Engineering-construction-oriented intelligent control method for production increase of coiled tubing

Abstract

The invention relates to the technical field of oil and gas field development, petroleum engineering and intelligent construction control, in particular to an engineering construction-oriented continuous oil pipe yield increase intelligent control method, which comprises the following steps of: 1, acquiring geometric parameters of a continuous oil pipe and rheological parameters of fluid in the pipe, and setting initial boundary conditions; 2, high-frequency pressure signals and flow signals in the ground pumping process are collected in real time, and well bottom apparent impedance is obtained through observation and calculation; 3, calculating the time derivative of the well bottom apparent impedance to construct a stratum softening factor representing the change trend of the stratum imbibition capacity, and generating a corresponding variable impedance flow control target; 4, inputting the variable impedance flow control target into a nonlinear model prediction controller, and driving a pump injection system to execute variable impedance servo control so as to realize self-adaptive matching of the stratum imbibition capacity; millisecond-level accurate matching of the pump injection system and the dynamic requirements of the stratum is achieved, and the yield increasing efficiency is remarkably improved.

Description

Technical Field

[0001] This invention relates to the fields of oil and gas field development, petroleum engineering and intelligent construction control technology, specifically to an intelligent control method for enhancing production in coiled tubing for engineering construction. Background Technology

[0002] During coiled tubing production enhancement operations, the complex downhole environment and variable fluid transport characteristics pose severe challenges to construction control. Existing control schemes usually rely on static surface parameters or simple pressure feedback, which makes it difficult to accurately perceive the true rheological state of the formation thousands of meters deep. Due to the significant frictional effects, momentum inertia, and large delay characteristics of fluid transport within slender tubing strings, pressure and flow signals acquired from the surface often overlap with the actual response depth at the bottom of the well, forming a signal noise wall that is difficult to penetrate. In addition, the fluid absorption capacity of the formation during fracturing or injection is highly nonlinear and dynamically variable. States such as fracture opening, sand blockage, or extension are fleeting. Traditional control architectures lack real-time tubing effect decoupling capabilities and adaptive adjustment mechanisms, resulting in pump injection rates that cannot be precisely matched with the dynamic needs of the formation. This not only limits production efficiency but may also induce mechanical damage accidents such as tubing helical buckling, making it difficult to support the requirements of high-precision and high-safety intelligent construction. Therefore, how to achieve real-time observation of the bottom hole rheological state and adaptive adjustment of the pumping impedance while filtering out interference from the tubing transmission, and improve the control accuracy and mechanical safety of the construction process, has become an urgent technical problem to be solved. Summary of the Invention

[0003] To address the aforementioned technical problems, this invention provides an intelligent control method for enhancing production in coiled tubing for engineering construction. Specifically, the technical solution of this invention includes: Step 1: Obtain the geometric parameters of the coiled tubing and the rheological parameters of the fluid inside the tubing. Based on the distributed parameter transmission line theory, discretize the coiled tubing into several impedance elements, construct the tubing-formation transmission line model, and set the initial boundary conditions. Step 2: Real-time acquisition of high-frequency pressure and flow signals during the surface pumping process, input into the tubing-formation transmission line model, use wave separation algorithm to decouple the tubing friction wave impedance and formation transient feedback, observe and calculate the bottom hole apparent impedance; Step 3: Calculate the time derivative of the apparent impedance at the bottom of the well to construct a formation softening factor that characterizes the changing trend of the formation's fluid absorption capacity, and map the formation softening factor onto a preset impedance-displacement phase plane composed of impedance and displacement to identify the current dynamic rheological response state of the formation and generate the corresponding variable impedance flow control target. Step 4: Input the variable impedance flow control target into the nonlinear model predictive controller, and combine it with the predicted output of the tubing-formation transmission line model to generate the adjustment command of the pump truck throttle, drive the pumping system to perform variable impedance servo control, so as to achieve adaptive matching of formation liquid suction capacity.

[0004] Preferably, step one includes: S11. Read the diameter, wall thickness, and length data of the coiled tubing, as well as the consistency coefficient and flow index of the working fluid; S12. Divide the coiled tubing into N discrete nodes along its length, and establish a second-order differential equation for each discrete node, which includes fluid induction, fluid capacity, and fluid resistance. S13. Concatenate the second-order differential equations of N discrete nodes to form a tubing-formation transmission line model, and define the surface pump inlet as the input port and the bottom formation as the load terminal to establish the system's transfer function matrix.

[0005] Preferably, step two includes: S21. Using high-frequency sensors deployed at the construction site, the ground pumping pressure waveform and flow waveform are synchronously collected at a preset sampling frequency. S22. Perform wavelet transform on the surface pumping pressure waveform and flow waveform to separate the downflow wave that represents the pumping input energy and the upflow wave that represents the formation reflection information. S23. Calculate the friction characteristic impedance of the tubing string using the tubing-formation transmission line model, subtract the friction characteristic impedance from the total impedance, and calculate the bottom hole apparent impedance that characterizes the transient fluid absorption capacity of the formation.

