Injection-production pipe suspension height adjusting method and device and injection-production pipe

Abstract

The embodiment of the invention provides an injection-production pipe suspension height adjusting method and device and an injection-production pipe, and relates to the technical field of salt cavern compression energy storage. The injection-production pipe suspension height adjusting method comprises the steps that first displacement data of a telescopic pipe and torque data of a screw mechanism at the current moment are obtained; inputting the first displacement data and the torque data into a displacement correction model to obtain second displacement data of the telescopic pipe; based on the pressure variation in the salt cavern at the current moment and the target suspension height of the injection-production pipe, determining trajectory data of the telescopic pipe in the prediction time domain from the current moment; according to the second displacement data and the track data, a control instruction of the screw mechanism at the current moment is generated through a screw transmission model, and the screw transmission model is used for representing the correlation between the screw transmission characteristic of the screw mechanism and the displacement of the telescopic pipe; and based on the control instruction, controlling a screw mechanism to adjust the suspension height of the telescopic pipe.

Description

TECHNICAL FIELD

[0001] Embodiments of the present specification relate to the technical field of salt cavern compressed energy storage, in particular to an injection-production pipe suspension height adjustment method and device and injection-production pipe. BACKGROUND

[0002] As the core component connecting the above-ground equipment and the underground gas storage, the injection-production pipe has always been developed around the improvement of high-pressure sealing, dynamic adaptability and corrosion resistance to meet the special working condition requirements of salt cavern energy storage. In the early development of salt cavern compressed energy storage, the injection-production pipe technology was relatively simple, mainly using traditional pipeline transportation technology to inject compressed air into the salt cavern and extract it. Such injection-production pipes are usually of fixed length and mainly made of metal, although they have certain pressure resistance and corrosion resistance, but they have obvious limitations in adapting to different salt cavern depths, complex working conditions and improving injection-production efficiency.

[0003] The commonly used injection-production pipe suspension height is usually fixed. During the process of releasing high-pressure air for power generation, due to the continuous change of the gas pressure inside the salt cavern, the fixed suspension height cannot accurately match the dynamic characteristics of the gas pressure in real time, resulting in a decrease in gas extraction efficiency. SUMMARY

[0004] The purpose of the embodiments of the present specification is to provide an injection-production pipe suspension height adjustment method and device and injection-production pipe to overcome the problem that the fixed suspension height in the prior art cannot accurately match the dynamic characteristics of the gas pressure in real time, resulting in a decrease in gas extraction efficiency.

[0005] To solve the above technical problems, the specific technical solutions of the embodiments of the present specification are as follows:

[0006] On the one hand, the embodiments of the present specification provide an injection-production pipe suspension height adjustment method,

[0007] The method comprises:

[0008] obtaining first displacement data of the telescopic pipe and torque data of the screw mechanism at the current time;

[0009] inputting the first displacement data and the torque data into a displacement correction model to obtain second displacement data of the telescopic pipe;

[0010] based on the pressure change amount in the salt cavern at the current time and the target suspension height of the injection-production pipe, determining trajectory data of the telescopic pipe in a prediction time domain from the current time;

[0011] generating a control instruction of the screw mechanism at the current time using a screw transmission model according to the second displacement data and the trajectory data, the screw transmission model being used to represent the correlation between the screw thread transmission characteristics of the screw mechanism and the displacement of the telescopic pipe;

[0012] Based on the control instruction, the screw mechanism is controlled to adjust the overhanging height of the telescopic pipe.

[0013] Further, the injection-production pipe further comprises a plurality of displacement sensors uniformly arranged on the outer wall of the telescopic pipe in the axial direction and a torque sensor arranged in the screw lifting mechanism.

[0014] The first displacement data of the telescopic pipe and the torque data of the screw mechanism at the current time are obtained, comprising:

[0015] The first displacement data of the telescopic pipe from the last historical time to the current time measured by the plurality of displacement sensors is obtained.

[0016] The torque data of the screw mechanism at the current time measured by the torque sensor is obtained.

[0017] Further, the injection-production pipe further comprises a plurality of inclination sensors uniformly arranged on the pipe opening of the telescopic pipe.

[0018] The displacement correction model is coupled with an extended Kalman displacement model, a strong tracking displacement model and an unscented Kalman displacement model, and the extended Kalman displacement model, the strong tracking displacement model and the unscented Kalman displacement model are respectively used to process linear correlation characteristics, nonlinear correlation characteristics and mutation correlation characteristics between the torque of the screw mechanism and the displacement of the telescopic pipe.

[0019] The first displacement data and the torque data are input into the displacement correction model to obtain the second displacement data of the telescopic pipe, comprising:

[0020] The inclination change amount of the telescopic pipe from the last historical time to the current time measured by the plurality of inclination sensors is obtained.

[0021] The first displacement data, the torque data and the inclination change amount are respectively input into the extended Kalman displacement model, the strong tracking displacement model and the unscented Kalman displacement model to obtain third displacement data, fourth displacement data and fifth displacement data.

[0022] The third displacement data, the fourth displacement data and the fifth displacement data are fused to obtain the second displacement data.

[0023] Further, the fusion of the third displacement data, the fourth displacement data and the fifth displacement data to obtain the second displacement data comprises:

[0024] Based on the third, fourth, and fifth displacement data, the confidence levels of the extended Kalman displacement model, the strong tracking displacement model, and the unscented Kalman displacement model are determined respectively; wherein, the confidence level of each model is determined based on at least the following indicators: the whitening degree of the model's innovation sequence, the consistency between the model's displacement data and the displacement data of other models, the convergence speed of each model, and the computational complexity of each model.

[0025] The confidence scores of the extended Kalman displacement model, the strong tracking displacement model, and the unscented Kalman displacement model are coupled to obtain the confidence score of the displacement correction model.

[0026] Maximize the confidence of the displacement correction model to obtain the optimal confidence of the extended Kalman displacement model, the strong tracking displacement model, and the unscented Kalman displacement model at the current time.

[0027] Based on the optimal confidence levels of the extended Kalman displacement model, the strong tracking displacement model, and the unscented Kalman displacement model at the current moment, the third displacement data, the fourth displacement data, and the fifth displacement data are fused to obtain the second displacement data.

[0028] Furthermore, the injection and extraction tube also includes multiple pressure sensors evenly distributed at the inlet of the telescopic tube;

[0029] The method for determining the trajectory data of the telescopic tube in the prediction time domain from the current moment, based on the pressure change within the salt cavern at the current moment and the target suspension height of the injection-production tube, includes:

[0030] Obtain the expansion and contraction amount of the telescopic tube at the previous historical moment;

[0031] Based on the expansion and contraction amount and the second displacement data, determine the expansion and contraction amount of the telescopic tube at the current moment;

[0032] Determine the suspension height of the telescopic pipe at the current moment based on the amount of expansion and contraction.

[0033] Acquire the pressure change within the salt cave from the previous historical moment to the current moment, as measured by the plurality of pressure sensors;

[0034] Based on the pressure change, the current suspension height of the telescopic pipe, and the target suspension height of the injection and extraction pipe, trajectory data of the telescopic pipe within the prediction time domain starting from the current moment is generated.

[0035] Further, the step of generating trajectory data of the telescopic pipe within the prediction time domain from the current moment based on the pressure change, the current suspension height of the telescopic pipe, and the target suspension height of the injection / production pipe includes:

[0036] Based on the pressure change and the current suspension height of the telescopic pipe, the latent state probability distribution of the fluid flow field inside the salt cave is inferred through a preset salt cave flow field model; wherein, the latent state probability distribution includes at least the probability distribution of the intensity and position of the flow field eddy, the probability distribution of the sediment suspension state, the probability distribution of the flow stability, and the probability distribution of the fluid velocity and direction at the inlet of the telescopic pipe.

[0037] Using the hidden state probability distribution, the current suspension height of the telescopic tube, and the target suspension height as constraints, a Markov decision process spanning multiple moments in the prediction time domain is constructed.

[0038] Solve the Markov decision process to generate trajectory data at multiple times within the prediction time domain.

[0039] Furthermore, the step of generating control commands for the helical mechanism at the current moment using the helical transmission model based on the second displacement data and trajectory data includes:

[0040] Based on the second displacement data and trajectory data, a sequence of control commands for the helical mechanism at multiple moments within the control time domain starting from the current moment is generated using a helical transmission model; wherein the length of the control time domain is less than the length of the prediction time domain.

[0041] Furthermore, the spiral mechanism includes a rotary lift and a threaded lifting rod; the rotary lift is used to drive the threaded lifting rod to rotate, so that the threaded lifting rod moves up and down in a spiral manner; the threaded lifting rod is used to push the telescopic tube to move axially; the injection and extraction tube also includes a fixing ring welded to the outer wall of the main tube, a limiter for bolting the rotary lift to the fixing ring, and a sliding mechanism; the sliding mechanism includes multiple grooved slide rails evenly arranged axially on the inner wall of the telescopic tube, and multiple pulleys arranged in the grooved slide rails;

[0042] The step of controlling the spiral mechanism to adjust the suspension height of the telescopic tube based on the control command includes:

[0043] Based on the control command, the rotation speed and direction of the rotary lift are controlled to drive the threaded lifting rod to rotate accordingly, thereby causing the telescopic tube to move axially under the push of the threaded lifting rod.

[0044] On the other hand, an adjustment device for the suspended height of an injection-production pipe is provided. The injection-production pipe includes a main pipe fixed in a salt cavern injection-production well, a telescopic pipe coaxially sleeved on the outer wall of the main pipe, and a spiral mechanism for driving the telescopic pipe to move along the axial direction of the main pipe.

[0045] The device includes:

[0046] The acquisition module is used to acquire the first displacement data of the telescopic tube and the torque data of the screw mechanism at the current moment;

[0047] The correction module is used to input the first displacement data and torque data into the displacement correction model to obtain the second displacement data of the telescopic tube;

[0048] The determination module is used to determine the trajectory data of the telescopic tube in the prediction time domain from the current moment, based on the pressure change in the salt cavern and the target suspension height of the injection and production tube at the current moment.

