Method for controlling the leakage flow rate of phase change material at high flow rates

Abstract

This invention discloses a method for controlling the leakage velocity of phase change materials (PCMs) under high flow rates, comprising: acquiring initial configuration data of the target leakage channel and characteristic model parameters of the PCM; real-time acquisition of fluid state observation sequences of the leakage channel; inverting the real-time equivalent geometric parameters and flow field characteristic parameters of the channel based on the observation sequences; estimating the residence time window of the injected grout in the channel; estimating the phase change solidification time window by combining the characteristic model parameters and the observation sequences; determining the control criterion index based on the ratio of the residence time window to the phase change solidification time window; generating an adaptive control vector based on the control criterion index, driving the actuator to adjust the injection parameters until the flow rate control termination condition is met. This invention utilizes the solidification and residence of the PCM in the leakage channel to form a hindrance, reducing the water flow velocity in the high-velocity leakage channel to the workable range of traditional grouting methods, creating low-velocity conditions for subsequent seepage prevention treatment.

Description

Technical Field

[0001] This invention relates to the field of seepage prevention and plugging technology in water conservancy projects, and in particular to a method for controlling the leakage velocity of phase change materials under high flow rates. Background Technology

[0002] High dams, such as those with asphalt concrete core walls, are commonly used in water resource development. The sealing of their seepage prevention system is crucial to the overall structural safety of the project. Under the constraints of complex geological and construction environments, high-velocity seepage is easily induced during the impoundment period. If high-velocity seepage is not controlled in the early stages, the continuous hydraulic fracturing and scouring effects will lead to the loss of fine-particle materials inside the dam body, and may even cause engineering accidents such as dam failure. Therefore, controlling the flow velocity of seepage channels under harsh dynamic water conditions helps maintain the seepage prevention effectiveness and structural integrity of hydraulic structures.

[0003] Currently, the mainstream materials for dam seepage control mainly include concrete cutoff wall materials, grouting materials, and various composite cutoff materials. Existing seepage control processes utilize materials engineering methods such as nano-modification or organic-inorganic composites to improve the impermeability and mechanical strength of the grout itself, performing grouting under normal or pressurized pressure in still or slow-flow environments. Engineering practice shows that when the water flow velocity in the seepage channel is low, for example, below 2 m / s, cement grout and chemical grout injection methods can remain and complete solidification, achieving seepage prevention and sealing effects. For flow velocities below 2 m / s, underwater non-dispersible concrete repair methods, by increasing the viscosity of the grout and improving the setting speed, have also been successfully applied. However, when the water flow velocity in the seepage channel exceeds 2 m / s, cement grout and chemical grout will be dispersed and diluted by the water flow during injection, making it difficult to remain in the channel; underwater non-dispersible concrete sealing technology also struggles to achieve effective sealing under these flow velocity conditions. Therefore, 2 m / s constitutes the critical flow velocity threshold for the effectiveness of existing grouting methods.

[0004] Existing dynamic water plugging technologies have extremely low material retention rates when dealing with complex and corrosive flow fields exceeding critical flow velocities, leading to plugging failures or even worsening leaks. Summary of the Invention

[0005] Purpose of the invention: The present invention aims to solve the problems of the prior art and provide a method for controlling the leakage flow rate of phase change materials under high flow rate.

[0006] Technical solution: On the one hand, a method for controlling the leakage velocity of phase change materials at high flow rates is provided, including:

[0007] Obtain the initial configuration data of the target leakage channel, as well as the characteristic model parameters of the phase change material;

[0008] Real-time acquisition of fluid state observation sequences in the leakage channel;

[0009] Based on the fluid state observation sequence, the real-time equivalent geometric parameters and flow field characteristic parameters of the leakage channel are inverted;

[0010] Based on this (real-time equivalent geometric parameters and flow field characteristic parameters), the residence time window of the injected grout in the seepage channel is estimated;

[0011] By combining the characteristic model parameters of phase change materials with the fluid state observation sequence, the phase change solidification time window of the injected slurry is estimated;

[0012] The control criteria index is determined based on the ratio between the residence time window and the phase change curing time window;

[0013] An adaptive control vector is generated based on the control criterion index. The actuator is then driven to adjust the injection parameters based on the adaptive control vector until the flow rate control termination condition is met.

