Intelligent oil extraction safety control system and method

Abstract

The invention discloses an intelligent oil extraction safety control system and method, and relates to the field of oil extraction. The system comprises a monitoring module, a control module, an early warning module, an execution module and a remote monitoring center. The monitoring module collects parameters such as wellhead pressure, temperature, indicator diagram and working fluid level in real time; the control module removes invalid data through data cleaning, performs working condition diagnosis, fluid production capacity calculation and working fluid level inversion by adopting an improved algorithm, and judges abnormality in combination with a parameter threshold database; the early warning module gives an alarm in time, the execution module carries out linkage processing, and the remote monitoring center realizes remote regulation and control. The oil extraction safety and efficiency are improved, and practicability is high.

Description

Technical Field

[0001] This invention relates to the field of oil extraction, specifically to an intelligent oil extraction safety control system and method. Background Technology

[0002] Oil extraction is a crucial part of the petroleum industry. Its working environment is complex and variable, posing numerous safety hazards, such as excessively high wellhead pressure, abnormal temperatures, and oil and gas leaks. If these hazards are not handled properly, they can easily lead to safety accidents, causing casualties and property damage. Currently, traditional oilfield safety control relies mainly on manual inspections and simple monitoring equipment, which suffers from problems such as untimely monitoring, delayed early warnings, and low processing efficiency. Manual inspections are limited by time and space, making it difficult to achieve real-time and comprehensive monitoring of the oilfield process. Problems are often only discovered after an accident occurs, failing to provide early warnings and timely intervention. Simple monitoring equipment has limited functionality, monitoring only a subset of parameters, and has limited data processing capabilities, making it unable to accurately assess and quickly respond to anomalies. Therefore, developing an intelligent oil production safety control system and method that can monitor in real time, provide timely early warnings, and quickly handle abnormal situations is of great significance for improving the safety and efficiency of oil production operations. Summary of the Invention

[0003] The purpose of this invention is to provide an intelligent oil production safety control system and method to solve the problems mentioned in the background art.

[0004] To achieve the above objectives, the present invention provides the following technical solution: an intelligent oil production safety control system, comprising a monitoring module for real-time acquisition of various parameters during the oil production process via sensors, including but not limited to wellhead pressure, temperature, flow rate, oil and gas concentration, equipment operating status parameters, dynamometer card data, and dynamic fluid level related data; a control module connected to the monitoring module for receiving the parameters acquired by the monitoring module and analyzing and processing the parameters, including data cleaning, operating condition diagnosis, production volume pump efficiency calculation, and dynamic fluid level inversion conversion; an early warning module connected to the control module for issuing an early warning signal when the control module detects abnormal parameters; an execution module connected to the control module for performing corresponding operations according to the instructions of the control module to handle abnormal situations; and a remote monitoring center connected to the control module via a communication module for receiving parameter information and early warning signals sent by the control module and sending control instructions to the control module. The control module includes a parameter threshold database and an anomaly data model. Data cleaning filters and selects data based on the anomaly data model, removing unqualified data. The operating condition diagnosis includes two algorithms: an improved single-function dynamometer diagram (SDM) graphic diagnosis and an improved historical dynamometer diagram feature value comparison diagnosis, used to diagnose abnormal well conditions in the tubing string. The production rate pump efficiency calculation is used to calculate the production rate of the collected oil pumps. The dynamic fluid level inversion and conversion uses an improved acoustic method and an improved pressure recovery method to calculate dynamic fluid level data in real time, accurately obtaining the dynamic fluid level depth of the oil well. The parameter threshold database stores the normal range thresholds for various parameters during oil production. The anomaly data model is constructed based on the characteristics of abnormal data during oil production.

[0005] Preferably, the execution module includes a valve control unit, a pump control unit, and an emergency shutdown unit. The valve control unit is used to control the opening and closing of the wellhead valves, the pump control unit is used to control the start, stop, and speed of the oil production pump, and the emergency shutdown unit is used to stop the operation of the oil production equipment in an emergency.

[0006] Preferably, the communication module adopts a wireless communication method, which includes one or more of 4G, 5G, LoRa or NB-IoT.