[0049] The generation module is used to generate the control command of the screw mechanism at the current moment based on the second displacement data and trajectory data using the screw transmission model. The screw transmission model is used to characterize the relationship between the screw mechanism's thread transmission characteristics and the displacement of the telescopic tube.

[0050] The control module is used to control the spiral mechanism to adjust the suspension height of the telescopic tube based on the control command.

[0051] Furthermore, embodiments of this specification also provide an injection and production pipe, including a main pipe fixed in a salt cavern injection and production well, a telescopic pipe coaxially sleeved on the outer wall of the main pipe, a spiral mechanism for driving the telescopic pipe to move axially along the main pipe, and a controller; the controller is used to execute the above-mentioned method for adjusting the suspension height of the injection and production pipe.

[0052] As can be seen from the technical solutions provided in the embodiments of this specification above, these embodiments can obtain the first displacement data of the telescopic tube and the torque data of the screw mechanism at the current moment; input the first displacement data and torque data into the displacement correction model to obtain the second displacement data of the telescopic tube; based on the pressure change in the salt cavern at the current moment and the target suspension height of the injection-production tube, determine the trajectory data of the telescopic tube in the prediction time domain from the current moment; according to the second displacement data and trajectory data, use the screw transmission model to generate the control command of the screw mechanism at the current moment, the screw transmission model is used to characterize the correlation between the screw mechanism's thread transmission characteristics and the telescopic tube's displacement; based on the control command, control the screw mechanism to adjust the suspension height of the telescopic tube. By establishing a displacement correction model and fusing multi-source data of displacement and torque to perform real-time correction and optimization of directly measured displacement data, the problem of single sensor being easily interfered with and having insufficient measurement accuracy under complex working conditions in the salt cavern is effectively solved, providing reliable state feedback for precise control of the suspension height. Furthermore, based on real-time pressure changes within the salt cavern and the target suspension height, displacement trajectory data can be generated in the prediction time domain, enabling proactive prediction of operating condition trends. This allows for advance planning of suspension height adjustment paths based on dynamic gas pressure characteristics, effectively overcoming control lag and improving injection-production efficiency. By constructing a helical transmission model that couples the thread transmission characteristics of the helical mechanism with the displacement of the telescopic tube, the mechanical characteristics of the transmission system are fully considered during the control command generation stage. This allows the control commands to effectively compensate for nonlinear factors such as transmission backlash and elastic deformation, significantly improving the accuracy and stability of displacement control. By sensing pressure changes in real time and generating predicted trajectories, the system can quickly adapt to dynamic fluctuations in gas pressure within the salt cavern. Combined with the real-time correction function of the displacement correction model, the adaptability and robustness under complex salt cavern conditions are comprehensively enhanced. In addition, precise displacement control avoids ineffective movement and over-adjustment of the telescopic tube, reducing mechanical wear and energy loss. Trajectory planning can also effectively reduce mechanical impact, thereby extending equipment lifespan. Attached Figure Description

[0053] To more clearly illustrate the technical solutions in the embodiments or prior art of this specification, the accompanying drawings used in the description of the embodiments or prior art will be briefly introduced below.

[0054] Figure 1 This is a schematic diagram of the structure of an injection-production pipe provided in the embodiments of this specification;

[0055] Figure 2 This is a schematic diagram of an injection / production pipe being run into a well, as provided in the embodiments of this specification.

[0056] Figure 3 This is a three-dimensional structure of an injection / collection tube provided in the embodiments of this specification;

[0057] Figure 4This is a three-dimensional structural diagram of an injection-production pipe running into a well, provided in an embodiment of this specification.

[0058] Figure 5 This is a partial three-dimensional structural diagram of an injection-production pipe being run into a well, provided in an embodiment of this specification.

[0059] Figure 6 This is a schematic diagram of the connection between a rotary lift and a spiral lifting rod provided in the embodiments of this specification;

[0060] Figure 7 This is a schematic diagram of the connection between the internal turbine and the spiral lifting rod of a rotary lift provided in the embodiments of this specification;

[0061] Figure 8 This is a flowchart illustrating a method for adjusting the suspension height of an injection / production pipe, as provided in the embodiments of this specification.

[0062] Figure 9 This is a schematic diagram of the structure of an adjustment device for the suspended height of an injection / production pipe provided in the embodiments of this specification.

[0063] The reference numerals in the above figures are as follows:

[0064] 1. Supervisor;

[0065] 2. Telescopic pipe;

[0066] 3. Grooved slide rail;

[0067] 4. Pulleys;

[0068] 5. Rotary elevator;

[0069] 6. Limit switch;

[0070] 7. Threaded lifting rod;

[0071] 8. Retaining ring;

[0072] 9. Fix the locking ring;

[0073] 10. Displacement sensor;

[0074] 11. Pressure sensor;

[0075] 12. Packer;

[0076] 13. Worm gear;

[0077] 14. Bearings;

[0078] 15. Gear rack;

[0079] 16. Turbine;

[0080] 17. Servo motor;

[0081] 18. Torque sensor;

[0082] 19. Tilt sensor. Detailed Implementation

[0083] The technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this specification, and not all embodiments. Based on the embodiments in this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this specification.

[0084] It should be noted that the terms "first," "second," etc., used in this specification and the aforementioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, apparatus, product, or device that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0085] In some embodiments, during the gas extraction power generation process, the high-pressure air within the salt cavern is released, resulting in a decrease in the internal gas pressure. According to fluid mechanics, at the same nozzle location, a decrease in gas pressure means a decrease in gas density, which in turn reduces the pressure difference driving the gas flow towards the nozzle. This leads to a decrease in gas velocity, a reduction in the amount of gas extracted per unit time, and a decrease in power generation. For the gas injection energy storage process, the opposite applies, but an optimal flow field also exists.

[0086] In some embodiments, the bottom of the salt cavern typically contains residual brine and sediment sloughed off from the cavern walls. During gas extraction, the flow field near the inlet changes as the gas pressure decreases and the flow rate varies. If the flow rate is too low, the gas may not be able to effectively purge away liquids and fine particles near the inlet, and may instead easily entrain droplets and sediment into the pipeline. This not only corrodes and wears the pipeline and equipment, but also directly reduces the quality and energy of the output air.

[0087] In some embodiments, when gas extraction causes a decrease in intracavitary pressure, extending the telescopic tube downwards reduces the suspension height, allowing the tube opening to be closer to the bottom of the salt cavern. According to Bernoulli's principle, in areas with smaller flow cross-sections, the fluid accelerates to maintain mass conservation. Simultaneously, the static pressure at the tube opening is relatively higher. By reducing the suspension height, the gas velocity and pressure at the tube opening can be locally increased under lower overall cavity pressure, effectively overcoming the velocity reduction problem caused by low pressure and maintaining high gas extraction efficiency. Real-time adjustment of the suspension height ensures the tube opening remains within a safe and efficient range. Specifically, in the high-pressure phase, the suspension height can be appropriately increased to avoid excessive scouring of the bottom sediment. In the low-pressure phase, precisely reducing the suspension height increases the flow velocity while avoiding ineffective suction zones formed under low pressure due to excessive height, thus significantly reducing the intake of liquid and sediment.

[0088] Figure 1 This is a structural diagram of an injection-production pipe provided in the embodiments of this specification.

[0089] In some embodiments, the injection-production pipe includes a main pipe 1 fixed in a salt cavern injection-production well, a telescopic pipe 2 coaxially sleeved on the outer wall of the main pipe, and a screw mechanism for driving the telescopic pipe 2 to move axially along the main pipe 1.

[0090] Reference Figure 1 , Figure 2 , Figure 3 , Figure 4 , Figure 5 , Figure 6 and Figure 7 As shown, the main pipe 1 can serve as a basic support structure. It can be connected to the ground equipment through the wellhead fixing device and remain stationary during the well running process, providing structural stability for the entire injection and production pipeline system.

[0091] The telescopic tube 2 is coaxially sleeved outside the main tube 1, forming a sliding fit. Driven by a screw mechanism, the telescopic tube 2 can move up and down along the axial direction of the main tube 1, thereby dynamically adjusting the suspension height of the injection / production tube inlet within the salt cavern. This coaxial sleeve design ensures the guiding accuracy of the telescopic tube 2 during movement, effectively preventing radial swaying and deflection. The axial movement of the telescopic tube 2 allows for dynamic adjustment of the injection / production tube's suspension height, enabling real-time optimization of injection / production efficiency based on changes in gas pressure within the salt cavern. Furthermore, the coaxial sleeve structure ensures good guidance, improving the system's stability and reliability under high pressure and high flow rate conditions.

[0092] The screw mechanism may include a drive motor and a threaded transmission assembly. Its specific operation is as follows: the drive motor receives commands from the control system, generates rotational motion, and transmits torque to the threaded transmission assembly via a coupling. This assembly converts the motor's rotational motion into precise linear displacement, thereby pushing or pulling the telescopic tube 2 along the axial direction of the main pipe 1. The screw mechanism utilizes the self-locking characteristic of the threaded transmission to achieve reliable locking at any position, ensuring the telescopic tube 2 remains stable at the target suspension height. The screw mechanism provides high-precision displacement control and reliable position holding capability, ensuring the accuracy and stability of the suspension height adjustment.

[0093] In some embodiments, the injection and extraction tube further includes a plurality of displacement sensors 10 uniformly arranged along the axial direction on the outer wall of the telescopic tube 2, and a torque sensor 18 disposed in the spiral lifting mechanism.

[0094] Reference Figure 1 As shown, the displacement sensor 10 can adopt a high-precision magnetic scale structure, evenly distributed along the axial direction of the outer wall of the telescopic pipe. It can detect the linear displacement of the telescopic pipe relative to the main pipe and convert the mechanical position change into a corresponding electrical signal output. Each sensor independently measures local displacement, and the readings from all sensors can be combined using a data fusion algorithm to calculate the overall accurate displacement and attitude information of the telescopic pipe. This multi-point distributed measurement design effectively overcomes errors that may occur in single-point measurements due to pipe bending or vibration. The multi-point layout of the displacement sensor 10 provides redundant measurement channels, ensuring reliable system operation even when a single sensor fails, significantly improving the measurement's fault tolerance and system reliability. Through multi-sensor data fusion, the system can identify the bending state of the telescopic pipe and compensate for and correct the displacement measurement values, improving the displacement estimation accuracy to the millimeter level.