[0006] Preferably, step three includes: S31. Calculate the rate of change of the apparent impedance at the bottom of the well over time to obtain the formation softening factor S; S32. Construct an impedance-displacement phase plane with bottom hole apparent impedance as the abscissa and pumping displacement as the ordinate; S33. Compare the current formation softening factor S with the preset steady-state dead zone threshold, locate the current working point on the impedance-displacement phase plane, and generate a dynamic rheological response state based on the position of the working point.

[0007] Preferably, step three also includes: S34. Evaluate the formation softening factor S and obtain the classification results of the dynamic rheological response state, including: if the formation softening factor S is less than the negative value of the steady-state dead zone threshold, it indicates that the formation impedance is significantly reduced and the fracture is in the opening process, and the dynamic rheological response state is marked as compliant; if the formation softening factor S is greater than the positive value of the steady-state dead zone threshold, it indicates that the formation impedance is significantly increased and the formation is in the closing or blocking process, and the dynamic rheological response state is marked as rejection; if the absolute value of the formation softening factor S is less than or equal to the steady-state dead zone threshold, it indicates that the formation impedance is in a steady state, and the dynamic rheological response state is marked as equilibrium. S35. Generate variable impedance flow control target based on dynamic rheological response state: For compliant state, generate follow command to maximize displacement growth rate; for rejection state, generate oscillation command of high frequency pressure pulse; for equilibrium state, generate pressure stabilization command to maintain current displacement.

[0008] Preferably, step four includes: S41. Use the variable impedance flow control target generated in S35 as a reference trajectory and input it into the nonlinear model predictive controller. S42. Using the tubing-formation transmission line model as the prediction model, the optimal control sequence for future times is calculated by rolling optimization within the preset prediction time domain. S43. Take the first element of the optimal control sequence as the current adjustment command and send it to the pump truck's actuator to adjust the throttle opening so that the actual output impedance approaches the variable impedance flow control target.

[0009] Preferably, the method further includes the following abnormal state protection steps: S51. Perform spectral analysis on the upward wave separated in S22 and extract characteristic frequency components; S52. Determine the degree of buckling of coiled tubing downhole based on the amplitude variation of characteristic frequency components; S53. When the degree of buckling exceeds the preset mechanical safety limit, the adjustment command in step four is forcibly interrupted, and an emergency pump stop signal is generated.

[0010] Preferably, steps one through four are all run on an edge computing gateway deployed at the construction site. The edge computing gateway is connected to the pump truck control system and ground sensors via a fieldbus protocol to achieve millisecond-level data interaction and control closed loop.

[0011] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention constructs a digital twin model based on the distributed parameter transmission line theory and combines it with a wave separation algorithm, which can effectively filter out tubing friction and inertial interference from aliased ground signals and truly restore the apparent impedance at the bottom of the well. This solves the perception problem caused by signal noise walls in deep well construction, enabling the control system to penetrate thousands of meters of tubing to directly observe the transient fluid absorption capacity of the formation. 2. By introducing a formation softening factor and an impedance-displacement phase plane identification mechanism, the system can intelligently diagnose different rheological states of the formation, such as compliance, rejection, or equilibrium. Combined with nonlinear model predictive control technology, it can predict and offset the large hysteresis effect of long tubing transmission in advance, achieving millisecond-level precise matching between the pumping system and the dynamic needs of the formation, and significantly improving production efficiency. 3. This invention innovatively utilizes the upward wave spectrum analysis to invert the mechanical state of the tubing string. In the absence of downhole sensors, it can assess the buckling degree of the coiled tubing in real time by monitoring changes in characteristic frequencies. When a severe risk of helical buckling is identified, it can immediately trigger an emergency pump stop, effectively preventing tubing string lock-up or breakage accidents and ensuring the physical safety of equipment in complex operating environments. 4. The control algorithm of this invention is deployed on the edge computing gateway throughout the entire process, and data closed loop is realized through fieldbus, eliminating the dependence on cloud network. This architecture eliminates remote transmission delay and ensures that the system still has extremely high response speed and robustness in harsh field environments, meeting the stringent requirements of high-frequency production increase operations for control system stability and real-time interaction. Attached Figure Description

[0012] The present invention will be further explained below with reference to the accompanying drawings and embodiments: Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation

[0013] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0014] Example 1: Please see Figure 1 A smart control method for enhancing production in coiled tubing for engineering construction, comprising the following steps: Step 1: Obtain the geometric parameters of the coiled tubing and the rheological parameters of the fluid inside the tubing. Based on the distributed parameter transmission line theory, discretize the coiled tubing into several impedance elements, construct the tubing-formation transmission line model, and set the initial boundary conditions. Step 2: Real-time acquisition of high-frequency pressure and flow signals during the surface pumping process, input into the tubing-formation transmission line model, use wave separation algorithm to decouple the tubing friction wave impedance and formation transient feedback, observe and calculate the bottom hole apparent impedance; Step 3: Calculate the time derivative of the apparent impedance at the bottom of the well to construct a formation softening factor that characterizes the changing trend of the formation's fluid absorption capacity, and map the formation softening factor onto a preset impedance-displacement phase plane composed of impedance and displacement to identify the current dynamic rheological response state of the formation and generate the corresponding variable impedance flow control target. Step 4: Input the variable impedance flow control target into the nonlinear model predictive controller, and combine it with the prediction output of the tubing-formation transmission line model to generate the adjustment command of the pump truck throttle, drive the pumping system to perform variable impedance servo control, so as to achieve adaptive matching of formation liquid suction capacity. The core of this embodiment lies in constructing a closed-loop control system capable of decoupling the tubing effect and formation response in real time. This system establishes a digital twin to describe the dynamic transport characteristics of fluid in a slender tubing string. During this process, the system acquires the geometric parameters of the coiled tubing, including diameter, wall thickness, and length, as well as the rheological parameters of the fluid inside the tubing, such as consistency coefficient and flow index. Based on the distributed parameter transmission line theory, the coiled tubing, which is thousands of meters long, is mathematically discretized into several series impedance units, where each unit simulates a small segment of the fluid, including inertia, elasticity, and frictional characteristics. Initial boundary conditions are set, which are typically the steady-state pressure and flow distribution at the previous moment. In order to observe the true state of the formation through the noise wall of the tubing string, the system uses a high-frequency sensor installed at the pump truck outlet to collect high-frequency pressure and flow signals during the surface pumping process in real time, and inputs these signals into the aforementioned tubing-formation transmission line model. Using a wave separation algorithm, the system can distinguish between the downflow wave emitted by the pump truck and the upflow wave reflected back from the formation at the bottom of the well. By eliminating the influence of friction wave impedance along the tubing string, the system calculates the apparent impedance at the bottom of the well. This impedance refers to the transient resistance of the formation rock pores to the injected fluid after eliminating tubing friction and fluid inertia effects. Based on this, the system calculates the time derivative of the apparent impedance at the bottom of the well and constructs the formation softening factor. This factor is a dimensionless index that characterizes the changing trend of formation fluid absorption capacity. Its physical meaning lies in quantifying the rate at which formation fractures open or close. The system maps this factor to a preset impedance-displacement phase plane composed of impedance and displacement. It should be noted that in this embodiment and throughout the text, the terms flow rate and displacement refer to the volume of fluid flowing through the cross section of the tubing per unit time, i.e., volumetric flow rate. Both are the same physical quantity when describing the pumping rate. On this phase plane, different regions represent different formation dynamic rheological response states, such as fracture opening, sand blockage, and steady-state extension. The system generates a corresponding variable impedance flow control target based on the current operating point's position on the phase plane. The system executes variable impedance servo control, inputting the generated variable impedance flow control target into the nonlinear model predictive controller. Combining the prediction output of the transmission line model, the controller calculates the optimal control sequence for future moments and generates adjustment commands for the pump truck throttle, driving the pumping system to actively adapt the output impedance to the transient impedance of the formation.

[0015] Example 2: Step one includes: S11. Read the diameter, wall thickness, and length data of the coiled tubing, as well as the consistency coefficient and flow index of the working fluid; S12. Divide the coiled tubing into N discrete nodes along its length, and establish a second-order differential equation for each discrete node, which includes fluid induction, fluid capacity, and fluid resistance. S13. Concatenate the second-order differential equations of N discrete nodes to form a tubing-formation transmission line model, and define the surface pump inlet as the input port and the bottom formation as the load terminal to establish the system's transfer function matrix.

[0016] This embodiment specifies the modeling process and establishes the mathematical foundation of the system; the system reads the inner diameter, wall thickness, and length of the coiled tubing, as well as the density, bulk modulus, consistency coefficient, and flow index of the working fluid from the construction design sheet or database; The coiled tubing is divided into multiple discrete nodes with specific lengths along its length. A dynamic equation incorporating fluid induction, fluid capacity, and fluid resistance is established for the i-th node. To further improve the simulation accuracy of the distributed parameter system and accurately capture the minute fluctuations in fluid transport within a slender tubing string, this embodiment is strictly based on fluid dynamics transport line theory. The dynamic description of a single node is decoupled into momentum conservation equations and mass conservation equations. The dynamic equation set for a single node is as follows: Considering the conservation logic of pressure and flow rate in discrete space, and to ensure the consistency of physical dimensions, the above equations are discretized using a central difference scheme. Their accurate nodal dynamics iterative description is as follows: In this system of equations, the first Pressure changes at nodes are influenced by inflow rates. outflow The difference determines the spatial cascade of mass and momentum.