[0007] Preferably, the specific implementation logic of the control module is as follows: Data cleaning: Filtering and screening data according to the anomaly data model to remove unqualified data. Specifically: receiving data collected by the monitoring module; extracting data features; comparing the extracted features with the features in the anomaly data model; if the data features match the anomaly data model, it is determined to be unqualified data and removed. Operating condition diagnosis: Improved single-power diagram graphical diagnostic algorithm: The single-power diagrams acquired by the monitoring module are preprocessed to remove noise interference. A Gaussian filtering algorithm is used, and the formula is as follows: ,in These are the filtered pixel values. For pixel coordinates, The standard deviation is, and The value is set to 1-3 based on the noise level of the dynamometer; extract the characteristic parameters of the single dynamometer, including but not limited to the maximum and minimum load values ​​and stroke length; compare the characteristic parameters with those of the standard dynamometer and calculate the degree of difference using the following formula: ,in For the degree of difference, The first dynamometer card to be diagnosed One feature parameter, The first standard dynamometer diagram One feature parameter, The number of feature parameters; when the difference When the threshold is exceeded, it is determined to be an abnormal operating condition of the corresponding type; Improved historical dynamometer card feature value comparison and diagnostic algorithm: Collect dynamometer card data of oil wells over a set period to establish a historical dynamometer card feature database; calculate the similarity between the feature values ​​of the current dynamometer card and those of historical dynamometer cards using a cosine similarity algorithm, the formula is as follows: ,in For similarity, The current work diagram is the first 1 eigenvalue, The first historical map Each feature value; when similarity When the value is below the set threshold, the characteristics of historical abnormal power diagrams are used to determine whether it is an abnormal working condition of the tubing. Production volume pump efficiency calculation: An improved production volume calculation algorithm is used, specifically as follows: The stroke, number of strokes, and pump diameter parameters of the oil pump are collected; the production volume is calculated according to the production volume calculation formula, which is: ,in For the liquid production rate, For correction factor, The settings are based on the properties of the fluid inside the well. For stroke, For the next stroke, Pump diameter, To improve pump efficiency, the calculation results were corrected, and the correction coefficients were obtained by fitting actual production data. Dynamic liquid surface inversion and conversion: First, an improved acoustic method is used for calculation: an acoustic signal is emitted and the reflected signal is received, and the propagation time of the acoustic wave is recorded. According to the formula Calculate the depth of the dynamic fluid surface ,in The speed at which sound waves propagate in the well medium. The correction is made based on the well temperature and pressure parameters, and the correction formula is as follows: ,in The speed of sound under standard conditions. This is the temperature correction factor. This is the pressure correction factor. The temperature inside the well. The wellbore pressure is given; then, the dynamic fluid level depth is calculated using the improved pressure recovery method, with the equation being: ,in For the well The pressure of constant time The original formation pressure, For production, For fluid viscosity, For penetration rate, For oil layer thickness, Porosity The overall compression coefficient is... Let be the wellbore radius, and the depth of the dynamic fluid level can be obtained by inversion using this equation.

[0008] Preferably, the specific calculation logic for calculating the liquid production volume according to the liquid production volume calculation formula is as follows: Step 1: Determine the correction factor Influencing factors and basic values: Based on the actual operating conditions of the oil well, the viscosity of the fluid inside the oil well is collected in real time through the monitoring module. gas content Moisture content Parameters, establishing fluid properties and A database of associated values; initially set according to the conventional classification of fluid properties. The basic value, that is When the fluid is of low viscosity Low gas content Low moisture content When crude oil, Set the value to 0.92; Step 2: Fitting based on actual production data Corrected model: Collect actual fluid production data of oil wells over the past 6 months. ,in The stroke data collected by the monitoring module within the corresponding time period was measured using a metering separator. Strike Pump diameter Pump efficiency and fluid property parameters; substitute the above data into the formula Reverse the historical process Value, that is Subsequently, the least squares method was used to... A multiple linear regression was performed with the fluid property parameters to obtain the fitting formula: ,in These are the regression coefficients; Step 3: Real-time calculation and dynamic correction value: The control module receives data collected in real time from the monitoring module. And fluid property parameters, first determined according to step 1. Then substitute the values ​​into the fitting formula from step 2 to calculate the real-time... Value; calculate the daily liquid production volume every 24 hours. Compared with the measured liquid production If the error exceeds 3%, the regression coefficients should be re-optimized. ,make sure The value always matches the actual working conditions; Step 4: Precise calculation and error control of liquid production: Real time Substitute the value into the formula The final liquid yield is obtained.

[0009] Preferably, the specific implementation logic of the improved acoustic method is as follows: Step I: Determine the basic value of sound wave propagation speed Based on the geological conditions of the oil well and the type of media within the well (such as a mixture of crude oil, water, and natural gas), and through laboratory simulations and historical data statistics, the standard conditions (temperature) are determined. ,pressure The fundamental value of the speed of sound propagation. ; Step II: Construct a temperature-pressure correction model: Collect data on different temperatures of oil wells over the past 3 months. ,pressure Measured values ​​of sound wave propagation speed under certain conditions ,in Data was simultaneously acquired using an acoustic logging tool, and corresponding monitoring data from temperature and pressure sensors were recorded. and Then, a multiple linear regression algorithm is used to... and By fitting the model, the corrected model is obtained: ;in This is the temperature correction factor. This is the pressure correction factor; Step III: Real-time parameter acquisition and calculation of corrected acoustic velocity: The monitoring module acquires the well temperature in real time through temperature and pressure sensors. and pressure The collected temperature and pressure Data is transmitted to the control module; the control module calls the steps in step I. The corrected model from step II automatically calculates the real-time sound wave propagation speed. : ; Step IV: Dynamically optimize the correction coefficient: Every 72 hours, the control module will adjust the calculated sound wave velocity value according to the correction. Compared with the measured values ​​of the synchronous sonic logging tool Perform comparisons and calculate the error rate. ;like If the result exceeds 2%, then refit the data based on the latest 100 sets of data. Coefficients, optimize and correct the model; Step V: Calculate the depth of the dynamic fluid level The control module uses the time difference between sound wave emission and reflection. (Data acquired by an acoustic sensor, accuracy ±1ms), combined with the corrected acoustic velocity. Through formula Calculate the depth of the dynamic fluid level. Because... It has been corrected for temperature and pressure to fit the actual well environment, compared to the traditional uncorrected sonic method. The calculation error is reduced, providing accurate data support for wellbore liquid accumulation early warning and oil pump condition adjustment in oil production safety control.