[0095] The torque sensor 18 can be embedded in the drive transmission chain of the screw jack mechanism, employing the strain gauge measurement principle. It accurately measures the torque output of the motor by detecting the minute deformation of the drive shaft under torque in real time. This sensor converts mechanical stress into an electrical signal and transmits it to the control system for real-time processing. The real-time torque data provided by the torque sensor 18 enables the control system to monitor the load status of the screw mechanism and take timely protective measures to prevent equipment damage in the event of overload, jamming, or other abnormal operating conditions.

[0096] The joint analysis of displacement and torque signals can provide a more comprehensive information basis for system status assessment, enabling the controller to optimize control parameters according to the actual load conditions, achieve more accurate displacement tracking control, help identify potential faults such as mechanical wear and insufficient lubrication in the early stage, and reduce the risk of unplanned downtime.

[0097] In some embodiments, the injection tube further includes a plurality of tilt sensors 19 evenly distributed at the inlet of the telescopic tube 2.

[0098] Reference Figure 1 As shown, the tilt sensor 19 can be an inertial measurement unit, evenly distributed in a ring around the end face of the pipe opening. It can measure the tilt angle and direction of the pipe opening relative to the horizontal plane in real time by detecting the component of gravitational acceleration on the sensor's sensitive axis. When the telescopic pipe bends or is subjected to uneven loads, the data from each sensor can be fused and processed to accurately calculate the spatial attitude angle of the pipe opening, providing an attitude compensation reference for displacement measurement.

[0099] The attitude data provided by the tilt sensor 19 can effectively compensate for displacement measurement errors caused by pipe bending, improving displacement estimation accuracy to a higher level. Through the collaborative measurement of multiple tilt sensors, the bending modes of the pipe can be identified.

[0100] In some embodiments, the injection tube further includes a plurality of pressure sensors 11 evenly distributed at the inlet of the telescopic tube 2.

[0101] Reference Figure 1 As shown, the pressure sensor 11 can be a high-frequency piezoresistive sensing element, also arranged in a ring around the pipe opening. It can sense the pressure exerted by the fluid medium on the sensitive diaphragm and convert the pressure signal into a corresponding electrical signal output. The arrangement of multiple sensors can capture the spatial pressure distribution around the pipe opening, providing data support for analyzing flow field characteristics.

[0102] The ring array arrangement of pressure sensors 11 accurately reflects the pressure distribution around the pipe opening, providing a direct basis for assessing flow field uniformity. High-frequency pressure measurement can capture the pressure pulsation characteristics within the salt cavern, providing a data foundation for eddy current identification and vibration early warning. Furthermore, the data cross-verification mechanism among multiple sensors enhances reliability; when one sensor malfunctions, cross-verification and compensation can be performed using data from adjacent sensors.

[0103] In some embodiments, the helical mechanism includes a rotary lift 5 and a threaded lifting rod 7; the rotary lift 5 is used to drive the threaded lifting rod 7 to rotate, so that the threaded lifting rod 7 moves up and down in a helical manner; the threaded lifting rod 7 is used to push the telescopic tube 2 to move axially.

[0104] Reference Figure 1 , Figure 6 and Figure 7As shown, the helical lifting mechanism can include a rotary lifting machine 5 and a threaded lifting rod 7, forming the core drive system of the injection-production pipe. The rotary lifting machine 5 can use a servo motor with a precision reducer to directly drive the threaded lifting rod 7 to rotate via the output shaft. The threaded lifting rod 7 can be made of high-strength alloy steel with precision trapezoidal threads machined on its surface. It forms a helical transmission pair with the nut inside the rotary lifting machine 5, converting the rotational motion into precise linear displacement. This helical transmission mechanism has a self-locking characteristic, which can reliably lock at any position, ensuring that the telescopic pipe remains stable at the target height. The helical transmission mechanism can provide precise displacement control and reliable position holding capability, meeting the precise adjustment requirements of the injection-production pipe's suspension height for salt cavern energy storage.

[0105] In some embodiments, the injection and extraction tube further includes a fixing ring 8 welded to the outer wall of the main tube 1, a limiter 7 for bolting the rotary elevator 5 to the fixing ring 8, and a sliding mechanism; the sliding mechanism includes a plurality of grooved slide rails 3 uniformly arranged axially on the inner wall of the telescopic tube, and a plurality of pulleys 4 arranged in the grooved slide rails 3.

[0106] Reference Figure 1 , Figure 6 and Figure 7 As shown, the fixed ring 8, as the main load-bearing structure, can transfer all the loads of the screw lifting mechanism to the main pipe.

[0107] The limiter 6 can be connected with high-strength bolts, which not only ensures the reliable connection between the rotating lift 5 and the fixed ring 8, but also prevents radial displacement of the lift during operation by precision machining of the mating surfaces.

[0108] The combination of the retaining ring and the limiter design ensures sufficient structural strength while providing convenient installation and maintenance conditions.

[0109] The sliding mechanism can consist of multiple grooved slide rails 3 evenly arranged axially along the inner wall of the telescopic tube and multiple pulleys 4 embedded in the slide rails. The grooved slide rails 3 can be made of wear-resistant alloy material to provide a precise guide path for the pulleys.

[0110] Pulley 4 can employ a bearing structure coated with engineering plastics, reducing the coefficient of friction while ensuring load-bearing capacity. Each pulley is fixed by a bracket and rolls smoothly along the slide rail as the telescopic tube moves.

[0111] This multi-point symmetrical pulley-rail system effectively limits the radial displacement of the telescopic tube, ensuring that it can maintain stable coaxial movement under the action of high-pressure airflow.

[0112] In some embodiments, the injection tube further includes a retaining ring 9.

[0113] Reference Figure 1As shown, the fixing ring 9 can be positioned at the termination point of the relative movement between the telescopic tube 2 and the main tube 1. The fixing ring 9 can be made of high-strength alloy steel and has an internal bidirectional locking tooth structure. When the telescopic tube moves to the predetermined position, the locking teeth precisely engage with the mating groove on the outer wall of the main tube via hydraulic drive or mechanical transmission, forming a rigid connection. The fixing ring 9 provides additional mechanical locking protection after the injection / production tube has completed its suspension height adjustment, preventing tubing movement that may occur during severe fluctuations in salt cavern pressure.

[0114] In some embodiments, the injection-production pipe further includes a packer 12 for preventing air leakage from the gaps in the pipe wall.

[0115] Reference Figure 2 As shown, the injection-production tubing may also include a packer 12 disposed between the main pipe 1 and the wellbore wall of the salt cavern. The packer 12 may be a hydraulically driven structure, consisting of a corrosion-resistant elastic sealing element, a support frame, and a hydraulic activation system. After the injection-production tubing is lowered to the target depth, hydraulic pressure is supplied to the packer through a surface pump station, causing the elastic sealing element to expand radially and tightly fit against the wellbore wall, forming a reliable annular seal. Its function is to effectively isolate the annular gap between the outer wall of the main pipe and the wellbore wall, preventing high-pressure gas leakage, while simultaneously preventing downhole brine and sediment from entering the working area of ​​the tubing string.

[0116] In some embodiments, the rotary lifting machine 5 includes a worm gear 13, a bearing 14, a rack 15 on the worm gear 13, a turbine 16, and a servo motor 17; when the worm gear 13 rotates in the forward direction, it drives the turbine 16 to rotate, causing the threaded lifting rod 7 to rotate clockwise and move downward, thereby extending the telescopic tube 2; when the worm gear 13 rotates in the reverse direction, it drives the turbine 16 to rotate, causing the threaded lifting rod 7 to rotate counterclockwise and move upward, thereby shortening the telescopic tube 2.

[0117] Reference Figure 1 , Figure 6 and Figure 7 As shown, the servo motor 17 can be used as a power source and is directly connected to the worm gear 13 through the output shaft to provide precise and controllable rotational motion.

[0118] The worm gear 13 can be made of high-strength alloy steel, and its surface helical rack 15 precisely meshes with the helical teeth of the turbine 16 to form a reliable power transmission path.

[0119] The bearings 14 can be high-precision angular contact ball bearings, which are respectively supported at both ends of the worm 13, ensuring the precise rotational positioning of the worm and being able to withstand the axial and radial loads generated during transmission.

[0120] The rotary lifting mechanism 5 operates as follows: When the servo motor 17 receives a forward rotation command, it drives the worm gear 13 to rotate clockwise. Through the meshing of the worm rack 15 and the worm wheel 16, the motion is transmitted to the worm wheel 16, causing the threaded lifting rod 7, which is coaxially connected to the worm wheel 16, to rotate synchronously. Due to the conversion effect of the threaded pair, the threaded lifting rod 7 generates a downward linear displacement while rotating, pushing the telescopic tube 2 to extend downward. Conversely, when the servo motor 17 rotates in the opposite direction, the worm gear 13 rotates counterclockwise, driving the threaded lifting rod 7 to rotate in the opposite direction through the worm wheel 16, generating an upward linear displacement, causing the telescopic tube 2 to shorten upward.

[0121] In some embodiments, the injection-production tube may include a controller. The controller may include a microcontroller unit (MCU) or a central processing unit (CPU). Of course, the controller may also include other devices capable of control functions, such as desktop computers, laptops, or other computer equipment. The controller can control the components by sending control commands to the components in the compressed energy storage and cyclic testing system. For example, it can acquire the first displacement data of the telescopic tube and the torque data of the helical mechanism at the current moment; input the first displacement data and torque data into a displacement correction model to obtain the second displacement data of the telescopic tube; based on the pressure change in the salt cavern at the current moment and the target suspension height of the injection-production tube, determine the trajectory data of the telescopic tube in the prediction time domain from the current moment; based on the second displacement data and trajectory data, use a helical transmission model to generate control commands for the helical mechanism at the current moment, whereby the helical transmission model characterizes the correlation between the threaded transmission characteristics of the helical mechanism and the displacement of the telescopic tube; based on the control commands, control the helical mechanism to adjust the suspension height of the telescopic tube.

