Intelligent oil and gas wellbore perforation sand control system and method

Abstract

The application discloses an intelligent oil and gas wellbore perforation sand control system and method, and a control system thereof comprises a perforator, multiple sets of aperture adjusting assemblies, multiple sets of angle adjusting assemblies, a data acquisition assembly and a control assembly; the aperture adjusting assembly is arranged in each perforation inlet, the aperture of the perforation is adjusted by the shape deformation of a ring-shaped SMA deformation layer through electrification; the angle adjusting assembly is arranged at each perforation outlet, the hollow shaft is deflected by driving the SMA wire arranged at an angle to adjust the jet angle; the data acquisition assembly acquires the fluid density, conductivity, sound velocity and temperature of the wellbore in real time; the control assembly outputs the electrification time length and current intensity of the SMA elements in the aperture and angle adjusting assemblies based on the deviation amount of the predicted sand production velocity and the threshold value based on a neural network prediction model, and actively adjusts the perforation aperture and angle. The application combines the shape memory alloy driving and intelligent prediction control, has a compact structure, a fast response, can actively adapt to the change of downhole working conditions, and takes into account the sand control and production capacity optimization.

Description

Technical Field

[0001] This invention relates to the field of oil well shaft sand control technology, specifically an intelligent oil and gas well shaft perforation sand control system and method. Background Technology

[0002] In oil and gas well production, formation sand production is a major issue affecting extraction efficiency, equipment lifespan, and operational safety. When the reservoir rock strength is low or the production pressure differential is too large, formation sand particles can enter the wellbore with the fluid. This can cause minor damage such as abrasion of downhole tubing, pumps, valves, and surface equipment, or even wellbore collapse, perforation blockage, and well abandonment. Therefore, sand control measures are usually adopted during the well completion stage. Among these, perforation sand control technology involves creating perforations of specific shapes and sizes in the casing and cement sheath. While ensuring oil and gas flow, it uses natural sand arching or gravel filling to block formation sand migration.

[0003] Traditional perforation sand control solutions mostly employ fixed-parameter perforators, such as shaped charge perforators or hydraulic jets, to create channels with a fixed diameter and direction. The orifice size, orifice density distribution, and jet direction of these perforators are preset in a single operation and cannot be adjusted as downhole production conditions change. However, the production profile of an oil and gas well throughout its entire lifecycle is extremely complex: factors such as rising water cut, decreasing reservoir pressure, and changes in sand particle size distribution all significantly impact sand control requirements and production optimization goals. Fixed-parameter perforations often face the awkward situation of "effective sand control in the initial stage, limited production in the middle stage, and increased sand production in the later stage," making it difficult to achieve a dynamic balance between sand control reliability and production.

[0004] To overcome these problems, the industry has developed various adjustable perforation devices. For example, hydraulically driven sliding-sleeve perforation valves can open or close some perforations according to surface commands, thereby changing the effective perforation density. Another example is some mechanically adjustable perforator devices that use a motor-driven mechanical transmission mechanism to drive a rotating sleeve or baffle to adjust the flow area. However, these solutions generally suffer from complex structures and numerous moving parts, making integration within the confined space of the wellbore extremely difficult, significantly increasing the tool's outer diameter and length, and raising the risk of hydraulic leakage. Furthermore, hydraulic or motor-driven systems have slow response times, making it difficult to make real-time adjustments to rapid fluctuations in fluid characteristics, resulting in a lag problem. Summary of the Invention

[0005] The purpose of this invention is to provide an intelligent oil and gas well perforation sand control system and method to solve the problems of difficult integration and lag in adjustment of current sand control perforators in the narrow space of the well.