[0010] Preferably, an intelligent oil production safety control method includes the following steps: Step S1: The monitoring module collects parameters and sends them to the control module; Step S2: The control module cleans and analyzes the data, and determines whether there is any abnormality by combining the results of working condition diagnosis, production pump efficiency calculation and dynamic liquid level inversion conversion. Step S3: If an anomaly is detected, the control module sends an early warning command to the early warning module, the early warning module issues an early warning signal, and at the same time, the control module sends corresponding control commands to the execution module according to the anomaly. Step S4: The execution module receives control commands and performs corresponding operations to handle abnormal situations; Step S5: The control module sends parameter information and early warning signals to the remote monitoring center through the communication module. The remote monitoring center sends control commands to the control module as needed to remotely regulate the oil extraction process.

[0011] Compared with the prior art, the beneficial effects of the present invention are: Improving data reliability and analysis accuracy: Monitoring data is cleaned using an anomaly data model, accurately removing unqualified data and providing a high-quality data foundation for subsequent analysis, thus solving the problem of misjudgment caused by data interference in traditional monitoring. The control module integrates multiple optimization algorithms, and the operating condition diagnosis adopts an improved single-power diagram graphic diagnosis and historical power diagram feature value comparison diagnosis algorithm, which significantly improves the accuracy of abnormal operating condition identification for wells with abnormal tubing such as sucker rod breakage and tubing leakage.

[0012] Achieving accurate calculation of oil production parameters: Introducing correction coefficients into production rate calculations By dynamically adjusting the parameters based on the properties of the fluid in the well and optimizing them through fitting with actual data, the error between the actual fluid production and the actual production rate is small when the parameters are accurate, overcoming the large error of the traditional effective stroke method. The dynamic fluid level inversion calculation adopts an improved sonic method and pressure recovery method. The sonic velocity is corrected for temperature and pressure, and the pressure recovery calculation integrates multiple formation parameters, which significantly reduces the error in the dynamic fluid level depth calculation, providing accurate data for wellbore fluid accumulation early warning and oil pump control.

[0013] Enhancing the timeliness and effectiveness of safety control: The monitoring module collects multi-dimensional parameters such as wellhead pressure, temperature, and dynamometer card data in real time. The control module quickly analyzes and processes these parameters, and the early warning module promptly alerts in case of abnormalities, simultaneously sending instructions to the execution module. Through operations such as valve regulation, pump speed adjustment, or emergency shutdown, rapid response and handling of anomalies are achieved, solving the problems of slow and inefficient traditional manual inspections. The remote monitoring center receives data and early warning information in real time via wireless communication (4G / 5G / LoRa, etc.) and can remotely issue control commands to achieve remote collaborative management, which is particularly suitable for ensuring the safety of oil production operations in complex or harsh environments.

[0014] Improving overall efficiency and economy of oil production operations: Precise production calculation and dynamic fluid level monitoring provide a scientific basis for optimizing oil production parameters. Dynamic adjustment of parameters such as pump efficiency and stroke through execution modules achieves high-efficiency oil production and reduces energy consumption. Early warning and fault diagnosis capabilities reduce downtime due to sudden equipment failures, lowering maintenance costs and accident losses, while also reducing manual inspection workload and saving labor costs.

[0015] Enhanced system adaptability and scalability: The system parameter threshold database can be flexibly adjusted according to different well conditions, and the abnormal data model is dynamically optimized during the oil production process, making it suitable for various oil well geological conditions and production stages. The modular design facilitates functional expansion, allowing for the addition of new monitoring parameters or control units as needed, meeting the upgrading requirements of oil production technology development.

[0016] In summary, this invention comprehensively improves the safety, reliability, and economy of oil production operations through a collaborative mechanism of intelligent monitoring, precise analysis, and rapid response, and has significant practical value and promotional significance. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the system structure of the present invention; Figure 2 This is a schematic diagram of the method flow of the present invention. Detailed Implementation

[0018] 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.

[0019] Example 1 Please see Figure 1The present invention provides a technical solution: an intelligent oil production safety control system, including a monitoring module, which is used to collect various parameters in the oil production process in real time through sensors. The parameters include, but are not limited to, wellhead pressure, temperature, flow rate, oil and gas concentration, equipment operating status parameters, dynamometer card data, and dynamic fluid level related data. It should be noted that the monitoring module includes multiple sensors, each used to collect different parameters. For example, a pressure sensor is used to collect wellhead pressure, a temperature sensor is used to collect wellhead temperature, a flow sensor is used to collect oil and gas velocity and flow rate, a gas sensor is used to detect oil and gas concentration, an equipment status sensor is used to monitor the operating status parameters of the oil production equipment, and a dynamometer card sensor is used to collect dynamometer card data, etc.