[0014] In conjunction with the first aspect, the fluid state observation sequence includes real-time acquisition of the volumetric flow rate of the leakage channel, the pressure difference between the two ends of the channel, and the fluid temperature;

[0015] The initial configuration data includes the channel nominal parameters and the phase transition temperature of the phase change material.

[0016] Combining the first aspect, the real-time equivalent geometric parameters and flow field characteristic parameters of the inverted leakage channels include:

[0017] The Reynolds number of the current fluid is calculated based on the fluid state observation sequence, and then compared with the flow state discrimination threshold to obtain the flow state discrimination result.

[0018] When the flow regime discrimination result indicates a laminar flow state, the equivalent diameter of the channel is obtained by inversion based on the laminar flow pressure drop model, the pressure difference at both ends of the channel and the volumetric flow rate of the leakage channel.

[0019] When the flow regime discrimination result indicates a non-laminar flow state, the equivalent diameter and equivalent friction of the channel are obtained by jointly solving the Darcy Morse pressure drop model.

[0020] The average flow velocity of the channel is calculated based on the equivalent diameter of the channel and the volumetric flow rate of the leakage channel.

[0021] The equivalent diameter of the channel is used as the real-time equivalent geometric parameter, and the average flow velocity of the channel is used as the flow field characteristic parameter.

[0022] In conjunction with the first aspect, in obtaining the equivalent channel diameter and equivalent friction based on the Darcy-Moore resistance-voltage drop model, at least one of the following strategies is employed to achieve equation closure:

[0023] The equivalent friction at the previous sampling time is used to approximate the equivalent friction at the current time.

[0024] Alternatively, the relationship between the friction coefficient and the Reynolds number can be introduced and substituted into the Darcy friction pressure drop model for simultaneous solution.

[0025] Combining the first aspect, based on real-time equivalent geometric parameters and flow field characteristic parameters, the residence time window of the injected grout in the seepage channel is estimated, including:

[0026] The equivalent length of the leakage channel is extracted from the real-time equivalent geometric parameters, and the average flow velocity of the channel is extracted from the flow field characteristic parameters.

[0027] The residence time window is determined based on the quotient of the channel's equivalent length and the channel's average flow velocity.

[0028] In conjunction with the first aspect, control criteria indicators are determined based on the ratio between the residence time window and the phase change curing time window, including:

[0029] Divide the residence time window by the phase change curing time window to calculate the capture index, which characterizes the degree of physical-time matching.

[0030] The capture index is compared with the capture index threshold to determine the result.

[0031] When the capture index is greater than or equal to the capture index threshold, it is confirmed that the current environment is feasible for flow rate control, and the capture index is used as the control criterion.

[0032] In conjunction with the first aspect, the adaptive control vector should contain at least the following control components:

[0033] The injection temperature component used to adjust the initial thermodynamic state of the slurry;

[0034] Injection rate component used to control the release rhythm of slurry space;

[0035] Nucleation triggering intensity component used for real-time intervention in the slurry phase change process.

[0036] Combining the first aspect, an adaptive control vector is generated based on the control criterion index, including a priority control strategy:

[0037] When the control criterion index does not reach the feasibility threshold for flow rate control, the adaptive control vector is adjusted first to shorten the phase change curing time window.

[0038] When the adjusted phase change curing time window reaches the physical lower limit of the material and the control criterion index still does not meet the requirements, a flow control command is generated to temporarily reduce the average flow rate of the channel, and the residence time window is extended until the control criterion index meets the preset conditions.

[0039] In conjunction with the first aspect, the actuator is driven by an adaptive control vector to adjust the injection parameters until the flow rate control termination condition is met, including:

[0040] During the process of adjusting the injection parameters of the drive actuator, the latest fluid state observation sequence is continuously collected;

[0041] The latest fluid state observation sequence is used as dynamic input to trigger the recalculation of real-time equivalent geometric parameters, residence time window and phase change solidification time window at high frequency;

[0042] The adaptive control vector is dynamically corrected based on the recalculated and updated control criterion index, forming a closed-loop control for process data reinjection until the fluid state observation sequence meets the velocity control termination condition, which represents the channel velocity decreasing below the target velocity threshold.