[0006] The technical solution of this invention is: An intelligent oil and gas well perforation sand control system includes a perforator, multiple sets of orifice diameter adjustment components, multiple sets of angle adjustment components, a data acquisition component, and a control component. The perforator has multiple perforations. The multiple sets of orifice diameter adjustment components are correspondingly installed within the perforation wall near the inlet of each perforation. Each orifice diameter adjustment component includes a first annular support and an annular SMA deformation layer. The first annular support is fixed to the inner wall of the perforation, and the annular SMA deformation layer is connected to the inner ring surface of the first annular support. The annular SMA deformation layer reduces its inner diameter when electrically heated. The multiple sets of angle adjustment components are correspondingly installed at the outlet of each perforation. Each angle adjustment component includes a second annular support, a hollow shaft, and two sets of SMA wires. The second annular support is fixed to the base of the perforator located at the perforation outlet. The hollow shaft is disposed within the second annular support, with one end connected to the perforation outlet via a flexible hose. The two sets of SMA wires are respectively arranged parallel to each other at both ends of the hollow shaft. Arranged 180° apart along the circumference of the hollow shaft, each set of SMA wires has its two ends connected to the end of the hollow shaft and the inner wall of the second annular support, respectively. The data acquisition component is used to collect the density, conductivity, sound velocity, and temperature parameters of the fluid in the wellbore in real time. The control component is electrically connected to the data acquisition component, the annular SMA deformation layer, and the SMA wires. The control component includes a data processing module and a PID controller. The data processing module has a preset prediction model based on a neural network. The prediction model is used to receive the density, conductivity, sound velocity, and temperature parameters of the fluid in the wellbore and output the predicted value of the sand production rate within a set time period. The data processing module subtracts the predicted sand production rate from the preset sand control sand production rate threshold to obtain the deviation value. The PID controller is used to receive the deviation value and generate the energizing duration and current intensity signal for adjusting the perforation diameter and jet angle required for sand control based on the deviation value, and apply the corresponding current to the annular SMA deformation layer and the SMA wires for sand control.

[0007] Preferably, as a further improvement of the present invention, the material of the annular SMA deformation layer and the material of the SMA wire are both nickel-titanium-based shape memory alloys with a deformation temperature threshold of 80℃~120℃.

[0008] Preferably, as a further improvement of the present invention, the plurality of perforations are divided into a first perforation assembly and a second perforation assembly, both of which are evenly distributed along the length direction of the perforator body, and the first perforation assembly and the second perforation assembly are arranged alternately.

[0009] Preferably, as a further improvement of the present invention, the data acquisition component includes a density sensor, a sound velocity sensor, multiple annular conductivity probes, and a temperature sensor; the density sensor is installed inside the wellbore for measuring fluid density; the sound velocity sensor is arranged in the perforation section for inverting the gas-liquid two-phase ratio through the sound wave propagation time difference; multiple annular conductivity probes are fixed on the inner wall of the wellbore and are uniformly distributed circumferentially along the inner wall of the wellbore for measuring the fluid conductivity distribution of the wellbore cross-section; the temperature sensor is set on the outer wall of the perforator for measuring the fluid temperature at the current location in real time.

[0010] Preferably, as a further improvement of the present invention, the data acquisition component further includes a pressure sensor disposed on the outer wall of the perforator and electrically connected to the control component. The pressure sensor is used to measure the fluid pressure at the current position in real time. When a sudden pressure change is detected that exceeds a set threshold, it is fed back to the control component, which directly adjusts the perforation diameter and angle.

[0011] This invention also discloses an intelligent method for controlling sand control during perforation in oil and gas wells, implemented using the aforementioned control system, comprising the following steps: Acquire dynamic data on the density, conductivity, sound velocity, and temperature of the fluid inside the wellbore; The dynamic data of density, conductivity, sound velocity and temperature of the fluid in the wellbore are input into the pre-trained neural network prediction model. The neural network prediction model associates historical time series data with the current dynamic data of density, conductivity, sound velocity and temperature of the fluid in the wellbore by setting a preset time step, determines the dynamic change law of fluid parameters and outputs the predicted value of sand production rate in the future set time period. The deviation between the predicted sand discharge rate and the preset sand control sand discharge rate threshold is obtained by subtracting the predicted value from the actual value. The deviation value is input into the PID controller. The PID controller generates the energizing duration and current intensity signals for adjusting the perforation diameter and jet angle required for sand control based on the deviation value, and applies corresponding current pulses to the SMA deformation layer and SMA filament for sand control.