[0017] in: For the flow through the first The flow rate of a unit originates from the model's state variables; :No. The pressure on the nodes originates from the model's state variables; in this system of dynamic equations, the subscripts... This indicates the spatial number of the current discrete node, corresponding to the near-surface upstream end of the coiled tubing, with the subscript... This indicates that the downstream node in the wellbore direction adjacent to this node achieves continuous transmission of pressure and flow in the tubing space through this cascade relationship; The length of a single discrete node is derived from the discretization setting and is used for spatial discretization calculations. The fluid sensing value per unit length is derived from geometric parameter calculations, and its calculation formula is: in, For fluid density, The cross-sectional area of ​​the pipe, physically representing the inertial effect of the fluid; The liquid volume per unit length is derived from fluid property calculations, and its calculation formula is: in, Bulk modulus of fluid, physically meaning the compressibility elasticity of the fluid; The fluid resistance per unit length is derived from fluid dynamics calculations and its physical meaning is the frictional loss between the fluid and the pipe wall; for non-Newtonian fluids, it is calculated using the read consistency coefficient. and flow index The calculation formula is as follows: in, For pipe diameter, To linearize the reference flow rate; in this embodiment, to ensure the determinism and accuracy of the model parameters, The selection logic is defined as follows: Take the rated displacement set under the current operating conditions of the pumping system, i.e., the target flow rate; if it is in the variable displacement test phase, calculate in real time the average measured flow rate within the previous sliding time window, for example, 10 seconds, as the target flow rate. To dynamically adapt the flow index The fluid viscous resistance varies with shear rate; the system cascades the equations of all nodes to form a state-space equation, defining the surface pump inlet as the input port, the input variables as pump pressure and discharge rate (i.e., pump flow rate), and the bottom formation as the load terminal, thereby establishing the system's transfer function matrix. Specifically, this is achieved by transforming the second-order differential equations of N discrete nodes into a state-space form: The state vector is defined as follows: System matrix For a striped matrix, its non-zero elements are determined by the unit length parameter, as defined below: For flow state quantity Its derivative term corresponds to the matrix The Row, element: For pressure state quantity Its derivative term corresponds to the matrix The Row, element: Input matrix Based on the boundary conditions set at the ground port, the pump inlet pressure is... and traffic The derivative term of the first node is mapped to the mathematical closed loop from the discrete dynamic equations to the state-space description of the system, which describes the frequency domain mapping relationship between the surface input and the bottom-hole output.

[0018] Example 3: Step two includes: S21. Using high-frequency sensors deployed at the construction site, the ground pumping pressure waveform and flow waveform are synchronously collected at a preset sampling frequency. S22. Perform wavelet transform on the surface pumping pressure waveform and flow waveform to separate the downflow wave that represents the pumping input energy and the upflow wave that represents the formation reflection information; S23. Calculate the friction characteristic impedance of the tubing string using the tubing-formation transmission line model, subtract the friction characteristic impedance from the total impedance, and calculate the bottom hole apparent impedance that represents the formation's transient fluid absorption capacity.

[0019] This embodiment details the specific methods for extracting effective information from aliased signals. It utilizes a high-frequency pressure sensor and flow meter deployed at the pump truck outlet to simultaneously acquire ground pumping pressure and flow waveforms at a sampling frequency greater than or equal to 1000Hz. Wavelet transform is applied to the ground pumping pressure and flow waveforms to separate the downlink wave representing the pumping input energy and the uplink wave representing the formation reflection information. Specifically, the system employs discrete wavelet transform to perform multi-scale decomposition and reconstruction of the original signal to extract effective frequency band signals and filter out noise. The wavelet transform formula is: in, The scaling parameter controls the scaling of the wavelet function, corresponding to different frequency components; The displacement parameter controls the translation of the wavelet function on the time axis, corresponding to time positioning; For the mother wavelet function after scaling and displacement The transformed complex conjugate function; in this embodiment, to ensure the accuracy of signal decomposition and reconstruction, Daubechies4 (db4) is specifically selected as the mother wavelet function. Due to its excellent tight support and orthogonality characteristics, it is suitable for capturing transient change features in pressure wave signals; based on the clean waveform after wavelet reconstruction, the system applies traveling wave theory formulas for wave separation, the calculation formula of which is: in: It is a downward wave, derived from calculation, and its physical meaning is a waveform that represents the pump input energy; It is an upward wave, derived from calculation, and its physical meaning is a waveform that represents the information reflected by the strata. The ground-based pressure data, processed by wavelet analysis, originates from the sensor. The ground-based flow rate, after wavelet processing, originates from the sensor. The characteristic wave impedance of the tubing is derived from the model parameters, and its calculation formula is as follows: in, For fluid density, The cross-sectional area of ​​the pipe is... The fluid velocity represents the physical impedance of the tubing string itself against waves. The friction loss along the tubing string is calculated using a transmission line model. This loss is then subtracted from the total impedance to obtain the apparent impedance at the bottom of the well. Here, the total impedance physically corresponds to the overall system response observed from the surface, including the tubing string transmission effect. The step of subtracting the friction loss is mathematically implemented using a time-shift reconstruction algorithm based on traveling waves. To eliminate amplitude attenuation and phase lag caused by tubing string friction, an amplitude correction factor is introduced into the system. ,in The attenuation coefficient varies with frequency; the corrected apparent impedance at the bottom of the well. The following physical derivation process is followed: According to traveling wave theory, the bottom hole pressure With traffic The traveling wave that can be separated from the ground and compensated for time delay is expressed as: but The formula for calculating the apparent impedance at the bottom of the well at a given time is: The formula explicitly includes the characteristic wave impedance. The participation in the calculation reflects the mathematical essence of subtracting the characteristic wave impedance along the path, ensuring the physical accuracy of the impedance solution. in, Obtained by backtracking calculation The apparent impedance at the bottom of the well at any given time is derived from calculations. For the current moment The acquired and separated upflow wave data; For storage in the history buffer Downstream wave data at any given time; The one-way propagation time of the wave in the tubing is derived from geometric parameter calculations, and its calculation formula is as follows: in, For the length of the tubing, The speed of sound in a fluid; For estimation The bottom displacement at any given time is derived from the historical state of the transmission line model observer. The specific estimation method is as follows: The string-to-formation transmission line model constructed in step one is invoked. This model is essentially a system of N cascaded second-order differential equations, which is transformed into a state-space form. The calculation formula is as follows: in, Includes pressure and flow status of all discrete nodes. The ground pressure and flow data were collected in real time in step two. The Runge-Kutta numerical integration algorithm was used to solve the state equation in real time. To ensure the stability of the numerical solution of the distributed parameter system, the integration step size was set by the algorithm. Must meet In this embodiment, the step size Unified settings Thus, the bottom node, i.e., the Nth node, is obtained at time [time]. Traffic estimates; This embodiment utilizes wavelet transform and traveling wave separation algorithms to completely separate formation feedback signals and pumping interference signals in the time domain. In deep well high-pressure construction scenarios, this method allows the control system to directly observe the reflection characteristics at the bottom of the well from the surface, effectively solving the problem of misjudgment of formation conditions caused by severe signal aliasing and low signal-to-noise ratio. Furthermore, time alignment ensures the executability of the algorithm in the real-time system. During real-time calculation, specific measures are taken for the initial stage of system startup. Time window, historical downwave buffer The initial boundary conditions set in step one are used uniformly for filling to eliminate singular value jumps at the moment the algorithm starts.