[0122] As can be seen from the above embodiments of the injection-production pipe provided in this specification, by coaxially sleeved the telescopic pipe outside the main pipe, a stable double-layer sleeve guiding structure is constructed. This design provides a precise mechanical guiding reference for the axial movement of the telescopic pipe, effectively suppressing radial sway and deflection, ensuring the motion accuracy and stability during the suspension height adjustment process, and laying a solid mechanical foundation for the system to achieve precise control. This telescopic injection-production pipe design enables the system to dynamically adjust the suspension height of the injection-production pipe according to the real-time changes in gas pressure within the salt cavern during gas extraction, effectively optimizing the gas flow field characteristics at the pipe inlet, promoting efficient gas entry into the pipeline, thereby significantly improving power generation efficiency and the overall system operating efficiency.

[0123] In some embodiments, in salt cavern scenarios, the machining errors, wear, and transmission backlash of mechanical transmission components such as the threaded lifting rod 7, turbine 16, and worm gear 13 are significantly amplified under frequent reversals and heavy-load conditions, resulting in nonlinear positioning deviations. Secondly, the tens-of-meters-long telescopic pipe 2 and the slender threaded lifting rod 7 undergo elastic deformation under load, leading to a non-negligible difference between the control command displacement and the actual displacement. Furthermore, the complex environmental factors inside the salt cavern, including high-speed airflow impact, frictional resistance between the pipe and the well wall, and the viscosity of the brine medium, all exert unpredictable random disturbances on the telescopic pipe 2, further affecting its positioning stability.

[0124] In this context, relying solely on displacement sensing devices to directly measure the displacement data of the telescopic pipe will not effectively compensate for the aforementioned systematic errors and random interference, resulting in a significant deviation between the measurement results and the actual displacement, which cannot meet the precise control requirements of the injection and production pipe suspension height of the salt cavern energy storage system.

[0125] To meet the precise control requirements of the injection-production pipe suspension height in salt cavern energy storage systems, this specification provides a method for adjusting the injection-production pipe suspension height. Figure 8 This is a flowchart illustrating a method for adjusting the suspension height of an injection / production pipe, as provided in an embodiment of this specification. In practice, it may include the following steps:

[0126] S10: Obtain the first displacement data of the telescopic tube and the torque data of the screw mechanism at the current moment.

[0127] In some embodiments, step S10 may specifically include: acquiring first displacement data of the telescopic tube from the previous historical moment to the current moment, measured by a plurality of displacement sensors 10; and acquiring torque data of the screw mechanism at the current moment, measured by a torque sensor 18.

[0128] The first displacement data can be a set of data directly measured by multiple displacement sensors 10, reflecting the relative position change of the telescopic tube 2 from the previous historical moment to the current moment. In specific implementation, a data acquisition command can be triggered according to a preset sampling period, and the real-time measurement values ​​of each displacement sensor 10 can be read sequentially. Each displacement sensor 10 independently measures the axial movement of the telescopic tube 2 at its installation position, forming a multi-channel displacement data stream. Through the distributed layout of multiple displacement sensors, measurement redundancy is formed. When a single sensor fails, the system can still maintain basic functions through the remaining sensors, significantly improving the fault tolerance and reliability of the control system.

[0129] Torque data can be a physical quantity that characterizes the torque required for the screw lifting mechanism to drive the load, measured in real time by torque sensor 18. In specific implementation, torque sensor 18 can detect the micro-strain generated by the drive shaft under torque, convert the strain signal into a standard electrical signal through a signal conditioning circuit, and then perform analog-to-digital conversion by a data acquisition card to finally obtain a digital torque measurement value.

[0130] At any given moment, the controller can send synchronous acquisition commands to multiple distributed displacement sensors 10. Each displacement sensor 10, based on its detection principle, converts the linear displacement of the telescopic tube 2 relative to the main tube 1 into a corresponding pulse signal or digital signal. The controller can read these signals using a high-speed counter or a dedicated interface chip and apply a scaling transformation formula to convert them into actual displacement values, forming a first displacement dataset. The first displacement measurements from multiple locations within the first displacement dataset can be fused to obtain the first displacement data. Similarly, the controller can simultaneously acquire torque information via torque sensors 18 installed in the screw mechanism drive chain.

[0131] Simultaneous acquisition of two different physical quantities, displacement and torque, can reflect both the kinematic state of the actuator and its dynamic load, providing a more comprehensive data foundation for subsequent state estimation and control decisions.

[0132] S20: Input the first displacement data and torque data into the displacement correction model to obtain the second displacement data of the telescopic tube.

[0133] In some embodiments, the above displacement correction model is coupled with the extended Kalman displacement model, the strong tracking displacement model, and the unscented Kalman displacement model to handle the linear correlation characteristics, nonlinear correlation characteristics, and abrupt correlation characteristics between the torque of the screw mechanism and the displacement of the telescopic tube, respectively.

[0134] The displacement correction model can be an intelligent estimation model based on the fusion of multiple displacement estimation models. It can perform error compensation and accuracy optimization on the directly measured first displacement data and output more accurate and reliable second displacement data, i.e., optimized displacement estimate.

[0135] The Extended Kalman Displacement Model (EDM) can leverage the approximately linear dynamic characteristics of a system operating in a stable state. It employs Taylor expansion to perform first-order local linearization of the state of the transmission system comprised of a helical mechanism and a telescopic tube. Specifically, the model calculates the Jacobian matrix of the transmission system's state and observation equations at the current moment, transforming the nonlinear problem into a linear Gaussian filtering framework for solution. Through a predictive-corrective two-step recursive algorithm, state prediction is first performed based on the transmission system's dynamic model. Then, the prediction results are weighted and corrected using sensor observation data, thereby achieving a minimum variance estimate of the system state. This model can provide high-precision displacement estimation results when the system operates smoothly and has low nonlinearity.

[0136] The strong tracking displacement model can address situations where transmission system parameters change abruptly or unmodeled dynamics exist by introducing a time-varying fading factor to adjust the filter gain in real time. This model calculates the deviation between the actual and theoretical covariance of the innovation sequence, and then constructs an optimization problem based on this deviation to address the adaptive fading factor. By adjusting the weights of the state prediction error covariance matrix online, the filter is forced to maintain its ability to track abrupt changes in state. Even when the system's dynamic characteristics change drastically, this model can maintain high estimation accuracy and effectively suppress filter divergence.

[0137] Unscented Kalman displacement models can handle strongly nonlinear system characteristics using deterministic sampling strategies. Based on the Sigma point sampling rule, this model selects a set of statistically representative sample points in the state space, ensuring that the mean and covariance of these sample points are consistent with the current state distribution. These Sigma points are then propagated through a nonlinear system function, and the propagated point set is weighted and reconstructed to obtain the mean and covariance of the state prediction. This method can accurately capture the higher-order moments of the state distribution after the nonlinear transformation, significantly reducing linearization errors and demonstrating superior estimation performance when the system exhibits strongly nonlinear characteristics.

[0138] Based on this, at each current moment, the controller can read the real-time measurement values ​​from multiple tilt sensors 19, and use a coordinate transformation algorithm to uniformly convert the local tilt angle measurement values ​​of each sensor to the global coordinate system, calculate the actual attitude angle of the telescopic pipe in three-dimensional space, and thus obtain the tilt angle change from the previous historical moment to the current moment. By fusing tilt angle data, the displacement measurement error caused by pipe bending deformation is effectively compensated, and the measurement accuracy under complex working conditions is improved.

[0139] Furthermore, the first displacement data, torque data, and tilt angle change can be simultaneously input into three parallel-running filters: the extended Kalman displacement model can predict and update using state-space equations, outputting third displacement data that considers the linear dynamics of the system; the strong-tracking displacement model can monitor the statistical characteristics of the innovation sequence in real time, adaptively adjust the gain matrix, and output fourth displacement data that has the ability to quickly track abrupt states; the unscented Kalman displacement model can process the nonlinearity of the system through unscented transformation, outputting fifth displacement data that accurately reflects the nonlinear dynamics. The parallel architecture of the three filters enables the system to simultaneously cope with multiple dynamic characteristics such as linearity, nonlinearity, and abrupt changes, significantly enhancing the robustness of state estimation.

[0140] The controller can calculate the confidence weight of each model based on the real-time performance indicators of each filter using a game theory-based multi-objective optimization algorithm. Finally, a weighted fusion algorithm is used to synthesize the three displacement estimates into the final second displacement data. This dynamic weight allocation mechanism based on real-time performance evaluation ensures that the most suitable estimation strategy is used under any operating condition, achieving optimal estimation accuracy.

[0141] S30: Based on the pressure change in the salt cavern and the target suspension height of the injection and production pipe at the current moment, determine the trajectory data of the telescopic pipe in the prediction time domain starting from the current moment.

[0142] In some embodiments, the pressure change can be time-series data that reflects the dynamic changes in air pressure inside the salt cavern, measured by an array of pressure sensors.

[0143] In some embodiments, the target suspension height can be set based on the optimal working position of the injection and production pipe, dynamically determined according to the actual operating conditions of the salt cavern energy storage system. This height is specifically determined based on a pressure-suspension height-efficiency mapping model pre-established through computational fluid dynamics simulation and field experimental data. This mapping model can be obtained through multiphysics fluid simulation analysis, tailored to the specific geometry of the salt cavern cavity and the characteristics of the injection and production pipe system. This model can systematically simulate the comprehensive efficiency characteristics of the injection and production process under different combinations of cavity pressure and injection / production pipe suspension heights. Efficiency is defined as the ratio of the effective energy of the injected gas to the total energy consumed or output by the system per unit time, comprehensively considering multiple factors such as airflow resistance loss, eddy current energy loss, and liquid carrying capacity. Based on the above analysis, an efficiency contour map or a database of high-efficiency working intervals is ultimately constructed, thereby establishing a complete mapping model. This model can directly output the theoretical suspension height value that achieves optimal efficiency under any real-time pressure condition. During actual system operation, pressure sensors continuously monitor changes in the internal pressure of the salt cavern. The controller collects the average air pressure value within a recent time period at a preset cycle and queries the air pressure-suspended height-efficiency mapping model. Based on the input average air pressure value, the model outputs the corresponding theoretical optimal suspended height, which is the target suspended height setpoint for the current moment.