[0043] In conjunction with the first aspect, before obtaining the initial configuration data of the target leakage channels and the characteristic model parameters of the phase change material, pre-calibration for offline model building is also included, namely:

[0044] Acquire solidification observation records of phase change materials under different water temperature boundary conditions and fluid shear environments, and construct a cleaning calibration dataset;

[0045] Based on the cleaning calibration dataset, the parameters were identified and fitted, and an empirical model of phase transition time window was constructed to extract the characteristic model parameters.

[0046] Beneficial effects: By injecting phase change material into high-velocity seepage channels, it undergoes phase change and solidifies within the channels to form a barrier, reducing the water flow velocity within the channels to the working range of traditional grouting methods (below 2m / s). This creates low-velocity construction conditions for subsequent seepage prevention treatment using grouting materials. For smaller seepage channels or cracks, phase change grout can also directly achieve flow rate control during the flow rate control process. Attached Figure Description

[0047] Figure 1 This is a flowchart illustrating a method for controlling the leakage velocity of phase change materials at high flow rates, as provided in an embodiment of this application.

[0048] Figure 2 This is a flowchart illustrating the real-time equivalent geometric parameters and flow field characteristic parameters of the inverted leakage channel provided in the embodiments of this application.

[0049] Figure 3 This is a flowchart illustrating how to estimate the residence time window of injected slurry in a seepage channel based on real-time equivalent geometric parameters and flow field characteristic parameters, as provided in the embodiments of this application.

[0050] Figure 4 A flowchart for determining control criteria indicators based on the ratio between the residence time window and the phase change curing time window, provided for embodiments of this application.

[0051] Figure 5This is a flowchart illustrating how an adaptive control vector-driven actuator adjusts injection parameters until the flow rate control termination condition is met, as provided in an embodiment of this application. Detailed Implementation

[0052] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0053] It should be noted that the terms "first," "second," etc., in the specification and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a predetermined order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the invention described herein can be implemented in sequences other than those illustrated or described herein. Furthermore, the terms "including" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus 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 apparatus.

[0054] To solve the above problem, it was found that:

[0055] Because the dynamic changes of the internal flow field are difficult to predict, the robustness and adaptability of the fixed-parameter grouting process under harsh hydrodynamic conditions are still insufficient.

[0056] Among them, the performance of the treatment material is the decisive factor in the seepage prevention effect. Traditional grouting materials have reduced the permeability coefficient through nano-modification and organic-inorganic composite technology. However, under the flushing of high-velocity water, they are difficult to remain in the seepage channel to achieve sealing, resulting in the phenomenon of failure to seal or more leakage despite sealing.

[0057] Furthermore, phase change materials (PCMs) are increasingly being introduced into the field of leakage control due to their controllable curing, good flowability, and environmental adaptability. They exhibit advantages in temperature control, directional solidification, and adaptive sealing. Currently, most research focuses on still water environments or simplified operating conditions, lacking systematic research on the flow rate control mechanism and sealing mechanism of PCMs under high flow rate environments.

[0058] In this invention, for ease of description and understanding, some parameters are explained as follows:

[0059] Dynamic yield stress refers to the critical shear stress that a non-Newtonian fluid can withstand before rheological collapse.

[0060] max() corresponds to the function that retrieves the maximum value.

[0061] The residence time window is the mechanical boundary condition that restricts the feasibility of flow rate control in this scheme. It defines the maximum physical time span during which the phase change material released from the grouting port can remain inside the target channel before it solidifies under high-flow-rate hydrodynamic drive.

[0062] The quotient is the corresponding ratio.

[0063] The capture index is a dimensionless evaluation index used to measure the relative success or failure of the two opposing mechanisms of water erosion and material solidification.

[0064] The dynamic water shear rate characterizes the velocity gradient distribution generated by high-speed water flow on the inner wall of the channel or the surface of the retained material.

[0065] The phase change curing time window characterizes the physical time required for a phase change material to complete its liquid-solid phase transition in a dynamic water environment.

[0066] Adaptive control vector is a set of multi-dimensional commands output by the system to the on-site grouting equipment.

[0067] Empirical constitutive relations refer to the universal laws and equations obtained through calibration in fluid mechanics.

[0068] The equivalent length of the channel represents the limit of the longitudinal space in which the impermeable grout can adhere and form a sealing body in the leakage channel.

[0069] The flow regime discrimination threshold can be preset according to specific engineering geological conditions.