[0020] Example 4: In this embodiment, step three includes: S31. Calculate the rate of change of the apparent impedance at the bottom of the well over time to obtain the formation softening factor S; S32. Construct an impedance-displacement phase plane with bottom hole apparent impedance as the abscissa and pumping displacement as the ordinate; S33. Compare the current formation softening factor S with the preset steady-state dead zone threshold, locate the current working point on the impedance-displacement phase plane, and generate a dynamic rheological response state based on the position of the working point. Step 3 also includes: S34, evaluating the formation softening factor S to obtain the classification results of the dynamic rheological response state, including: a negative value of the formation softening factor S less than the steady-state dead zone threshold indicates a significant decrease in formation impedance and that the fracture is in the opening process, and the dynamic rheological response state is marked as compliant; a positive value of the formation softening factor S greater than the steady-state dead zone threshold indicates a significant increase in formation impedance and that the formation is in the closing or blocking process, and the dynamic rheological response state is marked as rejecting; a value of the absolute value of the formation softening factor S less than or equal to the steady-state dead zone threshold indicates that the formation impedance is in a steady state, and the dynamic rheological response state is marked as equilibrium. S35. Generate variable impedance flow control target based on dynamic rheological response state: For compliant state, generate follow command to maximize displacement growth rate; for rejection state, generate oscillation command of high frequency pressure pulse; for equilibrium state, generate pressure stabilization command to maintain current displacement.