[0012] Preferably, as a further improvement of the present invention, it also includes a closed-loop feedback control process, wherein the closed-loop feedback control process includes the following steps: Real-time acquisition of the density, conductivity, sound velocity change, and temperature change of the fluid after perforation adjustment; Obtain the target value of the desired fluid parameter corresponding to the current well condition, which is generated by the neural network prediction model based on historical stable sand production conditions; Calculate the deviation rate between the actual collected parameters and the target value. If the deviation rate is ≥10%, trigger the model retraining of the AI ​​decision unit, use the newly collected time series data to perform online incremental training on the neural network prediction model, and update the model weights. Based on the retrained neural network prediction model, the prediction of future fluid parameter changes is re-executed, a multidimensional compensation coefficient matrix is ​​generated, and the driving adjustment parameters are converted to perform secondary adjustment of the perforation diameter, effective perforation density, and angle.

[0013] Compared with the prior art, the beneficial effects of the present invention are: 0. By real-time monitoring of the flow state parameters of the three-phase fluids (oil, gas, and water) in the wellbore, combined with the analysis of the preset neural network prediction model in the control component, the predicted value of the future sand production rate is generated in advance. The deviation value between the predicted value of the future sand production rate and the preset sand control sand production rate threshold is obtained. The deviation value is processed by the PID controller to generate the energizing duration and current intensity signal of the perforation diameter and jet angle required for sand control. This actively adjusts the diameter and angle of the perforation, thereby realizing the leap from passive reaction to active prevention and solving the problem of lag in traditional perforation adjustment.

[0014] 2. Using shape memory alloy as the stepless adjustment actuator for aperture and angle, it has a compact structure and can control the deformation through current to adjust the perforation opening and angle. It can also adapt to environmental parameters such as temperature. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the structure of an intelligent oil and gas well perforation sand control system according to an embodiment of the present invention.

[0016] Figure 2 This is a schematic diagram of the aperture adjustment component of an intelligent oil and gas well perforation sand control system according to an embodiment of the present invention.

[0017] Figure 3 This is a schematic diagram of the angle adjustment component of an intelligent oil and gas well perforation sand control system according to an embodiment of the present invention.

[0018] Figure 4 This is a schematic flowchart of an intelligent oil and gas well perforation sand control method according to an embodiment of the present invention.

[0019] Figure 5 This is a flowchart illustrating the perforation angle adjustment control process in an intelligent oil and gas well perforation sand control method according to an embodiment of the present invention. Detailed Implementation

[0020] The following is combined with Figures 1-5The specific embodiments of the present invention will be described in detail below. In the description of the invention, it should be understood that the terms "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicating the orientation or positional relationship are based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, and are not intended to indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the present invention.

[0021] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature; in the description of the invention, unless otherwise stated, "a plurality of" means two or more.

[0022] Example 1 like Figures 1-3As shown, this embodiment of the invention provides an intelligent oil and gas wellbore perforation sand control system, including a perforator, multiple sets of orifice diameter adjustment components, multiple sets of angle adjustment components, a data acquisition component, and a control component. The perforator has multiple perforations. The multiple sets of orifice diameter adjustment components are correspondingly installed within the perforation wall near the inlet of each perforation. Each orifice diameter adjustment component includes a first annular support 71 and an annular SMA deformation layer 72. The first annular support 71 is fixed to the inner wall of the perforation, and the annular SMA deformation layer 72 is connected to the first annular support 71. On the inner ring surface, the annular SMA deformation layer 72 reduces its inner diameter when heated by electricity; multiple angle adjustment components are correspondingly arranged at the outlet of each perforation. The angle adjustment components include a second annular bracket 81, a hollow shaft 82, and two sets of SMA wires 83. The second annular bracket 81 is fixed to the base of the perforator located at the perforation outlet. The hollow shaft 82 is arranged inside the second annular bracket 81, and one end of the hollow shaft 82 is connected to the outlet of the perforation through a flexible tube. The two sets of SMA wires 83 are arranged parallel to each other at both ends of the hollow shaft 82, and are arranged along the... Hollow shafts 82 are arranged circumferentially at 180° intervals. The two ends of each set of SMA wires 83 are connected to the end of the hollow shaft 82 and the inner wall of the second annular support 81, respectively. The data acquisition component is used to collect the density, conductivity, sound velocity, and temperature parameters of the fluid in the wellbore in real time. The control component is electrically connected to the data acquisition component, the annular SMA deformation layer 72, and the SMA wires 83, respectively. The control component includes a data processing module and a PID controller. The data processing module has a preset prediction model based on a neural network. The prediction model is used to receive the density, conductivity, sound velocity, and temperature parameters of the fluid in the wellbore and output the predicted value of the sand production rate within a set time period. The data processing module subtracts the predicted value of the sand production rate from the preset sand control sand production rate threshold to obtain the deviation value. The PID controller is used to receive the deviation value and generate the energizing duration and current intensity signal for adjusting the perforation diameter and jet angle required for sand control based on the deviation value. It also applies the corresponding current to the annular SMA deformation layer 72 and the SMA wires 83 to control sand control, so as to dynamically adjust the perforation diameter and angle.