[0020] The control module, connected to the monitoring module, receives parameters collected by the monitoring module and analyzes and processes these parameters. This processing includes data cleaning, operational condition diagnosis, production volume and pump efficiency calculation, and dynamic fluid level inversion and conversion. The control module contains a parameter threshold database and an abnormal data model. Data cleaning filters and selects data based on the abnormal data model, removing unqualified data. Operational condition diagnosis includes two algorithms: an improved single-function dynamometer diagram (SDM) graphic diagnosis and an improved historical dynamometer diagram feature value comparison diagnosis. These algorithms accurately diagnose abnormal well conditions such as sucker rod breakage, tubing leakage, and pump leakage. The production volume and pump efficiency calculation is used to calculate the production volume of the pump. The dynamic fluid level inversion and conversion uses an improved acoustic method and an improved pressure recovery method to calculate dynamic fluid level data in real time, accurately obtaining the dynamic fluid level depth of the oil well, thereby comprehensively and accurately understanding the dynamic fluid level change patterns. When the collected parameters exceed the corresponding threshold after processing and analysis, the control module determines that the parameters are abnormal; the parameter threshold database stores the normal range thresholds of various parameters during the oil production process; the abnormal data model is constructed based on the abnormal data characteristics during the oil production process. The specific implementation logic of the control module is as follows: Data cleaning: Filtering and screening data according to the anomaly data model to remove unqualified data. Specifically: receiving data collected by the monitoring module; extracting data features; comparing the extracted features with the features in the anomaly data model; if the data features match the anomaly data model, it is determined to be unqualified data and removed. Operating Condition Diagnosis: Improved Single-Power Diagram Graphical Diagnosis Algorithm: The single-power diagrams collected by the monitoring module are preprocessed to remove noise interference. A Gaussian filtering algorithm is used, and the formula is as follows: ,in These are the filtered pixel values. For pixel coordinates, The standard deviation is, and The value is set to 1-3 based on the noise level of the dynamometer; extract the characteristic parameters of the single dynamometer, including but not limited to the maximum and minimum load values ​​and stroke length; compare the characteristic parameters with those of the standard dynamometer and calculate the degree of difference using the following formula: ,in For the degree of difference, The first dynamometer card to be diagnosed One feature parameter, The first standard dynamometer diagram One feature parameter, The number of feature parameters; when the difference When the threshold is exceeded, it is determined to be an abnormal operating condition of the corresponding type; Improved historical dynamometer card feature value comparison and diagnostic algorithm: Collect dynamometer card data of oil wells over a set period to establish a historical dynamometer card feature database; calculate the similarity between the feature values ​​of the current dynamometer card and those of historical dynamometer cards using a cosine similarity algorithm, the formula is as follows: ,in For similarity, The current work diagram is the first 1 eigenvalue, The first historical map Each feature value; when similarity When the value is below the set threshold, the characteristics of historical abnormal power diagrams are used to determine whether it is an abnormal working condition of the tubing. Production volume pump efficiency calculation: An improved production volume calculation algorithm is used, specifically as follows: The stroke, number of strokes, and pump diameter parameters of the oil pump are collected; the production volume is calculated according to the production volume calculation formula, which is: ,in For the liquid production rate, For correction factor, The settings are based on the properties of the fluid inside the well. For stroke, For the next stroke, Pump diameter, To improve pump efficiency, the calculation results were corrected, and the correction coefficients were obtained by fitting actual production data. Dynamic liquid surface inversion and conversion: First, an improved acoustic method is used for calculation: an acoustic signal is emitted and the reflected signal is received, and the propagation time of the acoustic wave is recorded. According to the formula Calculate the depth of the dynamic fluid surface ,in The speed at which sound waves propagate in the well medium. The correction is made based on the well temperature and pressure parameters, and the correction formula is as follows: ,in The speed of sound under standard conditions. This is the temperature correction factor. This is the pressure correction factor. The temperature inside the well. The pressure inside the well; where the speed of sound wave propagation is a factor. Make corrections and introduce temperature. and pressure The correction term, through the correction coefficient The adjustments made the sound wave velocity more closely match the actual well environment, improving the accuracy of dynamic fluid level depth calculation.

[0021] The following is about A brief explanation of the coefficients: Correction coefficient (Temperature Correction Term): The value ranges from 0.98 to 1.01. Referring to the "Influence of Temperature on Crude Oil Acoustic Wave Propagation" section in the Petroleum Engineering Handbook, and combining the temperature gradient data from 30 wells in Block C (temperature increases by 3°C for every 100m increase in well depth), linear interpolation is used to determine the corresponding temperature correction term for different well depths. value.

[0022] Correction coefficient (Environmental pressure correction term): Value range 0.95-1.03, based on the linear relationship between well pressure and sound wave velocity (following the sound wave propagation velocity formula in fluid mechanics). ),in The bulk modulus of the fluid. (The fluid density) was obtained by fitting pressure-sonic velocity data from 20 wells in Block A. , (This refers to the bottom hole pressure, in MPa). Then, the improved pressure recovery method is used to calculate the dynamic fluid level depth, as shown in the equation: ,in For the well The pressure of constant time Original formation pressure , For production , fluid viscosity , For penetration rate , Oil layer thickness , Porosity The overall compression coefficient , wellbore radius The dynamic fluid level depth is obtained through inversion using this equation. This method comprehensively considers multiple parameters such as production rate, fluid viscosity, and permeability. Compared to traditional methods, calculating the dynamic fluid level depth through equation inversion provides a more comprehensive reflection of the well conditions and improves the reliability of the results.

