when updating the average deviation calculated by the validity verification unit In other words, when the deviation exceeds the standard, a "deviation exceeds the standard" signal is automatically sent to the twin model parameter adaptive update module to trigger the backtracking mechanism.

[0042] After the backtracking mechanism is activated, the twin model parameter adaptive update module first retrieves the basic data on which this parameter update is based. This basic data includes the collected real-time data, the output data of the formation characteristic change model and the equipment parameter fluctuation model. By tracing the data source, the root cause of the deviation is located: for example, if it is found that the predicted value output by the model deviates greatly from the actual data, it is determined that the change law model parameters in the data preprocessing and change law modeling module are inaccurate.

[0043] Subsequently, the twin model parameter adaptive update module sends a parameter calibration command to the data preprocessing and variation pattern modeling module to recalibrate the parameters of the corresponding variation pattern model. For example, for the stratigraphic characteristic variation model, the formula is: If the porosity predicted by the model deviates significantly from the actual porosity, adjust the attenuation coefficient. The value was adjusted from the initial 0.002 m⁻³ to 0.0018 m⁻³; for the equipment parameter fluctuation model, the formula is as follows: If the predicted vibration frequency deviates significantly, adjust the attenuation coefficient. The value was adjusted from 0.02 Hz / h to 0.019 Hz / h. After calibration, the parameter update and validity verification process was repeated until the deviation met the requirements, forming a closed loop of "update-verification-backtracking-calibration" to ensure that the model always has high accuracy.

[0044] In this embodiment, the simulation result feedback optimization module includes: The deviation analysis engine is used to break down the deviation between the simulated output of the twin model and the actual collected data, locate the source of the deviation as insufficient acquisition frequency of the multi-dimensional dynamic data acquisition module or poor adaptability of the data preprocessing and change law modeling module to specific working conditions, and generate a deviation attribution report. It also includes a closed-loop optimization strategy unit, which is connected to the deviation analysis engine. This unit is used to adjust the acquisition frequency of corresponding parameters in the multi-dimensional dynamic data acquisition module based on the deviation attribution results, or to update the training samples of the machine learning algorithm in the data preprocessing and change pattern modeling module, or to adjust the correlation coefficient of the physical mechanism model.

[0045] Specifically, the deviation analysis engine uses a data comparison algorithm to compare the simulated data output by the twin model, such as simulated formation pressure and simulated equipment energy consumption, with the actual field data collected during the same period, time by time, and calculates the total deviation. (Formula equivalent to implementation 1), and the source of deviation is located through the deviation decomposition model: if data acquisition deviation The proportion exceeds 50%, of which The delayed data acquisition was determined to be due to insufficient acquisition frequency from the multi-dimensional dynamic data acquisition module; if there is a modeling deviation... The proportion exceeds 60%, of which The calculated values ​​from the model indicate poor adaptability of the data preprocessing and variation pattern modeling module to specific working conditions, such as high sand content and high pressure. The deviation analysis engine generates a deviation attribution report based on the positioning results, clearly identifying the deviation type, percentage, and improvement directions.

[0046] After receiving the deviation attribution report, the closed-loop optimization strategy unit formulates a targeted optimization plan: if the attribution result indicates insufficient acquisition frequency, it adjusts the stability threshold of the corresponding parameter in the multi-dimensional dynamic data acquisition module. For example, the stability threshold of formation pressure can be lowered from 0.8 MPa to 0.5 MPa to trigger higher-frequency data collection. If the attribution result indicates poor model fit, the training samples of the machine learning algorithm in the data preprocessing and variation pattern modeling module can be updated. For example, 300 sets of historical data for high sand content conditions can be added. High sand content conditions refer to conditions with sand content > 20%. Alternatively, the correlation coefficient of the physical mechanism model can be adjusted, such as the viscosity coefficient in Darcy's law. The correction factor was adjusted from 1.0 to 1.2 to improve the model's adaptability to special operating conditions. Through targeted optimization, simulation deviations were continuously reduced, and system reliability was improved.