[0023] The annular SMA deformation layer 72 is equipped with a current input contact 73, which is electrically connected to the control component. Applying a current pulse triggers the deformation of the annular SMA deformation layer 72. The annular SMA deformation layer 72 is made of a nickel-titanium-based shape memory alloy, with a deformation temperature threshold of 80℃~120℃. The deformation amount exhibits a non-linear relationship with temperature. When no power is applied, the initial aperture of the perforation is 8mm~12mm. A decrease in aperture is achieved by applying a current pulse to heat the annular SMA deformation layer 72, causing it to shrink. An increase in aperture is achieved by cooling or using a reverse current pulse to restore the material to its initial state. The control component achieves bidirectional adjustment by regulating the polarity or duration of the current pulse.

[0024] Among them, the SMA wire 83 is made of nickel-titanium-based shape memory alloy, with a deformation temperature threshold of 80℃~120℃. The perforation angle is adjusted by pulling the deflection shaft 82 through the SMA wire 83 to achieve channel deflection without affecting the aperture adjustment. When the control component shortens the deformation of the two sets of SMA wires 83 by adjusting the current pulse, since the two sets of SMA wires 83 are respectively arranged in parallel at both ends of the hollow shaft 82 and are arranged 180° apart along the circumference of the hollow shaft 82, during deformation, such as Figure 5 As shown, a pulling force is applied to the left at the upper end of the hollow shaft 82 and to the right at the lower end of the hollow shaft 82, driving the hollow shaft 82 to deflect. The deflection angle is controlled by the current pulse width of the second PID controller 4, which can achieve an angle adjustment process of 0° to 45°.

[0025] As an optimization scheme, multiple perforations are divided into a first perforation assembly 2 and a second perforation assembly 5, both of which are evenly distributed along the length direction of the shape memory alloy perforator body. The first perforation assembly 2 and the second perforation assembly 5 are arranged alternately, and the matching accuracy between the perforation density and the fluid state is improved by zonal adjustment.

[0026] Specifically, the monitoring module includes a non-contact density sensor, a high-frequency sound velocity sensor, an annular conductivity probe, and a temperature sensor 6. The non-contact density sensor is installed in the middle section of the wellbore and uses a gamma ray source and detector to measure fluid density based on the gamma ray attenuation principle. The high-frequency sound velocity sensor is arranged in the perforation section to invert the gas-liquid two-phase ratio through the sound wave propagation time difference. The annular conductivity probe is used to measure the conductivity distribution of the wellbore cross-section, and the temperature sensor 6 is set on the outer wall of the perforator to measure the fluid temperature at the current location in real time.