[0023] Here is an explanation: The improved single-function dynamometer graphic diagnostic algorithm: introduces a Gaussian filtering algorithm to remove noise, and adapts to different noise conditions by setting a standard deviation, thereby improving the dynamometer preprocessing effect; uses a difference calculation formula to quantify the difference between the dynamometer to be diagnosed and the standard dynamometer, making the anomaly judgment more objective and improving the diagnostic accuracy; the improved historical dynamometer feature value comparison diagnostic algorithm: uses a cosine similarity algorithm to calculate the similarity between the current and historical dynamometer feature values, judges anomalies by similarity threshold, and combines historical abnormal dynamometer features to enhance the ability to identify abnormal working conditions of the tubing, further improving the diagnostic accuracy rate; The specific calculation logic for the liquid production volume based on the liquid production volume calculation formula is as follows: Step 1: Determine the correction factor Influencing factors and basic values: Based on the actual operating conditions of the oil well, the viscosity of the fluid inside the oil well is collected in real time through the monitoring module. gas content Moisture content Parameters, establishing fluid properties and A database of associated values; initially set according to the conventional classification of fluid properties. The basic value, that is When the fluid is of low viscosity Low gas content Low moisture content When crude oil, The value is set to 0.92; when the fluid viscosity, gas content, or water content increases, it decreases according to the preset gradient. ,like hour, The price has been lowered to 0.88-0.90. The details are as follows: Should The coefficients are used to correct for the influence of fluid properties on in-well detection signals (such as acoustic waves and pressure). For example, the basic values ​​and gradient ranges are determined based on statistical analysis of historical data from 120 production wells in three typical oilfield blocks across the country (low-permeability oilfield block A, medium-high permeability oilfield block B, and high water-cut oilfield block C) over the past three years. Low viscosity (≤50mPa・s), low gas content (≤3%), and low water content (≤20%) crude oil: The acoustic propagation error data of 45 wells out of 120 wells that meet this condition were statistically analyzed, and the average correction coefficient was 0.92 (fitting error ≤3%) by linear regression fitting. When the fluid viscosity (50-100 mPa·s), air content (3%-8%), or water content (20%-40%) increases: Based on orthogonal experimental design, the sensitivity weights of the three parameters to sound wave attenuation are analyzed (viscosity: air content: water content = 4:3:3). The gradient descent method is used to determine the coefficient reduction gradient corresponding to each parameter increase (the coefficient is reduced by 0.01-0.02 for each increase), ultimately forming an adjustment range of 0.88-0.90.

[0024] The method for determining this (the actual production data fitting process, which is existing technology, and will be briefly described here): Step 1: Data Acquisition and Preprocessing: Collect 15 types of characteristic data from the target oilfield block, including fluid physical properties (viscosity, gas cut, water cut), well structure parameters (well depth, pipe diameter), and production dynamic data (production rate, bottom hole pressure). The nearest neighbor (KNN) algorithm is used to fill in missing values, and principal component analysis (PCA) is used to extract core influencing features (the top 5 principal components are retained, with a cumulative contribution rate of ≥92%). Step 2: Fitting Model Construction: Based on the random forest algorithm, establish a mapping relationship model between fluid properties, sound wave propagation error, and correction coefficient. Divide the preprocessed feature data in Step 1 into a training set (for model training) and a validation set (for error verification) in a 7:3 ratio. Step 3: Coefficient optimization and determination: The model is tuned using the grid search method and the adaptive moment estimation (Adam) algorithm to ensure that the model prediction error is ≤5%. Finally, the correction coefficient values ​​and gradient adjustment rules corresponding to different fluid property combinations are output. Step 4: Mine verification: A 6-month field test was conducted in 10 test wells in Block B. By adjusting the coefficient values ​​in real time, the acoustic detection error was reduced from 8.7% before optimization to 2.9% after optimization, verifying the effectiveness of the coefficient range.

[0025] Step 2: Fitting based on actual production data Corrected model: Collect actual fluid production data of oil wells over the past 6 months. ,in The stroke data collected by the monitoring module within the corresponding time period was measured using a metering separator. Strike Pump diameter Pump efficiency and fluid property parameters; substitute the above data into the formula Reverse the historical process Value, that is Subsequently, the least squares method was used to... A multiple linear regression was performed with the fluid property parameters to obtain the fitting formula: ,in These are the regression coefficients; it should be noted that... The regression coefficients are preset in advance by staff based on the actual oil extraction situation.

[0026] Step 3: Real-time calculation and dynamic correction value: The control module receives data collected in real time from the monitoring module. And fluid property parameters, first determined according to step 1. Then substitute the values ​​into the fitting formula from step 2 to calculate the real-time... Value; calculate the daily liquid production volume every 24 hours. Compared with the measured liquid production If the error exceeds 3%, the regression coefficients should be re-optimized. ,make sure The value always matches the actual working conditions; Step 4: Precise calculation and error control of liquid production: Real time Substitute the value into the formula The final liquid yield is obtained.