[0047] In this embodiment, the simulation result feedback optimization module further includes: The decision support enhancement unit, connected to the closed-loop optimization strategy unit, is used to simulate the impact of different acquisition parameters on acquisition efficiency or equipment lifespan based on the updated twin model, and output quantitative optimization suggestions.

[0048] Specifically, the decision support enhancement unit and the closed-loop optimization strategy unit are connected through a data bus. After the closed-loop optimization strategy unit completes the optimization of the data acquisition frequency or model parameters, it obtains the updated twin model data.

[0049] Based on the updated twin model, this unit constructs a multi-scenario simulation environment to simulate the impact of different acquisition parameters on acquisition efficiency or equipment lifespan. Different acquisition parameters include wellhead flow rate and production pressure. For example, it simulates the production efficiency when the wellhead flow rate is 15 m³ / h, 20 m³ / h, and 25 m³ / h, where production efficiency refers to the ratio of production volume to energy consumption, as well as the lifespan wear of key equipment components, such as the pumping unit crank. It also simulates the formation fluid flow state and pipeline wear degree when the production pressure is 12 MPa, 15 MPa, and 18 MPa.

[0050] Subsequently, the unit uses quantitative analysis algorithms to calculate quantitative indicators such as the improvement rate of acquisition efficiency and the extension of equipment life under various scenarios, and generates optimization suggestion reports. For example, the report clearly states that "when the wellhead flow rate is adjusted to 20 m³ / h, the acquisition efficiency will be improved compared to the current level, and the rate of equipment life loss will be reduced." This provides quantitative basis for staff to formulate acquisition plans, helps staff to find the optimal balance between acquisition efficiency and equipment protection, and improves the overall benefits of oil acquisition.

[0051] Please see Figure 2 The present invention also provides an oil extraction simulation method based on digital twin technology, applicable to any of the above-mentioned oil extraction simulation systems based on digital twin technology, comprising: Real-time acquisition of dynamic data on formation characteristics and equipment status features, and dynamic adjustment of acquisition frequency based on the fluctuation range of the acquired data; The collected dynamic data of formation characteristics and equipment status characteristics are preprocessed, and the formation characteristic change model and equipment parameter fluctuation model are dynamically trained based on the fusion model of oil extraction physical mechanism and machine learning, so as to predict the future short-term change trend of formation characteristic dynamic data and equipment status characteristics. Based on the weight of the influence of parameters on the acquisition process, the parameters of the twin model are prioritized, and a hierarchical incremental update mechanism is adopted according to the parameter priority. The deviation between the acquired real-time data and the current parameters of the twin model is compared in real time. When the deviation exceeds the preset accuracy threshold, the parameter correction amount is calculated based on the constructed formation characteristic change model and equipment parameter fluctuation model, and the corresponding formation characteristic parameters and equipment status parameters in the twin model are incrementally updated. A deviation analysis engine is built to compare the simulated output of the twin model with the actual field data collected during the same period, and to break down and attribute the deviation. Based on the bias attribution results, the data acquisition frequency or model parameters are optimized in reverse, and parameter sensitivity analysis is provided.

[0052] Specifically, Step 1: Real-time data acquisition and dynamic frequency adjustment. Distributed fiber optic sensors are deployed in the oil well formation to collect dynamic data on formation characteristics such as porosity and permeability; MEMS sensors are deployed on key components of the wellhead equipment, such as the pumping unit gearbox and wellhead valves, to collect equipment status characteristic data such as equipment vibration frequency and component temperature fluctuations. The amplitude of parameter fluctuations is calculated through an adaptive acquisition controller. (Parameter definition is the same as in Implementation 1), preset stable threshold ,when Reduce the sampling frequency when Increase the frequency of data collection to ensure both accuracy and cost-effectiveness.