[0027] Specifically, the data processing module includes an AI decision-making unit and an information processing unit. There are two sets of PID controllers: a first PID controller 3 and a second PID controller 4, both mounted on the perforator. The AI ​​decision-making unit constructs a prediction model based on an LSTM neural network, including a cascaded input layer, a hidden layer, and an output layer. The input layer receives time-series data on density, conductivity, sound velocity, and temperature. The hidden layer correlates historical data with the current state using a time step to predict fluid parameter changes within a set timeframe of 5 to 10 minutes. The output layer generates a predicted sand discharge velocity value for the next 5 to 10 minutes based on the prediction results from the hidden layer. The information processing unit has a preset sand discharge velocity threshold. It receives the predicted sand discharge velocity value and subtracts it from the preset threshold to obtain the deviation. The first PID controller... The controller 3 is electrically connected to each annular SMA deformation layer 72 in the information processing unit, the first perforation assembly 2, and the second perforation assembly 5. It is used to receive the deviation value output by the information processing unit, process it to generate the energizing duration and current intensity signal for adjusting the perforation diameter required for sand control, and apply the corresponding current to each annular SMA deformation layer 72 to adjust the diameter and density. The second PID controller 4 is electrically connected to each SMA wire 83 in the information processing unit, the first perforation assembly 2, and the second perforation assembly 5. It is used to receive the deviation value output by the information processing unit, process it to generate the energizing duration and current intensity signal for adjusting the perforation angle required for sand control, and apply the corresponding current to each annular SMA deformation layer 72. It controls the two diagonally arranged SMA wires to produce differentiated expansion and contraction deformation, which drives the hollow shaft 82 to deflect relative to the second annular support 81 to adjust the perforation angle.

[0028] In this embodiment, the first PID controller 3 and the second PID controller 4 are linked with the AI ​​decision unit. The first PID controller 3 is used to drive the current distribution of the SMA annular structure. By applying current pulses to cause the annular SMA deformation layer 72 to contract, the perforation diameter of the first perforation assembly 2 and the second perforation assembly 5 is adjusted. By applying current pulses to cause the annular SMA deformation layer 72 to contract to complete closure (perforation closed) or expand to the initial diameter (perforation open), the effective number of perforations is dynamically adjusted, thereby adjusting the perforation density of the first perforation assembly 2 and the second perforation assembly 5. The second PID controller 4 is used to adjust the perforation angle of the first perforation assembly 2 and the second perforation assembly 5 to achieve adaptive matching of the downhole environment.

[0029] Temperature sensor 6 is heated to the SMA phase transition point by the downhole ambient temperature or the embedded heating element. Temperature sensor 6 is used to monitor the ambient temperature in real time and feed the data back to the AI ​​decision unit. The embedded heating element is driven by a PID controller and is heated to the SMA phase transition point by current to trigger active deformation.

[0030] Temperature triggering is a passive response (e.g., when the downhole ambient temperature reaches the SMA phase transition point), while current pulse is an active control (rapidly triggering deformation through embedded heating elements). The two complement each other: temperature is used for slow environmental changes, and current is used for rapid adjustment. When the data acquisition component detects drastic fluctuations in parameters such as fluid density and conductivity (e.g., a sudden change in the gas-liquid ratio exceeding ±10%), indicating a potential sand inrush, the AI ​​decision unit immediately generates a compensation command. This command sends a current pulse through the PID controller, actively adjusting the perforation diameter or angle within seconds to suppress sand production. Based on an LSTM model, the AI ​​decision unit predicts that the fluid state will deteriorate within the next 5-10 minutes (e.g., a significant increase in sand production risk). The system will proactively initiate active control to optimize perforation parameters, achieving proactive sand control. Closed-loop calibration: When the closed-loop feedback control component detects a significant deviation (deviation rate ≥10%) between passive deformation or the actual effect of the previous adjustment and the target value, it triggers active control for precise compensation.

[0031] To complement the active control, the data acquisition component is also equipped with a pressure sensor 4, which is installed on the outer wall of the perforator and electrically connected to the control component. The pressure sensor 4 is used to monitor the fluid pressure in real time. When the pressure change exceeds the threshold (e.g., ±5%), it is fed back to the control component, which directly acts on the SMA material to trigger deformation.

[0032] Example 2 This embodiment is based on embodiment 1, such as... Figures 4-5 As shown, an intelligent method for controlling sand control during perforation in oil and gas wells is disclosed, which is implemented using the aforementioned control system and includes the following steps: S1. Acquire dynamic data on the density, conductivity, sound velocity, and temperature of the fluid inside the wellbore, and store them according to the sampling period to form continuous time series data.