[0144] In some embodiments, the prediction time domain can be the time range for controlling forward trajectory planning.

[0145] In some embodiments, trajectory data may be a set of optimal displacement setpoints arranged in a time series within the prediction time domain, containing at least displacement, velocity, and acceleration parameters.

[0146] In some embodiments, step S30 may specifically include: acquiring the expansion and contraction of the telescopic tube at the previous historical moment; determining the expansion and contraction of the telescopic tube at the current moment based on the expansion and contraction and the second displacement data; determining the suspension height of the telescopic tube at the current moment based on the expansion and contraction at the current moment; acquiring the pressure change in the salt cavern from the previous historical moment to the current moment as measured by the plurality of pressure sensors; and generating trajectory data of the telescopic tube in the prediction time domain from the current moment based on the pressure change, the suspension height of the telescopic tube at the current moment, and the target suspension height of the injection and extraction tube.

[0147] The historical expansion and contraction of the expansion pipe at a previous historical moment can be obtained. Combined with the second displacement data obtained through the displacement correction model at the current moment, the precise expansion and contraction at the current moment can be recursively determined. Based on the geometric relationship between this expansion and contraction and the fixed installation position of the main pipe, the actual suspension height of the expansion pipe opening relative to the bottom of the salt cave at the current moment can be calculated through coordinate transformation.

[0148] It can read the pressure time series collected from the previous historical moment to the current moment by a distributed pressure sensor network, remove measurement noise through digital filtering, and use differential calculation method to obtain dynamic characteristic parameters such as pressure change rate and change acceleration, thus constructing a complete description of the dynamic change of pressure.

[0149] By analyzing pressure change trends in real time, the generated trajectory can proactively adapt to the complex pressure dynamics within the salt cavern, avoiding control lag. Specifically, the controller can perform sliding window filtering and differential calculations on the pressure sequence to extract key characteristic parameters such as the rate of pressure change and acceleration. Based on these characteristics, an autoregressive integral moving average model is used to predict the pressure evolution trend over several control cycles. In the trajectory planning stage, the model predictive control algorithm can use the predicted pressure sequence as a feedforward input and generate a forward-looking displacement trajectory by solving a constrained optimization problem. This trajectory can adjust the velocity and acceleration distribution of the telescopic pipe in advance over time: when a rapid drop in pressure is predicted, the trajectory can increase the descent velocity of the telescopic pipe in advance, bringing the pipe opening closer to the high-efficiency gas extraction area earlier; when a stable pressure is predicted, the trajectory can optimize the deceleration curve to avoid over-adjustment. This forward-looking decision-making mechanism based on pressure trend prediction enables the control system to overcome the inherent response delay of traditional feedback control, achieving a leap from following changes to anticipating changes, significantly improving control quality and operational efficiency in dynamic pressure environments.

[0150] In some embodiments, the pressure change, current suspension height, and target suspension height can be input into a trajectory planner based on model predictive control. This planner generates the optimal trajectory through the following steps: establishing a predictive model that includes dynamic constraints and operational safety boundaries that couple the pressure change to the current suspension height; constructing a multi-objective optimization function with optimal energy efficiency as the primary objective, while also considering stability and safety; and continuously solving the multi-objective optimization function in the prediction time domain to generate an optimal displacement trajectory sequence that satisfies all constraints.

[0151] Specifically, a dynamic coupling equation between pressure change and suspension height can be established based on computational fluid dynamics principles, and the pressure fluctuations inside the salt cavern can be modeled in relation to the dynamic characteristics of the telescopic pipe. Simultaneously, the maximum allowable speed of the telescopic pipe, acceleration limits, and the minimum safe distance to avoid contact with the sediment at the bottom of the salt cavern are incorporated as hard constraints into the model framework.

[0152] Subsequently, a multi-objective optimization function can be constructed, with optimal energy efficiency as the primary objective while also considering stability and safety. This function can be implemented using a weighted summation: the energy efficiency objective quantifies the injection and extraction efficiency at different suspension heights, and the height-efficiency mapping relationship can be obtained by fitting experimental data; the stability objective penalizes abrupt acceleration changes and acceleration variations in the trajectory to ensure smooth operation of the mechanical system; the safety objective is constructed using a potential field function, which generates a sharply increasing penalty value when the trajectory approaches the safety boundary, forcing the trajectory away from the danger zone.

[0153] At each current moment, using the currently measured pressure data and suspension height as initial conditions, the prediction time domain is divided into multiple stages. The continuous optimization problem is transformed into a nonlinear programming problem using the direct multiple-shot method, and then the interior-point method is used for efficient solution, outputting the optimal displacement trajectory sequence for several future sampling periods. This sequence satisfies all physical constraints and safety boundaries, and achieves an optimal balance between energy efficiency, stability, and safety.

[0154] This trajectory planning based on the predictive time domain can anticipate future changes in operating conditions and plan the optimal motion path in advance, significantly improving injection and extraction efficiency. Furthermore, by simultaneously considering multiple objectives such as energy efficiency, safety, and equipment lifespan during trajectory generation, the motion trajectory is optimized, reducing unnecessary acceleration and braking processes, significantly lowering system energy consumption, and improving operational economy.

[0155] S40: Based on the second displacement data and trajectory data, use the helical transmission model to generate the control command for the helical mechanism at the current moment. The helical transmission model is used to characterize the relationship between the threaded transmission characteristics of the helical mechanism and the displacement of the telescopic tube.

[0156] In some embodiments, the screw drive model can be a mathematical model describing the dynamic characteristics of the screw lifting mechanism. This model can couple the dynamic relationship between the mechanical characteristics of the screw drive system, including parameters such as lead, backlash, friction coefficient, and transmission efficiency, and the displacement of the telescopic tube. By explicitly considering the nonlinear characteristics of the transmission system through the screw drive model, the influence of factors such as backlash and friction on control accuracy can be effectively compensated.

[0157] In some embodiments, the control command may be a specific command that drives the screw mechanism to perform an action, including parameters such as target position, speed of motion, and torque limit.

[0158] In some embodiments, step S40 may specifically include: generating a sequence of control commands for the helical mechanism at multiple moments in the control time domain starting from the current moment, based on the second displacement data and trajectory data using a helical transmission model; wherein the length of the control time domain is less than the length of the prediction time domain.

[0159] A state-space expression containing the dynamic characteristics of the screw mechanism can be constructed, where the state variables include the displacement of the telescopic tube, the speed of motion, the torsional deformation of the transmission system, etc., and the control variable is the motor drive signal.

[0160] Based on the second displacement and trajectory data at the current moment, a constrained optimization problem can be constructed within the control time domain, which is significantly shorter than the prediction time domain. The objective function of this optimization problem can comprehensively balance three performance indicators: the trajectory tracking accuracy term measures the transmission system's ability to follow the desired trajectory using a quadratic form of the displacement deviation; the control energy consumption term regularizes the motor output torque to improve energy efficiency; and the planner smoothness term ensures the continuity of mechanical motion by limiting the rate of change of the control variable. Simultaneously, the optimization process strictly adheres to multiple physical constraints: including the maximum speed limit of the servo motor, the peak torque capacity of the drive system, and the mechanical limitations of the helical transmission mechanism in terms of acceleration and transmission backlash. The shorter control time domain compared to the prediction time domain retains the advantage of using long-term predictive information for forward-looking decision-making while reducing computational complexity by limiting the number of optimization variables, thus meeting the requirements of real-time control.

[0161] Numerical optimization algorithms, such as quadratic programming or interior-point methods, can be used to solve this optimization problem online, obtaining a series of optimal control commands in the control time domain. Only the first control command in this sequence is used as the actual control output at the current moment, and the optimization calculation is re-performed in the next sampling period, forming a rolling optimization mechanism. This model-predictive-based rolling optimization mechanism can respond to system disturbances in a timely manner and maintain excellent tracking performance.

[0162] In some embodiments, at each current moment, only the control instruction corresponding to the current moment in the control instruction sequence is retained.

[0163] A control command sequence can be a set of future control commands arranged in chronological order in the control time domain, obtained by solving a model predictive control algorithm.

[0164] The control instruction corresponding to the current moment can specifically refer to the first one in the sequence, that is, the control instruction that needs to be executed immediately.

[0165] Specifically, at each current moment, the controller can acquire the current state of the injection and sampling system, including second displacement data, trajectory data, etc., and then solve the finite-time optimization problem based on the helical transmission model to generate a control sequence [u(k), u(k+1), ..., u(k+N-1)] containing N control commands. The first control command u(k) can be accurately extracted from this control sequence, converted into a specific servo motor drive signal, including target speed, direction, and torque limit, and sent to the servo motor of the helical mechanism. After the current control command is output, the controller can completely discard the remaining future control commands [u(k+1), ..., u(k+N-1)] and wait for the next sampling cycle to re-measure and optimize the state to generate a completely new control command sequence.

[0166] By re-optimizing in each cycle, the system can respond promptly to unforeseen disturbances such as sudden changes in salt cavern pressure and mechanical resistance, maintaining control accuracy. Feeding the latest actual displacement measurements back to the controller effectively corrects model prediction errors, forming a closed-loop compensation mechanism. This avoids reliance on pre-calculated future control commands, allowing the control strategy to dynamically adjust based on real-time operating conditions. Frequent optimization updates prevent error accumulation and system divergence, ensuring long-term operational reliability. Furthermore, retaining only currently necessary control commands reduces data storage and processing resource requirements, and calculation anomalies in a single cycle do not affect subsequent control quality; the system possesses rapid self-recovery capabilities.

[0167] S50: Based on the control command, control the spiral mechanism to adjust the suspension height of the telescopic tube.

[0168] In some embodiments, the control command may be a control signal obtained through rolling optimization calculation, which includes parameters such as target speed, rotation direction, acceleration curve and torque limit.

[0169] In some embodiments, step S50 may specifically include: controlling the rotation speed and rotation direction of the rotary lift based on the control command, so as to drive the threaded lifting rod to rotate accordingly, and then causing the telescopic tube to move axially under the push of the threaded lifting rod.