[0053] Step 2: Data Preprocessing and Pattern Modeling. The 3σ algorithm is used to remove outliers from the collected data (e.g., implementation method 1, with the same weighting). Linear interpolation is used to fill in any gaps in the data collection (e.g., implementation method 1, with the same weighting). Using the preprocessed data as input, a basic framework is constructed based on the physical mechanisms of oil extraction, such as Darcy's law. LSTM or ARIMA time-series prediction algorithms are introduced to train formation characteristic variation models and equipment parameter fluctuation models. Formation characteristic variation models include models of porosity variation with extraction volume, and equipment parameter fluctuation models include models of vibration frequency decay with operating time. This enables the prediction of short-term future trends in the data.

[0054] Step 3: Hierarchical incremental update of twin model parameters. Parameter priorities are determined using the analytic hierarchy process (AHP), dividing the model into a core influence layer and a secondary influence layer. Real-time deviations are calculated for the parameters in the core influence layer. ,when When the accuracy threshold is exceeded, the correction amount is calculated based on the model constructed in step 2, and the core parameters are updated in real time; the deviations of the auxiliary influence layer parameters are summarized according to the preset period, and the auxiliary parameters are fine-tuned in batches in combination with the long-term trend prediction results of the model to ensure that the model parameters are consistent with the actual scenario.

[0055] Step 4: Deviation Analysis and Attribution. A deviation analysis engine is built to compare the simulated output of the twin model with the actual field data collected during the same period. Simulated outputs include, for example, simulated mining volume and simulated equipment temperature. The total deviation is calculated and broken down into data acquisition deviation and modeling deviation. The sources of deviation are identified, such as insufficient acquisition frequency or poor model adaptability, and a deviation attribution report is generated.

[0056] Step 5: Reverse Optimization and Sensitivity Analysis. Based on the bias attribution results, adjust the data acquisition frequency (e.g., lower the stability threshold) or model parameters (e.g., supplement training samples, adjust the physical mechanism model coefficients); simultaneously calculate the rate of change of simulation results when each parameter changes by 1%, conduct parameter sensitivity analysis, identify key optimization directions, and provide support for optimizing the acquisition scheme.

[0057] This method solves the problems of lagging parameter updates and large deviations in existing simulation methods through a closed-loop process of "collection-modeling-update-optimization", thereby improving the accuracy of simulation results.

[0058] In summary, this invention effectively solves the problems of existing oil extraction simulation systems, such as only initially importing data, difficulty in adjusting twin model parameters in real time to adapt to actual data changes, leading to increased deviations and inconsistencies between simulation results and reality. Specifically, the multi-dimensional dynamic data acquisition module dynamically adjusts the acquisition frequency based on data fluctuations, ensuring dense and accurate data acquisition during periods of high volatility while avoiding redundant acquisition during stable periods. The data preprocessing and variation pattern modeling module removes outliers, fills in data gaps, and integrates oil extraction physics mechanisms with machine learning modeling to predict short-term data trends. The twin model parameter adaptive update module updates parameters hierarchically according to their influence weights, with real-time comparison and updates of core parameters, periodic batch fine-tuning of auxiliary parameters, and model calibration through validity verification and backtracking mechanisms to ensure parameters closely match reality. The simulation result feedback optimization module decomposes deviation attribution and back-optimizes acquisition frequency and model parameters, providing parameter sensitivity analysis and quantitative decision-making suggestions. Ultimately, this ensures the twin model accurately reflects the actual scenario, reduces simulation deviations, provides a reliable basis for optimizing acquisition schemes, and improves the safety and efficiency of oil extraction.