[0070] The consistency safety threshold can be determined based on the geometric scale of the actual leakage channel and the sensor sampling frequency, through engineering experience or on-site debugging.

[0071] Other parameters, such as the capture index threshold and the retention safety threshold, can also be determined by engineers based on factors such as safety redundancy, leakage severity, and sensor accuracy requirements in actual working conditions. Their specific values ​​can be flexibly adjusted according to the actual application scenario, and are not limited to a single value in this invention.

[0072] For example, the sampling period of the closed-loop control loop can be set to 100 milliseconds to match the time scale of flow field changes in the high-velocity channel.

[0073] To solve these problems, combined with Figures 1 to 5 The present invention will be specifically described through the following embodiments.

[0074] In some embodiments, a method for controlling the leakage velocity of a phase change material at high flow rates is provided, mainly including the following steps:

[0075] Step 101: Obtain the initial configuration data of the target leakage channel and the characteristic model parameters of the phase change material;

[0076] In this step, basic information on the engineering environment and materials is acquired. The initial configuration data includes pre-configured nominal channel parameters and the phase transition temperature of the phase change material. The nominal channel parameters can provide an estimate of the approximate aperture and length obtained during the design phase or early detection, offering an initial iterative benchmark for the inversion algorithm.

[0077] The phase transition temperature is the thermodynamic critical point at which a material changes from a liquid to a solid state.

[0078] The characteristic model parameters are a set of calibration coefficients that characterize the curing properties of phase change materials under predetermined supercooling and shear force.

[0079] Furthermore, obtaining the above parameters in advance can establish physical boundary conditions for the online time window matching algorithm.

[0080] Step 102: Real-time acquisition of fluid state observation sequences in the leakage channel;

[0081] Since the internal state of the seepage channel is difficult to observe directly, indirect measurement is required using an external sensor array. Specifically, the fluid state observation sequence includes real-time acquisition of the seepage channel volumetric flow rate, pressure difference across the channel, and fluid temperature. The volumetric flow rate can be obtained using an electromagnetic flowmeter. The pressure difference across the channel can be obtained by measuring the difference in pressure gauges placed upstream and downstream. The fluid temperature is monitored in real-time by a temperature sensor. Based on this, a dynamic time-series array reflecting the hydrodynamic state of the underground or dam body is constructed.

[0082] Furthermore, after obtaining the above observation sequence, in order to avoid random sensor noise caused by high-velocity turbulence interfering with subsequent inversion:

[0083] The real-time acquired fluid state observation sequences are subjected to time-series alignment and noise reduction. For example, a moving average filtering algorithm can be used to smooth the data.

[0084] The system identifies stable segments in the denoised fluid state observation sequence, extracts the stable intervals of the sequence fluctuations, and generates a stable condition indicator that characterizes the relatively static flow field.

[0085] The step of inverting the real-time equivalent geometric parameters and flow field characteristic parameters of the leakage channel is triggered only when the stable operating condition indicator is in an effective state.

[0086] For example, when the variance of the volumetric flow rate is less than a preset variance threshold within a consecutive preset sampling period, the operating condition stability flag is set to an active state. Subsequent state inversion steps are only triggered when the operating condition stability flag is active, thus improving the engineering's resistance to disturbances.

[0087] Step 103: Based on the fluid state observation sequence, invert the real-time equivalent geometric parameters and flow field characteristic parameters of the leakage channel;

[0088] Since actual leakage channels may be irregular holes or cracks, they need to be transformed into hydraulically computable equivalent regular pipe models.

[0089] For example, by using time-series data of flow rate and pressure difference, combined with an empirical model of pressure drop in fluid mechanics, real-time equivalent geometric parameters can be calculated, which may include the equivalent diameter and equivalent length of the channel. Simultaneously, flow field characteristic parameters, such as the average flow velocity and dynamic water shear rate of the channel, are derived by combining the volumetric flow rate.

[0090] Step 104: Based on real-time equivalent geometric parameters and flow field characteristic parameters, estimate the residence time window of the injected grout in the seepage channel;

[0091] Correspondingly, the residence time window refers to the physical time it takes for a fluid particle to go from entering the seepage channel to being completely carried out of the channel by the fluid. Furthermore, by using the equivalent channel length obtained from the inversion and the average channel velocity to calculate the quotient, the residence time window is used to characterize the limiting response time of the phase change material given by the high-velocity water flow, which can reflect the time boundary of the hydrodynamic constraint on the sealing action.