[0021] This embodiment details how to utilize impedance information for intelligent diagnosis and decision-making; the system calculates the rate of change of apparent impedance at the bottom of the well over time to obtain the formation softening factor, which is derived from real-time differential calculation; the formula is as follows: in: The formation softening factor, derived from calculation, physically represents the rate of change in the formation's fluid absorption capacity. This optimization problem is constrained by the following physical constraints within each control cycle to ensure the smoothness of throttle regulation and equipment safety. Its calculation formula is: in, and This represents the physical opening limit of the throttle valve. This represents the maximum permissible rate of change per step; simultaneously, to prevent drastic, unsteady deceleration in the pumping system, the maximum permissible rate of decrease per step is defined. ; By embedding the aforementioned constraint set into the optimization solver, the generated control instructions are ensured to remain within physically executable limits, thus preventing algorithm runaway caused by computational divergence. The apparent impedance at the bottom of the well is derived from the calculation in the previous stage. The system constructs an impedance-displacement phase plane with the apparent impedance at the bottom of the well as the pump displacement (i.e., pump flow rate) as the ordinate, and sets a steady-state dead zone threshold. This threshold is not an arbitrary constant, but is calculated based on the statistical characteristics of the high-frequency noise signal separated in step S22. The specific formula is as follows: in, The arithmetic mean of the background noise sequence. The softening factor background noise sequence measured during non-pumping periods is used to establish the steady-state determination boundary with a 99.7% confidence interval; the system compares the softening factor value at the current moment with this threshold. Compare the results; classify the current state based on the comparison results: if the softening factor is less than the negative value of the steady-state dead zone threshold, it is determined that the apparent impedance at the bottom of the well is significantly reduced, indicating that the formation microfractures are opening or extending, the formation's ability to accept fluids is enhanced, and the system marks the dynamic rheological response state as the compliant state. When the softening factor is positive and exceeds the steady-state dead zone threshold, the system determines that the apparent impedance at the bottom of the well has increased significantly, indicating sand blockage, fracture closure, or filtration channel blockage in the formation, and the formation rejects fluid injection. The system marks the dynamic rheological response state as a rejection state. When the absolute value of the softening factor is less than or equal to the steady-state dead zone threshold, the system determines that the apparent impedance at the bottom of the well is basically stable, indicating that the fracture is in a steady-state extension or filtration equilibrium stage. The system marks the dynamic rheological response state as an equilibrium state. Based on the classification results, variable impedance flow control targets are generated, and these qualitative flow control strategies are transformed into numerical impedance reference trajectories required by the nonlinear model predictive controller. ,in, To predict the number of steps and address the issue of controller input type mismatch: For compliant states, a follower instruction that maximizes displacement growth is generated, and its mathematical mapping and calculation formula are as follows: in, This is the equipment pressure limit. The displacement growth coefficient is used to ensure a stable displacement growth rate without damaging the equipment. In this embodiment, The specific value is set as follows The aim is to guide the impedance to decrease along a hyperbolic trajectory; for the rejection state, an oscillating command for generating a high-frequency pressure pulse is generated, the mathematical mapping of which is: in, Pulse intensity, The oscillation frequency, in this embodiment, is the pulse intensity. Set to 0.15, which is a 15% fluctuation range, oscillation frequency. Set as To avoid the first-order resonant frequency of the tubular column, the aim is to utilize the water hammer effect; It should be noted that the above parameters , and The specific values ​​were all determined based on simulation optimization using historical construction data; among them, The value balances the response speed of the production increase operation with the pressure fluctuation stability of the pumping system; and The selection of [the appropriate frequency] aims to generate a transient pulse sufficient to overcome the crack initiation pressure, while ensuring that the oscillation frequency avoids the natural frequency of the coiled tubing to prevent fatigue damage; for the equilibrium state, a pressure stabilization command is generated to maintain the current displacement, and its mathematical mapping is [formula missing]. To maintain a constant target impedance and ensure smooth operation.

[0022] Example 5: In this embodiment, step four includes: S41. Use the variable impedance flow control target generated in S35 as a reference trajectory and input it into the nonlinear model predictive controller. S42. Using the tubing-formation transmission line model as the prediction model, the optimal control sequence for future times is calculated by rolling optimization within the preset prediction time domain. S43. Take the first element of the optimal control sequence as the current adjustment command and send it to the pump truck's actuator to adjust the throttle opening so that the actual output impedance approaches the variable impedance flow control target.

[0023] This embodiment illustrates the specific implementation process of the controller; the generated variable impedance flow control target is used as a reference trajectory and input into the nonlinear model predictive controller; using the tubing-formation transmission line model as the prediction model, a cost function is defined to minimize the deviation between the predicted impedance and the target impedance, as well as the variation amplitude of the control quantity; the cost function formula is as follows: in: The future bottom-hole impedance predicted by the model is derived from the prediction model; The target value for the reference trajectory at a future time is derived from the flow control target generation module; The amount of change in the input, i.e., the throttle opening, is derived from the optimization variables; The weight coefficient matrix is ​​derived from preset parameters. In this embodiment, to ensure the uniformity of the dimensions of each term in the cost function and the convergence of the optimization, and to solve the algorithm failure problem caused by undefined parameters, Bryson's rule is used for parameter definition. To ensure the numerical robustness of the optimization objective, the cost function... Each item in the matrix needs to be dimensionless; weight matrix Set as a diagonal matrix, its diagonal elements ,in Take the system's maximum design impedance, for example Weight matrix Set as a diagonal matrix, its diagonal elements ,in Take the maximum permissible rate of change of throttle valve in a single step, for example... ; The number of time-domain steps for prediction is derived from the controller settings and represents the number of steps the model takes to predict forward. The number of time-domain steps, derived from the controller settings, represents the number of future control increment steps involved in the optimization calculation. Within each control cycle, the above optimization problem is solved to obtain the optimal control sequence, and the first element of the sequence is used as the current adjustment command. The unit is defined as the percentage of throttle opening, and its mapping relationship with the final control quantity of the actuator is as follows: ,in, The actual opening degree of the previous control cycle is sent to the pump truck actuator to adjust the throttle opening; This embodiment employs nonlinear model predictive control technology to explicitly address the nonlinearity and physical constraints of the system. In scenarios where the pumping system response has a large lag, the controller can predict the pressure wave transmission process in the tubing through rolling optimization, thereby acting in advance to offset the control lag caused by the long tubing and achieve high-precision impedance tracking.