[0059] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0060] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An oil extraction simulation system based on digital twin technology, characterized in that, include: A multi-dimensional dynamic data acquisition module is used to acquire dynamic data on formation characteristics and equipment status features in real time, and dynamically adjust the acquisition frequency according to the fluctuation range of the acquired data. The data preprocessing and change pattern modeling module is connected to the multi-dimensional dynamic data acquisition module. It is used to preprocess the dynamic data of formation characteristics and the data of equipment status characteristics acquired by the multi-dimensional dynamic data acquisition module, and to dynamically train the formation characteristic change model and the equipment parameter fluctuation model based on the fusion model of oil extraction physical mechanism and machine learning, so as to predict the future short-term change trend of the dynamic data of formation characteristics and the data of equipment status characteristics. The twin model parameter adaptive update module, connected to the data preprocessing and change pattern modeling module, is used to prioritize twin model parameters according to the influence weight of parameters on the acquisition process, and adopt a hierarchical incremental update mechanism according to the parameter priority to compare the deviation between the real-time data acquired by the multi-dimensional dynamic data acquisition module and the current parameters of the twin model in real time. When the deviation exceeds the preset accuracy threshold, the parameter correction amount is calculated based on the constructed formation characteristic change model and the equipment parameter fluctuation model, and the corresponding formation characteristic parameters and equipment status parameters in the twin model are incrementally updated. The simulation result feedback optimization module is connected to the twin model parameter adaptive update module and the multi-dimensional dynamic data acquisition module. It is used to build a deviation analysis engine, compare the simulated output of the twin model with the actual field data collected in the same period, decompose and attribute the deviation, and based on the deviation attribution results, reverse-optimize the acquisition frequency of the multi-dimensional dynamic data acquisition module or the model parameters of the data preprocessing and change law modeling module, and provide parameter sensitivity analysis function.

2. The oil extraction simulation system based on digital twin technology according to claim 1, characterized in that, The multi-dimensional dynamic data acquisition module includes: Formation sensing unit is used to deploy in oil well formations to collect dynamic data of at least one formation characteristic, including formation porosity, permeability, and formation pressure gradient, in real time. Microelectromechanical sensing modules are used to be deployed on key components of wellhead equipment to collect in real time at least one of the equipment status characteristics data, such as equipment vibration frequency, component temperature fluctuation, and sealing performance degradation. An adaptive acquisition controller is also provided, which is connected to the formation sensing unit and the microelectromechanical sensing module. It is used to dynamically adjust the acquisition frequency of the formation sensing unit and the microelectromechanical sensing module according to the fluctuation range of the formation characteristic dynamic data or the equipment status characteristic data. When the parameter fluctuation is less than a preset stability threshold, the acquisition frequency is reduced. When the parameter fluctuation exceeds the preset stability threshold, the acquisition frequency is automatically increased.

3. The oil extraction simulation system based on digital twin technology according to claim 1, characterized in that, The data preprocessing and change pattern modeling module includes: The multi-source data preprocessing unit is used to remove interference values ​​from the dynamic data of formation characteristics and the data of equipment status characteristics through an outlier identification algorithm, and to fill occasional acquisition gaps through a data completion algorithm. The system also includes a dynamic law modeling unit connected to the multi-source data preprocessing unit. This unit takes the preprocessed dynamic data of formation characteristics and equipment status characteristics as input, constructs a basic law framework based on the physical mechanism of oil extraction, and introduces a time-series prediction algorithm to dynamically train the formation characteristic change model and the equipment parameter fluctuation model.

4. The oil extraction simulation system based on digital twin technology according to claim 3, characterized in that, The formation characteristic change model trained by the dynamic law modeling unit is a formation porosity change curve model with mining volume, and the equipment parameter fluctuation model is a model of equipment vibration frequency decay with operating time.

5. The oil extraction simulation system based on digital twin technology according to claim 1, characterized in that, The twin model parameter adaptive update module divides the twin model parameters into core influence layer parameters and auxiliary influence layer parameters; The twin model parameter adaptive update module includes: The core parameter update unit is used to adopt a real-time comparison and instant update mode. When the deviation between the collected real-time data and the current core influence layer parameters of the twin model exceeds a preset accuracy threshold, the parameter correction amount is calculated based on the formation characteristic change model and the equipment parameter fluctuation model, and the corresponding core influence layer parameters in the twin model are incrementally updated. The auxiliary parameter update unit is used to perform batch fine-tuning of the corresponding auxiliary influence layer parameters in the twin model by summarizing the deviation of the collected data according to a preset periodic verification batch optimization mode, and combining the long-term trend prediction results of the formation characteristic change model and the equipment parameter fluctuation model.