[0033] S2. After preprocessing, the time series data is input into the preset neural network prediction model. The neural network prediction model associates historical time series data with the current wellbore fluid state through a preset time step to determine the dynamic change law of fluid parameters and predict the dynamic change trend of wellbore fluid density, conductivity, sound velocity and temperature in the future time period to determine the predicted value of sand production rate.

[0034] S3. Subtract the predicted sand discharge rate from the preset sand control and sand discharge rate threshold to obtain the deviation between the two.

[0035] S4. Input the deviation value into the PID controller. The PID controller generates the energizing duration and current intensity signal for adjusting the perforation diameter and jet angle required for sand control based on the deviation value, and applies the corresponding current pulse to the SMA deformation layer and SMA wire for sand control.

[0036] Furthermore, it also includes a closed-loop feedback control process, which includes the following steps: Real-time acquisition of the density, conductivity, sound velocity change, and temperature change of the fluid after perforation adjustment; Obtain the target value of the desired fluid parameter corresponding to the current well condition, which is generated by the neural network prediction model based on historical stable sand production conditions; Calculate the deviation rate between the actual collected parameters and the target value. If the deviation rate is ≥10%, trigger the model retraining of the AI ​​decision unit, use the newly collected time series data to perform online incremental training on the neural network prediction model, and update the model weights. Based on the retrained neural network prediction model, the prediction of future fluid parameter changes is re-executed, a multidimensional compensation coefficient matrix is ​​generated, and the driving adjustment parameters are converted to perform secondary adjustment of the perforation diameter, effective perforation density, and angle.

[0037] The training process of the LSTM prediction model includes: A training dataset was constructed, which included historical perforation parameters, fluid state data and corresponding sand production rate labels. Offline training was conducted using well logging data, core experimental data and historical sand production records of the target block, and a mapping relationship was established between laboratory parameters and real-time measurable parameters in the field. Sequence features are extracted using a time-step sliding window, with a window length of 10 to 15 minutes. The mean squared error loss function and the Adam optimizer are used to iteratively update the model parameters.

[0038] In a field test at an oilfield wellbore, a comparison was made with a traditional perforation system: Sand output: 4.2% (volume fraction) for the system of this invention, and 7.8% for the traditional system; Adjustable response time: 28 seconds for this invention, 150 seconds for the conventional system; Three-phase separation efficiency: 92% for this invention, compared to 78% for the traditional system.

[0039] In summary, this invention, through the synergy of AI prediction and SMA self-response, reduces sand output by ≥40% (compared to ≤15% with traditional methods), thereby improving sand control efficiency. The response time for perforation parameter adjustment is ≤30 seconds (compared to ≥2 minutes with traditional mechanical adjustment), thus optimizing response speed.

[0040] The above-disclosed embodiments are merely preferred embodiments of the present invention. However, the embodiments of the present invention are not limited thereto, and any variations that can be conceived by those skilled in the art should fall within the protection scope of the present invention.

Claims

1. An intelligent oil and gas wellbore perforation sand control system, characterized in that, include: A perforator with multiple perforations; Multiple orifice diameter adjustment components are respectively set in the orifice wall near the inlet of each of the perforations. Each orifice diameter adjustment component includes a first annular support and an annular SMA deformation layer. The first annular support is fixed on the inner wall of the perforation. The annular SMA deformation layer is connected to the inner ring surface of the first annular support. The annular SMA deformation layer reduces its inner diameter when heated by electricity. Multiple angle adjustment components are arranged one-to-one at the outlet of each of the perforations. Each angle adjustment component includes a second annular bracket, a hollow shaft, and two sets of SMA wires. The second annular bracket is fixed to the base of the perforator located at the outlet of the perforation. The hollow shaft is arranged inside the second annular bracket, and one end of the hollow shaft is connected to the outlet of the perforation through a flexible tube. The two sets of SMA wires are arranged parallel to each other at both ends of the hollow shaft and are arranged 180° apart along the circumference of the hollow shaft. The two ends of each set of SMA wires are connected to the end of the hollow shaft and the inner wall of the second annular bracket, respectively. The data acquisition component is used to collect parameters such as density, conductivity, sound velocity, and temperature of the fluid inside the wellbore. The control component is electrically connected to the data acquisition component, the annular SMA deformation layer, and the SMA wire, respectively. The control component includes a data processing module and a PID controller. The data processing module has a preset prediction model based on a neural network. The prediction model is used to receive the density, conductivity, sound velocity, and temperature parameters of the fluid in the wellbore and output the predicted value of the sand production rate within a set time period. The data processing module subtracts the predicted sand production rate from the preset sand control sand production rate threshold to obtain the deviation value. The PID controller is used to receive the deviation value and generate the energizing duration and current intensity signals for adjusting the perforation diameter and jet angle required for sand control based on the deviation value, and apply the corresponding current to the annular SMA deformation layer and the SMA wire for sand control.