[0027] Productivity Pump Efficiency Calculation: Innovative Design with Correction Factors The calculation formula, Based on the properties of the fluid in the well and fitted and corrected according to actual production data, the problem of large errors in the traditional effective stroke method is solved. When the parameters are accurate, the error between the actual fluid production and the actual production volume can be reduced, and the calculation accuracy is greatly improved.

[0028] The specific implementation logic of the improved acoustic method is as follows: Step I: Determine the basic value of sound wave propagation speed Based on the geological conditions of the oil well and the type of media within the well (such as a mixture of crude oil, water, and natural gas), and through laboratory simulations and historical data statistics, the standard conditions (temperature) are determined. ,pressure The fundamental value of the speed of sound propagation. ; Step II: Construct a temperature-pressure correction model: Collect data on different temperatures of oil wells over the past 3 months. ,pressure Measured values ​​of sound wave propagation speed under certain conditions ,in Data was simultaneously acquired using an acoustic logging tool, and corresponding monitoring data from temperature and pressure sensors were recorded. and Then, a multiple linear regression algorithm is used to... and By fitting the model, the corrected model is obtained: ; Step III: Real-time parameter acquisition and calculation of corrected acoustic velocity: The monitoring module acquires the well temperature in real time through temperature and pressure sensors. and pressure The collected temperature and pressure Data is transmitted to the control module; the control module calls the steps in step I. The corrected model from step II automatically calculates the real-time sound wave propagation speed. : ; Step IV: Dynamically optimize the correction coefficient: Every 72 hours, the control module will adjust the calculated sound wave velocity value according to the correction. Compared with the measured values ​​of the synchronous sonic logging tool Perform comparisons and calculate the error rate. ;like If the result exceeds 2%, then refit the data based on the latest 100 sets of data. Coefficients, optimize and correct the model; Step V: Calculate the depth of the dynamic fluid level The control module uses the time difference between sound wave emission and reflection. (Data acquired by an acoustic sensor, accuracy ±1ms), combined with the corrected acoustic velocity. Through formula Calculate the depth of the dynamic fluid level. Because... It has been corrected for temperature and pressure to fit the actual well environment, compared to the traditional uncorrected sonic method. The calculation error is reduced, providing accurate data support for wellbore liquid accumulation early warning and oil pump condition adjustment in oil production safety control.

[0029] Further explanation is needed regarding the depth of the dynamic fluid level. These are core safety parameters for oil extraction operations, and their values ​​directly affect: assessment of the operating condition of the oil pump; if... If the liquid level is too shallow (too low), the pump may experience accelerated wear due to "cavitation," and could even lead to sucker rod breakage (directly related to the "sucker rod breakage" diagnosis in the control module's "Operating Condition Diagnosis"); if Excessive depth (excessive fluid level) may lead to fluid accumulation in the wellbore, increasing wellhead pressure (linked to the "wellhead pressure monitoring" module).

[0030] Production volume control is based on the dynamic liquid level change reflecting the formation's liquid supply capacity. Combined with the "production volume calculation" of the control module, the pump stroke rate and valve opening can be adjusted through the execution module to avoid equipment overload or resource waste caused by insufficient or excessive liquid supply.

[0031] Security alert triggered: When If the preset safety threshold is exceeded (e.g., shallower than 100m or deeper than 500m, stored in the "parameter threshold database" of the control module), the control module will trigger the alarm of the early warning module and link the execution module to take measures (e.g., reduce the number of strokes or stop the machine).

[0032] An early warning module is connected to the control module. When the control module detects abnormal parameters, the early warning module issues an early warning signal. An execution module, connected to the control module, is used to perform corresponding operations according to the instructions of the control module to handle abnormal situations. The execution module includes a valve control unit, a pump control unit, and an emergency shutdown unit. The valve control unit is used to control the opening and closing of the wellhead valves, the pump control unit is used to control the start, stop, and speed of the oil production pump, and the emergency shutdown unit is used to stop the operation of the oil production equipment in an emergency.

[0033] The remote monitoring center is connected to the control module via a communication module. It is used to receive parameter information and early warning signals sent by the control module and to send control commands to the control module. The communication module adopts a wireless communication method, which includes one or more of 4G, 5G, LoRa, or NB-IoT.

[0034] Example 2 Please see Figure 2 An intelligent oil production safety control method includes the following steps: Step S1: The monitoring module collects parameters and sends them to the control module; Step S2: The control module cleans and analyzes the data, and determines whether there is any abnormality by combining the results of working condition diagnosis, production pump efficiency calculation and dynamic liquid level inversion conversion. Step S3: If an anomaly is detected, the control module sends an early warning command to the early warning module, the early warning module issues an early warning signal, and at the same time, the control module sends corresponding control commands to the execution module according to the anomaly. Step S4: The execution module receives control commands and performs corresponding operations to handle abnormal situations; Step S5: The control module sends parameter information and early warning signals to the remote monitoring center through the communication module. The remote monitoring center sends control commands to the control module as needed to remotely regulate the oil extraction process.