6. The oil extraction simulation system based on digital twin technology according to claim 5, characterized in that, The twin model parameter adaptive update module also includes: The validity verification unit is connected to the core parameter update unit and the auxiliary parameter update unit. After each parameter update, it compares the updated twin model simulation output with the actual field data collected at the same time through virtual reality comparison verification. If the deviation is less than a set threshold, the update is confirmed to be valid.

7. The oil extraction simulation system based on digital twin technology according to claim 6, characterized in that, When the update validity verification unit determines that the deviation exceeds the standard, the backtracking mechanism is automatically triggered, and the parameters of the corresponding change pattern model in the data preprocessing and change pattern modeling module are recalibrated by the twin model parameter adaptive update module.

8. The oil extraction simulation system based on digital twin technology according to claim 1, characterized in that, The simulation result feedback optimization module includes: The deviation analysis engine is used to break down the deviation between the simulated output of the twin model and the actual collected data, locate the source of the deviation as insufficient collection frequency of the multi-dimensional dynamic data acquisition module or poor adaptability of the data preprocessing and change law modeling module to specific working conditions, and generate a deviation attribution report. And a closed-loop optimization strategy unit, connected to the deviation analysis engine, used to adjust the acquisition frequency of the corresponding parameters in the multi-dimensional dynamic data acquisition module, or update the training samples of the machine learning algorithm in the data preprocessing and change law modeling module, or adjust the correlation coefficient of the physical mechanism model based on the deviation attribution results.

9. The oil extraction simulation system based on digital twin technology according to claim 8, characterized in that, The simulation result feedback optimization module also includes: The decision support enhancement unit, connected to the closed-loop optimization strategy unit, is used to simulate the impact of different acquisition parameters on acquisition efficiency or equipment lifespan based on the updated twin model, and output quantitative optimization suggestions.

10. A method for simulating oil extraction based on digital twin technology, applied to the oil extraction simulation system based on digital twin technology as described in any one of claims 1-9, characterized in that, include: Real-time acquisition of dynamic data on formation characteristics and equipment status features, and dynamic adjustment of the acquisition frequency based on the fluctuation range of the acquired data; The collected dynamic data of formation characteristics and the data of equipment status characteristics are preprocessed, and the formation characteristic change model and the equipment parameter fluctuation model are dynamically trained based on the fusion model of oil extraction physical mechanism and machine learning, so as to predict the future short-term change trend of the dynamic data of formation characteristics and the data of equipment status characteristics. The twin model parameters are prioritized according to the weight of their influence on the acquisition process. A hierarchical incremental update mechanism is adopted based on the parameter priorities to compare the deviation between the acquired real-time data and the current parameters of the twin model in real time. When the deviation exceeds the preset accuracy threshold, the parameter correction amount is calculated based on the constructed formation characteristic change model and the equipment parameter fluctuation model, and the corresponding formation characteristic parameters and equipment status parameters in the twin model are incrementally updated. A deviation analysis engine is constructed to compare the simulated output of the twin model with the actual field data collected during the same period, and to decompose and attribute the deviation. Based on the bias attribution results, the data acquisition frequency or model parameters are optimized in reverse, and parameter sensitivity analysis is provided.

Citation Information

Patent Citations

  • Equipment fault diagnosis method, device and system based on digital twin model

    CN110442936A

  • Digital twinning model construction method and device

    CN112200493A

  • Construction method of intelligent assembly system based on digital twinning

    CN114580083A

  • Cutter machining state real-time interaction and monitoring method based on digital twinning technology

    CN118493076A

  • Method for optimizing carbon dioxide injection of heavy oil reservoir

    CN118862484A