2. The intelligent oil and gas wellbore perforation sand control system according to claim 1, characterized in that, The material of the annular SMA deformation layer and the SMA wire are both nickel-titanium based shape memory alloys with a deformation temperature threshold of 80℃~120℃.

3. The intelligent oil and gas wellbore perforation sand control system according to claim 1, characterized in that, The plurality of perforations are divided into a first perforation assembly and a second perforation assembly, both of which are evenly distributed along the length of the perforator body, and the first perforation assembly and the second perforation assembly are arranged alternately.

4. The intelligent oil and gas wellbore perforation sand control system according to claim 1, characterized in that, The data acquisition component includes: A density sensor, installed inside the wellbore, is used to measure fluid density; A sound velocity sensor is placed in the perforation section to invert the gas-liquid two-phase ratio by the time difference of sound wave propagation. Multiple annular conductivity probes are fixed on the inner wall of the wellbore and are evenly distributed along the circumference of the inner wall of the wellbore to measure the fluid conductivity distribution of the cross-section of the wellbore. A temperature sensor is installed on the outer wall of the perforator to measure the fluid temperature at the current location in real time.

5. The intelligent oil and gas wellbore perforation sand control system according to claim 4, characterized in that, The data acquisition component also includes a pressure sensor, which is installed on the outer wall of the perforator and electrically connected to the control component. The pressure sensor is used to measure the fluid pressure at the current position in real time. When the detected pressure change exceeds a set threshold, it is fed back to the control component, which directly adjusts the perforation diameter and angle.

6. An intelligent method for controlling sand control during perforation in oil and gas wells, implemented using the control system described in any one of claims 1 to 5, characterized in that, Includes the following steps: Acquire dynamic data on the density, conductivity, sound velocity, and temperature of the fluid inside the wellbore; The dynamic data of density, conductivity, sound velocity and temperature of the fluid in the wellbore are input into the pre-trained neural network prediction model. The neural network prediction model associates historical time series data with the current dynamic data of density, conductivity, sound velocity and temperature of the fluid in the wellbore by setting a preset time step, determines the dynamic change law of fluid parameters and outputs the predicted value of sand production rate in the future set time period. The deviation between the predicted sand discharge rate and the preset sand control sand discharge rate threshold is obtained by subtracting the predicted value from the actual value. The deviation value is input into the PID controller. The PID controller generates the energizing duration and current intensity signals for adjusting the perforation diameter and jet angle required for sand control based on the deviation value, and applies corresponding current pulses to the SMA deformation layer and SMA filament for sand control.

7. The intelligent oil and gas wellbore perforation sand control method according to claim 6, characterized in that, It also includes a closed-loop feedback control process, which includes the following steps: Real-time acquisition of the density, conductivity, sound velocity change, and temperature change of the fluid after perforation adjustment; Obtain the target value of the desired fluid parameter corresponding to the current well condition, which is generated by the neural network prediction model based on historical stable sand production conditions; Calculate the deviation rate between the actual collected parameters and the target value. If the deviation rate is ≥10%, trigger the retraining of the neural network prediction model. Use the newly collected time series data to perform online incremental training on the neural network prediction model and update the model weights. Based on the retrained neural network prediction model, the prediction of future sand production speed is re-executed to achieve secondary adjustment of the perforation diameter and angle.

Images