[0170] The suspension height adjustment can be achieved by changing the axial position of the telescopic pipe in the salt cavern, so that its pipe opening reaches the optimal injection and production efficiency position under the current working conditions.

[0171] The servo motor of the helical mechanism can receive digital control commands from the controller and decompose them into specific motor control parameters through a built-in analytical algorithm. These parameters include the target speed, rotation direction indicator (forward / reverse), acceleration curve, and torque limit threshold. Based on the analyzed control parameters, the output characteristics of the servo motor are precisely adjusted through a three-loop control strategy consisting of a current loop, a speed loop, and a position loop. Specifically, this includes: precisely controlling the motor torque and speed using vector control technology; employing an S-curve acceleration / deceleration algorithm to ensure smooth motion; and monitoring the actual motor output in real time and comparing and correcting it with the target value.

[0172] The precise rotational motion of the servo motor is transmitted to the worm gear reducer via a coupling. After being reduced in speed and increased in torque by a preset reduction ratio, it drives the threaded lifting rod to perform the corresponding rotational motion. The rotation of the threaded lifting rod is converted into linear displacement of the telescopic tube through its precision trapezoidal thread.

[0173] Driven by the threaded lifting rod, the telescopic tube moves precisely axially along the preset sliding guide mechanism, namely the slide rail and pulley block.

[0174] This precise motion control avoids unnecessary acceleration and braking processes, and combined with a highly efficient transmission system, it significantly reduces energy consumption and improves the accuracy of suspension height adjustment.

[0175] In some embodiments, it can be determined whether the current suspension height of the telescopic tube is equal to the target suspension height; if it is equal, the control ends; if it is not equal, the steps of obtaining the control command at the current moment and adjusting the suspension height of the telescopic tube based on the control command at the current moment are executed iteratively until the current suspension height of the telescopic tube is equal to the target suspension height.

[0176] At each current moment, the absolute difference between the current suspension height and the target suspension height can be calculated. A preset accuracy threshold ε (e.g., ±2mm) can be set. When the absolute difference is less than or equal to ε, the heights are considered equal. When the height equality condition is met, the current operating parameters are saved to the historical database, and a position holding command is sent to the screw mechanism to maintain the current suspension height unchanged.

[0177] When the heights are not equal, the sensor data is reacquired, the complete S10-S50 control steps are executed, new control commands are generated, and the screw mechanism is driven to adjust the suspension height. This process is repeated until the target suspension height is reached.

[0178] This continuous closed-loop feedback and iterative correction ensures a gradual approach to and stabilization at the target altitude, eliminating steady-state errors. Control commands are recalculated based on the latest system state at each current moment, effectively addressing dynamic disturbances such as pressure fluctuations and changes in mechanical characteristics within the salt cavern. Furthermore, the iterative mechanism provides fault tolerance; random errors at a single moment do not affect the final control performance, demonstrating self-correction capabilities.

[0179] In some embodiments, the fusion of the third, fourth, and fifth displacement data in step S20 to obtain the second displacement data may specifically include: determining the confidence levels of the extended Kalman displacement model, the strong tracking displacement model, and the unscented Kalman displacement model based on the third, fourth, and fifth displacement data, respectively; wherein the confidence level of each model is determined based on at least the following indicators: the whitening degree of the model's innovation sequence, the consistency between the model's displacement data and other model displacement data, the convergence speed of each model, and the computational complexity of each model; coupling the confidence levels of the extended Kalman displacement model, the strong tracking displacement model, and the unscented Kalman displacement model to obtain the confidence level of the displacement correction model; maximizing the confidence level of the displacement correction model to obtain the optimal confidence levels of the extended Kalman displacement model, the strong tracking displacement model, and the unscented Kalman displacement model at the current time; and fusing the third, fourth, and fifth displacement data based on the optimal confidence levels of the extended Kalman displacement model, the strong tracking displacement model, and the unscented Kalman displacement model at the current time to obtain the second displacement data.

[0180] The autocorrelation function and Ljung-Box test can be performed on the innovation sequences of each model to determine the degree of matching between the model and the actual dynamics of the injection and collection pipe by evaluating whether they conform to the statistical characteristics of zero-mean white noise. The smaller the whitening test statistic, the higher the confidence score. The innovation sequence of each model can be the time series data of the difference between its output displacement data and the second displacement data.

[0181] The Kullback-Leibler divergence can be used to calculate the differences in the distributions of the three sets of displacement data, and hypothesis testing can be used to assess the level of consistency between the model outputs. The smaller the distribution difference, the higher the confidence level of consistency.

[0182] It can monitor the decay rate of the trace of the state estimation error covariance matrix of each model over time; the larger the convergence exponent, the better the dynamic response performance.

[0183] Computational efficiency can be evaluated by monitoring the CPU time utilization and memory consumption of each model in real time. The lower the resource consumption, the higher the complexity score.

[0184] A multi-objective decision-making model based on the analytic hierarchy process (AHP) can be constructed. A judgment matrix is ​​established for the four evaluation indicators, and the weights of each indicator are calculated using the eigenvector method. A weighted mean method is then used to couple the multi-dimensional confidence indices of each model to obtain a comprehensive confidence score.

[0185] An optimization problem based on Nash negotiation game can be established, with the objective function being to maximize the confidence of the overall displacement correction model while satisfying the fairness constraint of confidence allocation for each model. The Pareto optimal confidence allocation scheme can be obtained by solving the problem using the Lagrange dual method. The obtained optimal confidence can be normalized into fusion weight coefficients, and the second displacement data can be calculated using an uncertainty-based optimal weighted fusion algorithm.

[0186] By conducting multi-dimensional real-time performance evaluation, dynamic weight adjustments are made based on the actual performance of each model, significantly improving the situational adaptability of the fusion system. Simultaneously, estimation accuracy, algorithm consistency, convergence characteristics, and computational efficiency are optimized to ensure the fusion system achieves an optimal balance across multiple performance dimensions. When the performance of a model degrades due to sudden changes in operating conditions, its fusion weight can be automatically reduced, effectively isolating the negative impact of poor estimation on overall performance. This adaptive fusion method, based on multi-index confidence assessment and game theory optimization, fully leverages the technical advantages of each filtering algorithm, providing the injection-progression pipeline control system with high-precision and high-reliability displacement state sensing capabilities.

[0187] In some embodiments, the step S20 above, which generates trajectory data of the telescopic pipe within the prediction time domain from the current moment based on the pressure change, the current suspension height of the telescopic pipe, and the target suspension height of the injection-production pipe, may include: inferring the latent state probability distribution of the fluid flow field inside the salt cavern based on the pressure change and the current suspension height of the telescopic pipe using a preset salt cavern flow field model; wherein the latent state probability distribution includes at least the probability distribution of the intensity and position of the flow field eddy current, the probability distribution of the sediment suspension state, the probability distribution of the flow stability, and the probability distribution of the fluid velocity and direction at the inlet of the telescopic pipe; constructing a Markov decision process spanning multiple moments within the prediction time domain using the latent state probability distribution, the current suspension height of the telescopic pipe, and the target suspension height as constraints; and solving the Markov decision process to generate trajectory data for multiple moments within the prediction time domain.

[0188] A digital twin model of the salt cave flow field based on computational fluid dynamics can be established. This model can couple the multiphase flow dynamics, thermodynamics, and kinematic equations of solid particles in the salt cave. The pressure change sequence and the current suspension height can be used as observation inputs, and the model state can be dynamically assimilated with real-time measurement data using the Ensemble Kalman Filter algorithm. Furthermore, a particle set representing the flow field state can be generated using the Monte Carlo method, and the posterior probability distribution of each hidden state can be calculated through Bayesian updates.

[0189] The posterior probability distribution of each latent state can characterize the probabilistic description of the flow field state inside the salt cavern that cannot be directly observed, including: the probability density function of eddy intensity and location; the spatial distribution probability of sediment suspension concentration; the Markov transition probability of flow stability; and the joint probability distribution of the nozzle velocity vector. This method of predicting potential flow field risks through latent state probability distributions can proactively avoid eddy zones and sediment accumulation zones during the trajectory planning stage, significantly improving operational safety. Furthermore, this probabilistic framework-based decision-making method can effectively handle the uncertainty of flow field dynamics, ensuring reliability under complex operating conditions.

[0190] A Markov decision process includes a state space, an action space, a reward function, and a state transition model. Specifically, the state vector in the state space can include multi-dimensional variables such as suspension height, hidden state probability distribution, and pressure gradient. In the action space, controllable parameters such as the velocity and acceleration of the telescopic tube can be defined as decision actions. The reward function can comprehensively consider injection and extraction efficiency, energy consumption costs, safety risks, and equipment lifespan to establish a multi-objective reward function. The state transition model allows for a probability transition matrix considering uncertainties, established based on the physical laws of the flow field. By comprehensively considering multiple objectives such as efficiency, safety, and energy consumption, the generated trajectory achieves Pareto optimality among multiple performance indicators.

[0191] An asynchronous value iterative method can be used to solve the Bellman optimal equation of the Markov decision process, obtaining the optimal value function and generating a trajectory sequence in the prediction time domain. At each time step, the solution is re-solved based on the latest state information, achieving adaptive trajectory adjustment. A real-time replanning mechanism enables the generated trajectory to quickly respond to sudden changes in the flow field, and global optimization in the prediction time domain avoids short-sighted decisions, achieving optimal long-term performance.

[0192] In some embodiments, the current time can be the time of each preset time interval. The preset time interval can be 1 second or tens of seconds.

[0193] The current moment can be any point in time at which the controller performs periodic sampling at preset time intervals. The preset time interval can be flexibly configured according to the specific operating conditions and control accuracy requirements of the salt cavern compressed air energy storage system.