[0035] This invention discloses an intelligent oil production safety control system and method, relating to the field of oil production technology. The system includes a monitoring module, a control module, an early warning module, an execution module, and a remote monitoring center. The monitoring module collects parameters such as wellhead pressure, temperature, flow rate, dynamometer card data, and dynamic fluid level data in real time through sensors. The control module cleans the data, performs operational condition diagnosis, calculates production volume and pump efficiency, and performs dynamic fluid level inversion and conversion. It has a built-in parameter threshold database and anomaly data model, employs improved single dynamometer card graphic diagnosis and historical dynamometer card feature value comparison diagnosis algorithms, and innovatively designs a production volume calculation method with correction coefficients. It also accurately inverts the dynamic fluid level depth using improved acoustic and pressure recovery methods. The early warning module issues warnings when parameters are abnormal. The execution module performs operations such as valve regulation, pump speed adjustment, or emergency shutdown according to control commands. The remote monitoring center receives data and performs remote control. This invention can monitor and accurately diagnose anomalies in real time, respond quickly, and improve the safety, reliability, and efficiency of oil production operations, possessing high practical value.

[0036] 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 intelligent oilfield safety control system, characterized in that, The system includes a monitoring module for real-time acquisition of various parameters during the oil production process via sensors, including but not limited to wellhead pressure, temperature, flow rate, oil and gas concentration, equipment operating status parameters, dynamometer card data, and dynamic fluid level data; a control module connected to the monitoring module for receiving the parameters acquired by the monitoring module and analyzing and processing them, including data cleaning, operating condition diagnosis, production volume and pump efficiency calculation, and dynamic fluid level inversion and conversion; an early warning module connected to the control module for issuing an early warning signal when the control module detects abnormal parameters; an execution module connected to the control module for performing corresponding operations according to the instructions of the control module to handle abnormal situations; and a remote monitoring center connected to the control module via a communication module for receiving parameter information and early warning signals sent by the control module and sending control commands to the control module. The control module includes a parameter threshold database and an anomaly data model. Data cleaning filters and selects data based on the anomaly data model, removing unqualified data. The operating condition diagnosis includes two algorithms: an improved single-function dynamometer diagram (SDM) graphic diagnosis and an improved historical dynamometer diagram feature value comparison diagnosis, used to diagnose abnormal well conditions in the tubing string. The production rate pump efficiency calculation is used to calculate the production rate of the collected oil pumps. The dynamic fluid level inversion and conversion uses an improved acoustic method and an improved pressure recovery method to calculate dynamic fluid level data in real time, accurately obtaining the dynamic fluid level depth of the oil well. The parameter threshold database stores the normal range thresholds for various parameters during oil production. The anomaly data model is constructed based on the characteristics of abnormal data during oil production.

2. The intelligent oilfield safety control system according to claim 1, characterized in that: The execution module includes a valve control unit, a pump control unit, and an emergency shutdown unit. The valve control unit is used to control the opening and closing of the wellhead valves, the pump control unit is used to control the start, stop, and speed of the oil production pump, and the emergency shutdown unit is used to stop the operation of the oil production equipment in an emergency.

3. The intelligent oilfield safety control system according to claim 1, characterized in that: The communication module adopts a wireless communication method, which includes one or more of 4G, 5G, LoRa or NB-IoT.

4. The intelligent oil production safety control system according to claim 1, characterized in that: The specific implementation logic of the control module is as follows: Data cleaning: Filtering and screening data according to the anomaly data model to remove unqualified data. Specifically: receiving data collected by the monitoring module; extracting data features; comparing the extracted features with the features in the anomaly data model; if the data features match the anomaly data model, it is determined to be unqualified data and removed. Operating Condition Diagnosis: Improved Single-Power Diagram Graphical Diagnosis Algorithm: The single-power diagrams collected by the monitoring module are preprocessed to remove noise interference. A Gaussian filtering algorithm is used, and the formula is as follows: ,in These are the filtered pixel values. For pixel coordinates, The standard deviation is, and The value is set to 1-3 based on the noise level of the dynamometer; extract the characteristic parameters of the single dynamometer, including but not limited to the maximum and minimum load values ​​and stroke length; compare the characteristic parameters with those of the standard dynamometer and calculate the degree of difference using the following formula: ,in For the degree of difference, The first dynamometer card to be diagnosed One feature parameter, The first standard dynamometer diagram One feature parameter, The number of feature parameters; when the difference When the threshold is exceeded, it is determined to be an abnormal operating condition of the corresponding type; Improved historical dynamometer card feature value comparison and diagnostic algorithm: Collect dynamometer card data of oil wells over a set period to establish a historical dynamometer card feature database; calculate the similarity between the feature values ​​of the current dynamometer card and those of historical dynamometer cards using a cosine similarity algorithm, the formula is as follows: ,in For similarity, The current work diagram is the first 1 eigenvalue, The first historical map Each feature value; when similarity When the value is below the set threshold, the characteristics of historical abnormal power diagrams are used to determine whether it is an abnormal working condition of the tubing. Production volume pump efficiency calculation: An improved production volume calculation algorithm is adopted, and the specific calculation is as follows: Collect the stroke, number of strokes, and pump diameter parameters of the oil pump; The production volume is calculated using the production volume calculation formula, which is: ,in For the liquid production rate, For correction factor, The settings are based on the properties of the fluid inside the well. For stroke, For the next stroke, Pump diameter, To improve pump efficiency, the calculation results were corrected, and the correction coefficients were obtained by fitting actual production data. Dynamic liquid surface inversion and conversion: First, an improved acoustic method is used for calculation: an acoustic signal is emitted and the reflected signal is received, and the propagation time of the acoustic wave is recorded. According to the formula Calculate the depth of the dynamic fluid level ,in The speed at which sound waves propagate in the well medium. The correction is made based on the well temperature and pressure parameters, and the correction formula is as follows: ,in The speed of sound under standard conditions. This is a temperature correction factor. This is the pressure correction factor. The temperature inside the well. This refers to the pressure inside the well. Then, the improved pressure recovery method is used to calculate the dynamic fluid level depth, as shown in the equation: ,in For the well The pressure of constant time The original formation pressure, For production, For fluid viscosity, For penetration rate, For oil layer thickness, Porosity The overall compression coefficient is... Let be the wellbore radius, and the depth of the dynamic fluid level can be obtained by inversion using this equation.