[0024] Example 6: In this embodiment, the method further includes the following abnormal state protection steps: S51. Perform spectral analysis on the upward wave separated in S22 and extract characteristic frequency components; S52. Determine the degree of buckling of coiled tubing downhole based on the amplitude variation of characteristic frequency components; S53. In response to the buckling degree exceeding the preset mechanical safety limit, the adjustment command in step four is forcibly interrupted, and an emergency pump stop signal is generated. This embodiment adds an abnormal state protection function based on waveform analysis; it performs a fast Fourier transform on the separated upward wave to extract its power spectral density; and it monitors the amplitude changes of characteristic frequency components, which refer to specific resonant frequencies related to the coiled tubing helical buckling mode. In this embodiment, to guide those skilled in the art to accurately locate the monitoring frequency and ensure that the calculation results have a correct physical and mechanical basis, the specific frequency is disclosed. The specific calculation formula is as follows: in, The Young's modulus of the tubing material. Let the moment of inertia of the tubing cross section be... The effective mass per unit length, including the mass of the fluid inside the pipe, is calculated using the following formula: in, The structural mass per unit length of the coiled tubing. The density of the fluid inside the pipe, The cross-sectional area of ​​the fluid inside the pipe; The helical buckling pitch of the coiled tubing downhole is physically defined as the characteristic length of the buckling waveform. This parameter is related to the axial load on the tubing string. and radial clearance Correlation can usually be approximated as: The instruction manual does not clearly disclose how to obtain the axial load. lead to The inability to calculate is clarified in this embodiment. The acquisition path is as follows: using the tubing-formation transmission line model established in step two, the pressure value at the bottom hole node is calculated in real time. And calculate the axial load based on the hydraulic piston principle. Here, the secondary effects of gravity and friction are ignored to ensure the convergence of real-time calculations, thus achieving parameter closed-loop in the absence of downhole mechanical sensors. This formula characterizes the fundamental frequency characteristics of a tubing experiencing helical buckling within a confined space. When the tubing undergoes helical buckling under downhole pressure, the vibrational energy distribution of the system changes, resulting in a so-called resonant frequency drift phenomenon, manifested as energy shifting from the fundamental frequency to the buckling characteristic frequency. Transfer; therefore, by monitoring this specific frequency The amplitude change at a given point can reflect the degree of frequency drift; based on this, the system calculates the buckling index, using the following formula: in: The buckling index is derived from calculation; The amplitude of the current characteristic frequency is derived from spectrum analysis; This is the reference amplitude at this frequency under straight tube conditions, derived from calibration data; Considering the frequency drift that occurs during the buckling process of coiled tubing, in this embodiment... The logical definition of its value is: based on the calculated fundamental frequency. Centered on, within the frequency range A peak search is performed within this interval, and the maximum power spectral density within that interval is taken as the current characteristic amplitude; where, Set as the current base frequency ; Ensures the buckling index under string instability conditions It can accurately reflect the degree of energy transfer to high-frequency modes; in response to the buckling index exceeding the preset mechanical safety limit, the system determines that the tubing string has a serious risk of lock-up, immediately forcibly interrupts the adjustment command, generates an emergency pump stop signal, and prevents the tubing from breaking due to excessive buckling; in this embodiment, the value of the mechanical safety limit is set to 2.0, and this threshold is determined based on the material fatigue limit and geometric nonlinearity critical value of the continuous tubing. This indicates that the tubular string has entered the irreversible plastic deformation stage; In this embodiment, the numerical value of the mechanical safety limit is set as follows: This threshold is determined based on the material fatigue limit and geometric nonlinearity critical value of coiled tubing. This indicates that the tubing has entered an irreversible plastic deformation stage; in this protection logic, if a frequency drift is detected, i.e., the energy peak is within the search range... If a significant displacement occurs within the digital twin, the system will synchronously record the offset and feed it back as a correction factor to the geometric parameter model in step one, thereby achieving online self-correction of the digital twin. This embodiment innovatively utilizes fluid wave signals to invert the mechanical state of the tubing string, achieving fluid-structure interaction monitoring. In operation scenarios lacking downhole sensors, this solution can effectively prevent the most common mechanical buckling and locking accidents in coiled tubing construction without increasing hardware costs, ensuring the physical safety of the operating equipment.

[0025] Example 7: In this embodiment, steps one through four are all run on the edge computing gateway deployed at the construction site. The edge computing gateway is connected to the pump truck control system and ground sensors through the fieldbus protocol to achieve millisecond-level data interaction and control closed loop.

[0026] This embodiment defines the physical deployment architecture of the system; the above algorithm steps are encapsulated and run in an edge computing gateway deployed at the construction site; the gateway is connected to the pump truck control system via CAN bus, i.e. J1939 protocol, and connected to the ground high-frequency sensor via analog input module; it is equipped with a high-performance embedded processor, supports millisecond-level control closed-loop cycle, and completes all data acquisition, model calculation and instruction generation locally, without relying on cloud server; This embodiment adopts an edge computing architecture, which eliminates network latency and uncertainty in remote data transmission. In industrial field scenarios with no or weak network, this solution ensures the real-time performance and robustness of the control system, and meets the stringent requirements of high-frequency impedance matching control for millisecond-level response.