[0194] Specifically, the preset time interval can be set based on a trade-off of several factors: when the gas pressure inside the salt cavern changes drastically or during the injection-production switching phase, a shorter time interval (e.g., 1 second) can be used to ensure that the control system can respond quickly to changes in operating conditions; considering the mechanical inertia of the spiral mechanism and telescopic tube, the time interval must be greater than the minimum response time of the system to avoid equipment oscillation due to excessively high command frequency; a longer time interval (e.g., 30 seconds) can reduce the computational load on the controller and is suitable for operating conditions where the system is in steady-state operation or slowly changing conditions; frequent control adjustments will increase system energy consumption, and appropriately extending the interval under non-emergency conditions is beneficial to improving operational economy. This discrete control mechanism based on configurable time intervals provides a flexible and efficient control strategy for the salt cavern energy storage injection-production pipe system, enabling it to adapt to diverse operating scenarios and performance requirements.

[0195] In practical implementation, a high-precision timer can be maintained. When the clock count reaches a preset interval, a new cycle of data acquisition, state estimation, trajectory planning, and control command generation is triggered. This discretized time scheduling mechanism ensures both real-time control and efficient utilization of system resources.

[0196] As can be seen from the above embodiments of the method for adjusting the suspension height of the injection-production pipe provided in this specification, this embodiment can obtain the first displacement data of the telescopic pipe and the torque data of the screw mechanism at the current moment; input the first displacement data and torque data into the displacement correction model to obtain the second displacement data of the telescopic pipe; based on the pressure change in the salt cavern at the current moment and the target suspension height of the injection-production pipe, determine the trajectory data of the telescopic pipe in the prediction time domain from the current moment; according to the second displacement data and trajectory data, use the screw transmission model to generate the control command of the screw mechanism at the current moment, the screw transmission model is used to characterize the correlation between the screw mechanism's thread transmission characteristics and the telescopic pipe displacement; based on the control command, control the screw mechanism to adjust the suspension height of the telescopic pipe. By establishing a displacement correction model and fusing multi-source data of displacement and torque to perform real-time correction and optimization of directly measured displacement data, the problem of single sensor being easily interfered with and having insufficient measurement accuracy under complex working conditions in the salt cavern is effectively solved, providing reliable state feedback for precise control of the suspension height. Furthermore, based on real-time pressure changes within the salt cavern and the target suspension height, displacement trajectory data can be generated in the prediction time domain, enabling proactive prediction of operating condition trends. This allows for advance planning of suspension height adjustment paths based on dynamic gas pressure characteristics, effectively overcoming control lag and improving injection-production efficiency. By constructing a helical transmission model that couples the thread transmission characteristics of the helical mechanism with the displacement of the telescopic tube, the mechanical characteristics of the transmission system are fully considered during the control command generation stage. This allows the control commands to effectively compensate for nonlinear factors such as transmission backlash and elastic deformation, significantly improving the accuracy and stability of displacement control. By sensing pressure changes in real time and generating predicted trajectories, the system can quickly adapt to dynamic fluctuations in gas pressure within the salt cavern. Combined with the real-time correction function of the displacement correction model, the adaptability and robustness under complex salt cavern conditions are comprehensively enhanced. In addition, precise displacement control avoids ineffective movement and over-adjustment of the telescopic tube, reducing mechanical wear and energy loss. Trajectory planning can also effectively reduce mechanical impact, thereby extending equipment lifespan.

[0197] Based on the above-described method for adjusting the suspension height of the injection-production pipe, this specification also proposes an embodiment of a device for adjusting the suspension height of the injection-production pipe. The injection-production pipe includes a main pipe fixed in a salt cavern injection-production well, a telescopic pipe coaxially sleeved on the outer wall of the main pipe, and a screw mechanism for driving the telescopic pipe to move axially along the main pipe. For example... Figure 9 As shown, the adjustment device 900 for the suspension height of the injection-production pipe may specifically include the following modules:

[0198] The acquisition module 901 is used to acquire the first displacement data of the telescopic tube and the torque data of the screw mechanism at the current moment.

[0199] The correction module 902 is used to input the first displacement data and torque data into the displacement correction model to obtain the second displacement data of the telescopic tube.

[0200] The determination module 903 is used to determine the trajectory data of the telescopic pipe in the prediction time domain from the current moment, based on the pressure change in the salt cave and the target suspension height of the injection and production pipe at the current moment.

[0201] The generation module 904 is used to generate the control command of the screw mechanism at the current moment using the screw transmission model based on the second displacement data and trajectory data. The screw transmission model is used to characterize the relationship between the screw mechanism's thread transmission characteristics and the displacement of the telescopic tube.

[0202] The control module 905 is used to control the spiral mechanism to adjust the suspension height of the telescopic tube based on the control command.

[0203] In some embodiments, the injection and extraction tube further includes a plurality of displacement sensors uniformly arranged axially on the outer wall of the telescopic tube and a torque sensor disposed within the spiral lifting mechanism.

[0204] Based on this, the aforementioned acquisition module 901 can be specifically used for:

[0205] Acquire the first displacement data of the telescopic tube from the previous historical moment to the current moment, measured by the plurality of displacement sensors;

[0206] Obtain the torque data of the screw mechanism at the current moment as measured by the torque sensor.

[0207] In some embodiments, the injection and sampling pipe further includes a plurality of tilt sensors uniformly arranged at the inlet of the telescopic pipe. The displacement correction model is coupled with the extended Kalman displacement model, the strong tracking displacement model, and the unscented Kalman displacement model, which are used to handle the linear correlation characteristics, nonlinear correlation characteristics, and abrupt correlation characteristics between the torque of the screw mechanism and the displacement of the telescopic pipe, respectively.

[0208] Based on this, the aforementioned correction module 902 can be specifically used for:

[0209] The change in tilt angle of the telescopic tube from the previous historical moment to the current moment is obtained, as measured by the plurality of tilt sensors.

[0210] The first displacement data, torque data, and tilt angle change are respectively input into the extended Kalman displacement model, the strong tracking displacement model, and the unscented Kalman displacement model to obtain the third displacement data, the fourth displacement data, and the fifth displacement data.

[0211] The second displacement data is obtained by fusing the third, fourth, and fifth displacement data.

[0212] In some embodiments, the correction module 902 may also be used for:

[0213] Based on the third, fourth, and fifth displacement data, the confidence levels of the extended Kalman displacement model, the strong tracking displacement model, and the unscented Kalman displacement model are determined respectively; wherein, the confidence level of each model is determined based on at least the following indicators: the whitening degree of the model's innovation sequence, the consistency between the model's displacement data and the displacement data of other models, the convergence speed of each model, and the computational complexity of each model.

[0214] The confidence scores of the extended Kalman displacement model, the strong tracking displacement model, and the unscented Kalman displacement model are coupled to obtain the confidence score of the displacement correction model.

[0215] Maximize the confidence of the displacement correction model to obtain the optimal confidence of the extended Kalman displacement model, the strong tracking displacement model, and the unscented Kalman displacement model at the current time.

[0216] Based on the optimal confidence levels of the extended Kalman displacement model, the strong tracking displacement model, and the unscented Kalman displacement model at the current moment, the third displacement data, the fourth displacement data, and the fifth displacement data are fused to obtain the second displacement data.

[0217] In some embodiments, the injection and collection tube further includes a plurality of pressure sensors evenly distributed at the inlet of the telescopic tube.

[0218] Based on this, the aforementioned determining module 903 can be specifically used for:

[0219] Obtain the expansion and contraction amount of the telescopic tube at the previous historical moment;

[0220] Based on the expansion and contraction amount and the second displacement data, determine the expansion and contraction amount of the telescopic tube at the current moment;

[0221] Determine the suspension height of the telescopic pipe at the current moment based on the amount of expansion and contraction.

[0222] Acquire the pressure change within the salt cave from the previous historical moment to the current moment, as measured by the plurality of pressure sensors;

[0223] Based on the pressure change, the current suspension height of the telescopic pipe, and the target suspension height of the injection and extraction pipe, trajectory data of the telescopic pipe within the prediction time domain starting from the current moment is generated.

[0224] In some embodiments, the determining module 903 described above can also be used for:

[0225] Based on the pressure change and the current suspension height of the telescopic pipe, the latent state probability distribution of the fluid flow field inside the salt cave is inferred through a preset salt cave flow field model; wherein, the latent state probability distribution includes at least the probability distribution of the intensity and position of the flow field eddy, the probability distribution of the sediment suspension state, the probability distribution of the flow stability, and the probability distribution of the fluid velocity and direction at the inlet of the telescopic pipe.

[0226] Using the hidden state probability distribution, the current suspension height of the telescopic tube, and the target suspension height as constraints, a Markov decision process spanning multiple moments in the prediction time domain is constructed.

[0227] Solve the Markov decision process to generate trajectory data at multiple times within the prediction time domain.

[0228] In some embodiments, the generation module 904 described above can be specifically used for:

[0229] Based on the second displacement data and trajectory data, a sequence of control commands for the helical mechanism at multiple moments within the control time domain starting from the current moment is generated using a helical transmission model; wherein the length of the control time domain is less than the length of the prediction time domain.

[0230] In some embodiments, the aforementioned helical mechanism includes a rotary lift and a threaded lifting rod; the rotary lift is used to drive the threaded lifting rod to rotate, so that the threaded lifting rod moves up and down in a helical manner; the threaded lifting rod is used to push the telescopic tube to move axially; the injection and extraction tube also includes a fixing ring welded to the outer wall of the main tube, a limiter for bolting the rotary lift to the fixing ring, and a sliding mechanism; the sliding mechanism includes multiple grooved slide rails uniformly arranged axially on the inner wall of the telescopic tube, and multiple pulleys arranged in the grooved slide rails.

[0231] Based on this, the aforementioned control module 905 can be specifically used for:

[0232] Based on the control command, the rotation speed and direction of the rotary lift are controlled to drive the threaded lifting rod to rotate accordingly, thereby causing the telescopic tube to move axially under the push of the threaded lifting rod.