5. The intelligent oilfield safety control system according to claim 4, characterized in that: The calculation logic for the product flow rate pump efficiency is as follows: Step 1: Determine the correction factor Influencing factors and basic values: Based on the actual operating conditions of the oil well, the viscosity of the fluid inside the oil well is collected in real time through the monitoring module. gas content Moisture content Parameters, establishing fluid properties and A database of associated values; initially set according to the conventional classification of fluid properties. The basic value, that is When the fluid is of low viscosity Low gas content Low moisture content When crude oil, Set the value to 0.92; Step 2: Fitting based on actual production data Corrected model: Collect actual fluid production data of oil wells over the past 6 months. ,in The stroke data collected by the monitoring module within the corresponding time period was measured using a metering separator. Strike Pump diameter Pump efficiency and fluid property parameters; substitute the above data into the formula Reverse the historical process Value, that is Subsequently, the least squares method was used to... A multiple linear regression was performed with the fluid property parameters to obtain the fitting formula: ,in These are the regression coefficients; Step 3: Real-time calculation and dynamic correction value: The control module receives real-time data collected by the monitoring module. And fluid property parameters, first determined according to step 1. Then substitute the values ​​into the fitting formula from step 2 to calculate the real-time... Value; calculate the daily liquid production volume every 24 hours. Compared with the measured liquid production If the error exceeds 3%, the regression coefficients should be re-optimized. ,make sure The value always matches the actual working conditions; Step 4: Precise calculation and error control of liquid production: Real time Substitute the value into the formula The final liquid yield is obtained.

6. The intelligent oilfield safety control system according to claim 4, characterized in that: The specific implementation logic of the improved acoustic method is as follows: Step I: Determine the basic value of sound wave propagation speed Based on the geological conditions and well media type of the oil well, and through laboratory simulation and historical data statistics, the baseline value of sound wave propagation velocity under standard conditions was determined. ; Step II: Construct a temperature-pressure correction model: Collect data on different temperatures of oil wells over the past 3 months. ,pressure Measured values ​​of sound wave propagation speed under certain conditions ,in Data was simultaneously acquired using an acoustic logging tool, and corresponding monitoring data from temperature and pressure sensors were recorded. and ; Then, the multiple linear regression algorithm is used to... and By fitting the model, the corrected model is obtained: ,in This is a temperature correction factor. This is the pressure correction factor; Step III: Real-time parameter acquisition and calculation of corrected acoustic velocity: The monitoring module acquires the well temperature in real time through temperature and pressure sensors. and pressure The collected temperature and pressure Data is transmitted to the control module; the control module calls the steps in step I. The corrected model from step II automatically calculates the real-time sound wave propagation speed. : ; Step IV: Dynamically optimize the correction coefficient: Every 72 hours, the control module will adjust the calculated sound wave velocity value according to the correction. Compared with the measured values ​​of the synchronous sonic logging tool Perform comparisons and calculate the error rate. ;like If the result exceeds 2%, then refit the data based on the latest 100 sets of data. Coefficients, optimize and correct the model; Step V: Calculate the depth of the dynamic fluid level The control module uses the time difference between sound wave emission and reflection. Combined with the corrected sound wave velocity Through formula Calculate the depth of the dynamic liquid level.

7. An intelligent oilfield safety control method, applied to an intelligent oilfield safety control system as described in any one of claims 1-6, characterized in that, Includes the following steps: Step S1: The monitoring module collects parameters and sends them to the control module; Step S2: The control module cleans and analyzes the data, and determines whether there is any abnormality by combining the results of working condition diagnosis, production pump efficiency calculation and dynamic liquid level inversion conversion. Step S3: If an anomaly is detected, the control module sends an early warning command to the early warning module, the early warning module issues an early warning signal, and at the same time, the control module sends corresponding control commands to the execution module according to the anomaly. Step S4: The execution module receives control commands and performs corresponding operations to handle abnormal situations; Step S5: The control module sends parameter information and early warning signals to the remote monitoring center through the communication module. The remote monitoring center sends control commands to the control module as needed to remotely regulate the oil extraction process.

Images