[0027] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A smart control method for enhancing production in coiled tubing for engineering construction, characterized in that, The specific steps include: Step 1: Obtain the geometric parameters of the coiled tubing and the rheological parameters of the fluid inside the tubing. Based on the distributed parameter transmission line theory, discretize the coiled tubing into several impedance elements, construct the tubing-formation transmission line model, and set the initial boundary conditions. Step 2: Real-time acquisition of high-frequency pressure and flow signals during the surface pumping process, input into the tubing-formation transmission line model, use wave separation algorithm to decouple the tubing friction wave impedance and formation transient feedback, observe and calculate the bottom hole apparent impedance; Step 3: Calculate the time derivative of the apparent impedance at the bottom of the well to construct a formation softening factor that characterizes the changing trend of the formation's fluid absorption capacity, and map the formation softening factor onto a preset impedance-displacement phase plane composed of impedance and displacement to identify the current dynamic rheological response state of the formation and generate the corresponding variable impedance flow control target. Step 4: Input the variable impedance flow control target into the nonlinear model predictive controller, and combine it with the predicted output of the tubing-formation transmission line model to generate the adjustment command of the pump truck throttle, drive the pumping system to perform variable impedance servo control, so as to achieve adaptive matching of formation liquid suction capacity.

2. The intelligent control method for enhancing production in coiled tubing for engineering construction according to claim 1, characterized in that, Step one includes: S11. Read the diameter, wall thickness, and length data of the coiled tubing, as well as the consistency coefficient and flow index of the working fluid; S12. Divide the coiled tubing into N discrete nodes along its length, and establish a second-order differential equation for each discrete node, which includes fluid induction, fluid capacity, and fluid resistance. S13. Concatenate the second-order differential equations of N discrete nodes to form the tubing-formation transmission line model, and define the surface pump inlet as the input port and the bottom formation as the load terminal to establish the system's transfer function matrix.

3. The intelligent control method for enhancing production in coiled tubing for engineering construction according to claim 2, characterized in that, Step two includes: S21. Using high-frequency sensors deployed at the construction site, the ground pumping pressure waveform and flow waveform are synchronously collected at a preset sampling frequency. S22. Perform wavelet transform on the surface pumping pressure waveform and flow waveform to separate the downflow wave that represents the pumping input energy and the upflow wave that represents the formation reflection information. S23. Calculate the friction characteristic impedance of the tubing string using the tubing-formation transmission line model, subtract the friction characteristic impedance from the total impedance, and calculate the bottom hole apparent impedance that characterizes the transient fluid absorption capacity of the formation.

4. The intelligent control method for enhancing production in coiled tubing for engineering construction according to claim 3, characterized in that, Step three includes: S31. Calculate the rate of change of the apparent impedance at the bottom of the well over time to obtain the formation softening factor S; S32. Construct an impedance-displacement phase plane with bottom hole apparent impedance as the abscissa and pumping displacement as the ordinate; S33. Compare the current formation softening factor S with the preset steady-state dead zone threshold, locate the current operating point on the impedance-displacement phase plane, and generate the dynamic rheological response state based on the position of the operating point.

5. The intelligent control method for enhancing production in coiled tubing for engineering construction according to claim 4, characterized in that, Step three also includes: S34. Evaluate the formation softening factor S to obtain the classification results of the dynamic rheological response state, including: if the formation softening factor S is less than a negative value of the steady-state dead zone threshold, it indicates that the formation impedance is significantly reduced and the fracture is in the opening process, and the dynamic rheological response state is marked as compliant; if the formation softening factor S is greater than a positive value of the steady-state dead zone threshold, it indicates that the formation impedance is significantly increased and the formation is in the closing or blocking process, and the dynamic rheological response state is marked as rejection; if the absolute value of the formation softening factor S is less than or equal to the steady-state dead zone threshold, it indicates that the formation impedance is in a steady state, and the dynamic rheological response state is marked as equilibrium. S35. Generate variable impedance flow control target based on dynamic rheological response state: For compliant state, generate follow command to maximize displacement growth rate; for rejection state, generate oscillation command of high frequency pressure pulse; for equilibrium state, generate pressure stabilization command to maintain current displacement.

6. The intelligent control method for enhancing production in coiled tubing for engineering construction according to claim 5, characterized in that, Step four includes: S41. Use the variable impedance flow control target generated in S35 as a reference trajectory and input it into the nonlinear model predictive controller. S42. Using the tubing-formation transmission line model as the prediction model, the optimal control sequence for future times is calculated by rolling optimization within the preset prediction time domain. S43. Take the first element of the optimal control sequence as the current adjustment command and send it to the pump truck's actuator to adjust the throttle opening so that the actual output impedance approaches the variable impedance flow control target.

7. The intelligent control method for enhancing production in coiled tubing for engineering construction according to claim 3, characterized in that, The method also includes the following abnormal state protection steps: S51. Perform spectral analysis on the upward wave separated in S22 and extract characteristic frequency components; S52. Determine the degree of buckling of coiled tubing downhole based on the amplitude variation of characteristic frequency components; S53. When the degree of buckling exceeds the preset mechanical safety limit, the adjustment command in step four is forcibly interrupted, and an emergency pump stop signal is generated.

8. The intelligent control method for enhancing production in coiled tubing for engineering construction according to claim 1, characterized in that, Steps one through four all run on an edge computing gateway deployed at the construction site. The edge computing gateway is connected to the pump truck control system and ground sensors via a fieldbus protocol to achieve millisecond-level data interaction and control closed loop.

Images