[0233] As can be seen from the above embodiments of the injection-production pipe suspension height adjustment device provided in this specification, this embodiment can acquire the first displacement data of the telescopic pipe and the torque data of the screw mechanism at the current moment; input the first displacement data and torque data into the displacement correction model to obtain the second displacement data of the telescopic pipe; based on the pressure change in the salt cavern at the current moment and the target suspension height of the injection-production pipe, determine the trajectory data of the telescopic pipe in the prediction time domain from the current moment; according to the second displacement data and trajectory data, use the screw transmission model to generate the control command of the screw mechanism at the current moment, the screw transmission model is used to characterize the correlation between the screw mechanism's thread transmission characteristics and the telescopic pipe displacement; based on the control command, control the screw mechanism to adjust the suspension height of the telescopic pipe. By establishing a displacement correction model and fusing multi-source data of displacement and torque to perform real-time correction and optimization of directly measured displacement data, the problem of single sensor being easily interfered with and having insufficient measurement accuracy under complex working conditions in the salt cavern is effectively solved, providing reliable state feedback for precise control of suspension height. Furthermore, based on real-time pressure changes within the salt cavern and the target suspension height, displacement trajectory data can be generated in the prediction time domain, enabling proactive prediction of operating condition trends. This allows for advance planning of suspension height adjustment paths based on dynamic gas pressure characteristics, effectively overcoming control lag and improving injection-production efficiency. By constructing a helical transmission model that couples the thread transmission characteristics of the helical mechanism with the displacement of the telescopic tube, the mechanical characteristics of the transmission system are fully considered during the control command generation stage. This allows the control commands to effectively compensate for nonlinear factors such as transmission backlash and elastic deformation, significantly improving the accuracy and stability of displacement control. By sensing pressure changes in real time and generating predicted trajectories, the system can quickly adapt to dynamic fluctuations in gas pressure within the salt cavern. Combined with the real-time correction function of the displacement correction model, the adaptability and robustness under complex salt cavern conditions are comprehensively enhanced. In addition, precise displacement control avoids ineffective movement and over-adjustment of the telescopic tube, reducing mechanical wear and energy loss. Trajectory planning can also effectively reduce mechanical impact, thereby extending equipment lifespan.

[0234] It should be understood that in the various embodiments of this specification, the sequence number of each process does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this specification.

[0235] It should also be understood that, in the embodiments of this specification, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this specification generally indicates that the preceding and following related objects have an "or" relationship.

[0236] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0237] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0238] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0239] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational tasks to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The task is a function specified in one or more boxes.

[0240] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for adjusting the suspended height of an injection / production pipe, characterized in that, The injection and production pipe includes a main pipe fixed in the salt cavern injection and production well, a telescopic pipe coaxially sleeved on the outer wall of the main pipe, and a spiral mechanism for driving the telescopic pipe to move along the axial direction of the main pipe. The method includes: Obtain the first displacement data of the telescopic tube and the torque data of the screw mechanism at the current moment; Input the first displacement data and torque data into the displacement correction model to obtain the second displacement data of the telescopic tube; Based on the pressure change in the salt cavern and the target suspension height of the injection and production pipe at the current moment, determine the trajectory data of the telescopic pipe in the prediction time domain from the current moment. Based on the second displacement data and trajectory data, the control command of the screw mechanism at the current moment is generated using the screw transmission model. The screw transmission model is used to characterize the relationship between the screw mechanism's thread transmission characteristics and the displacement of the telescopic tube. Based on the control command, the screw mechanism is controlled to adjust the suspension height of the telescopic tube.

2. The method according to claim 1, characterized in that, The injection and extraction pipe also includes multiple displacement sensors evenly arranged along the axial direction on the outer wall of the telescopic pipe, and a torque sensor arranged in the spiral lifting mechanism; The acquisition of the first displacement data of the telescopic tube and the torque data of the screw mechanism at the current moment includes: Acquire the first displacement data of the telescopic tube from the previous historical moment to the current moment, measured by the plurality of displacement sensors; Obtain the torque data of the screw mechanism at the current moment as measured by the torque sensor.

3. The method according to claim 1, characterized in that, The injection and collection tube also includes multiple tilt sensors evenly distributed at the inlet of the telescopic tube; The displacement correction model is coupled with the extended Kalman displacement model, the strong tracking displacement model, and the unscented Kalman displacement model. The extended Kalman displacement model, the strong tracking displacement model, and the unscented Kalman displacement model are respectively used to handle the linear correlation characteristics, nonlinear correlation characteristics, and abrupt correlation characteristics between the torque of the screw mechanism and the displacement of the telescopic tube. The step of inputting the first displacement data and torque data into the displacement correction model to obtain the second displacement data of the telescopic tube includes: The change in tilt angle of the telescopic tube from the previous historical moment to the current moment is obtained, as measured by the plurality of tilt sensors. The first displacement data, torque data, and tilt angle change are respectively input into the extended Kalman displacement model, the strong tracking displacement model, and the unscented Kalman displacement model to obtain the third displacement data, the fourth displacement data, and the fifth displacement data. The second displacement data is obtained by fusing the third, fourth, and fifth displacement data.

4. The method according to claim 3, characterized in that, The process of fusing the third, fourth, and fifth displacement data to obtain the second displacement data includes: Based on the third, fourth, and fifth displacement data, the confidence levels of the extended Kalman displacement model, the strong tracking displacement model, and the unscented Kalman displacement model are determined respectively; wherein, the confidence level of each model is determined based on at least the following indicators: the whitening degree of the model's innovation sequence, the consistency between the model's displacement data and the displacement data of other models, the convergence speed of each model, and the computational complexity of each model. The confidence scores of the extended Kalman displacement model, the strong tracking displacement model, and the unscented Kalman displacement model are coupled to obtain the confidence score of the displacement correction model. Maximize the confidence of the displacement correction model to obtain the optimal confidence of the extended Kalman displacement model, the strong tracking displacement model, and the unscented Kalman displacement model at the current time. Based on the optimal confidence levels of the extended Kalman displacement model, the strong tracking displacement model, and the unscented Kalman displacement model at the current moment, the third displacement data, the fourth displacement data, and the fifth displacement data are fused to obtain the second displacement data.

5. The method according to claim 1, characterized in that, The injection and extraction tube also includes multiple pressure sensors evenly distributed at the inlet of the telescopic tube; The method for determining the trajectory data of the telescopic tube in the prediction time domain from the current moment, based on the pressure change within the salt cavern at the current moment and the target suspension height of the injection-production tube, includes: Obtain the expansion and contraction amount of the telescopic tube at the previous historical moment; Based on the expansion and contraction amount and the second displacement data, determine the expansion and contraction amount of the telescopic tube at the current moment; Determine the suspension height of the telescopic pipe at the current moment based on the amount of expansion and contraction. Acquire the pressure change within the salt cave from the previous historical moment to the current moment, as measured by the plurality of pressure sensors; Based on the pressure change, the current suspension height of the telescopic pipe, and the target suspension height of the injection and extraction pipe, trajectory data of the telescopic pipe within the prediction time domain starting from the current moment is generated.

6. The method according to claim 5, characterized in that, The step of generating trajectory data of the telescopic pipe within the prediction time domain from the current moment, based on the pressure change, the current suspension height of the telescopic pipe, and the target suspension height of the injection / production pipe, includes: Based on the pressure change and the current suspension height of the telescopic pipe, the latent state probability distribution of the fluid flow field inside the salt cave is inferred through a preset salt cave flow field model; wherein, the latent state probability distribution includes at least the probability distribution of the intensity and position of the flow field eddy, the probability distribution of the sediment suspension state, the probability distribution of the flow stability, and the probability distribution of the fluid velocity and direction at the inlet of the telescopic pipe. Using the hidden state probability distribution, the current suspension height of the telescopic tube, and the target suspension height as constraints, a Markov decision process spanning multiple moments in the prediction time domain is constructed. Solve the Markov decision process to generate trajectory data at multiple times within the prediction time domain.

7. The method according to claim 1, characterized in that, The step of generating control commands for the helical mechanism at the current moment using a helical transmission model based on the second displacement data and trajectory data includes: Based on the second displacement data and trajectory data, a sequence of control commands for the helical mechanism at multiple moments within the control time domain starting from the current moment is generated using a helical transmission model; wherein the length of the control time domain is less than the length of the prediction time domain.

8. The method according to claim 1, characterized in that, The spiral mechanism includes a rotary lift and a threaded lifting rod; the rotary lift drives the threaded lifting rod to rotate, causing the threaded lifting rod to move up and down in a spiral manner; the threaded lifting rod pushes the telescopic tube to move axially; the injection and extraction tube also includes a fixing ring welded to the outer wall of the main tube, a limiter for bolting the rotary lift to the fixing ring, and a sliding mechanism; the sliding mechanism includes multiple grooved slide rails evenly arranged axially on the inner wall of the telescopic tube, and multiple pulleys arranged in the grooved slide rails. The step of controlling the spiral mechanism to adjust the suspension height of the telescopic tube based on the control command includes: Based on the control command, the rotation speed and direction of the rotary lift are controlled to drive the threaded lifting rod to rotate accordingly, thereby causing the telescopic tube to move axially under the push of the threaded lifting rod.

9. A device for adjusting the suspension height of an injection / production pipe, characterized in that, The injection and production pipe includes a main pipe fixed in the salt cavern injection and production well, a telescopic pipe coaxially sleeved on the outer wall of the main pipe, and a spiral mechanism for driving the telescopic pipe to move along the axial direction of the main pipe. The device includes: The acquisition module is used to acquire the first displacement data of the telescopic tube and the torque data of the screw mechanism at the current moment; The correction module is used to input the first displacement data and torque data into the displacement correction model to obtain the second displacement data of the telescopic tube; The determination module is used to determine the trajectory data of the telescopic tube in the prediction time domain from the current moment, based on the pressure change in the salt cavern and the target suspension height of the injection and production tube at the current moment. The generation module is used to generate the control command of the screw mechanism at the current moment based on the second displacement data and trajectory data using the screw transmission model. The screw transmission model is used to characterize the relationship between the screw mechanism's thread transmission characteristics and the displacement of the telescopic tube. The control module is used to control the spiral mechanism to adjust the suspension height of the telescopic tube based on the control command.

10. An injection-production tube, characterized in that, The injection-production pipe includes a main pipe fixed in the salt cavern injection-production well, a telescopic pipe coaxially sleeved on the outer wall of the main pipe, a spiral mechanism for driving the telescopic pipe to move axially along the main pipe, and a controller; the controller is used to perform the method as described in any one of claims 1-8.

Images