[0092] Step 105: Combine the characteristic model parameters of the phase change material with the fluid state observation sequence to estimate the phase change solidification time window of the injected slurry;

[0093] Among them, characteristic model parameters include, for example, initial configuration data, characteristic model parameters of phase change materials, fluid state observation sequences, and flow field characteristic parameters.

[0094] In this embodiment, the supercooling is calculated by extracting the fluid water temperature from the observation sequence and the phase change temperature in the initial configuration, and the shear stress state is extracted by combining the flow field characteristics. The dynamic environmental factors are input into the pre-constructed empirical solidification model, and coupled calculations are performed with the characteristic model parameters to obtain the current phase change solidification time window.

[0095] Step 106: Determine the control criterion index based on the ratio between the residence time window and the phase change curing time window;

[0096] Traditional dynamic water grouting relies on manual trial and error, which can easily lead to the grout being washed away before it solidifies. Therefore, the flow rate control problem in high-velocity leakage channels is transformed into an automatic control criterion based on time.

[0097] Alternatively, the ratio of the residence time window to the phase change curing time window is calculated to obtain a dimensionless parameter characterizing the degree of physical-time matching. This ratio is then used as a control criterion.

[0098] For example, when the residence time window is not less than the phase change curing time window, i.e. the ratio is ≥1.0, it means that the slurry has enough time to complete curing and adhering to the wall before being flushed out of the channel, forming a barrier in the leakage channel to reduce the water flow velocity in the channel.

[0099] Step 107: Generate an adaptive control vector based on the control criterion index, and drive the actuator to adjust the injection parameters based on the adaptive control vector until the preset flow rate control termination condition is met.

[0100] When the control criterion index reflects that the curing time is too long or the residence time is too short, an adaptive control vector containing a predetermined adjustment strategy is generated, such as increasing the initial heating temperature of the slurry or changing the output rate of the injection pump.

[0101] Furthermore, the actuator dynamically adjusts the grouting action based on the vector. The grouting process and the data acquisition process are synchronized and looped to form a closed loop until the average flow velocity of the channel obtained by inversion decreases below the preset target flow velocity threshold and remains below it for a preset time. At this point, the flow velocity control termination condition is triggered, and grouting stops.

[0102] In another embodiment, the inversion is further refined as follows:

[0103] Step 201: Calculate the Reynolds number of the current fluid based on the fluid state observation sequence, and compare it with the preset flow state discrimination threshold to obtain the flow state discrimination result;

[0104] Accordingly, the volumetric flow rate and the estimated equivalent diameter are extracted, and the Reynolds number of the current fluid is calculated using the existing Reynolds number formula, combined with the fluid density and dynamic viscosity. After the calculation is completed, the Reynolds number is compared with a preset flow regime discrimination threshold.

[0105] By comparing the results, the output indicates whether the current flow channel is in a laminar or non-laminar state. The resistance characteristics of fluids differ under different flow states; therefore, this determination is a prerequisite for selecting the correct pressure drop inversion model.

[0106] Step 202: When the flow regime discrimination result indicates a laminar flow state, the equivalent diameter of the channel is obtained by inversion based on the laminar pressure drop model, the pressure difference between the two ends of the channel in the fluid state observation sequence, and the volumetric flow rate of the leakage channel.

[0107] For example, when the Reynolds number is less than or equal to a preset flow regime discrimination threshold, the flow field is dominated by viscous forces, with parallel streamlines and no lateral mixing. In this case, the laminar pressure drop model is invoked, specifically using the Poiseuille pressure drop physical formula.

[0108] In this model, the pressure drop exhibits a strictly linear relationship with the volumetric flow rate and is inversely proportional to the fourth power of the equivalent diameter. By substituting the measured pressure difference across the channel and the volumetric flow rate of the leakage channel into the laminar flow pressure drop model, the equivalent diameter of the channel can be solved through algebraic rearrangement.

[0109] Step 203: When the flow regime discrimination result indicates a non-laminar flow state, the equivalent diameter and equivalent friction of the channel are obtained by jointly solving based on the Darcy-Mohr pressure drop model, the pressure difference at both ends of the channel and the volumetric flow rate of the leakage channel.

[0110] For example, when the Reynolds number exceeds a preset flow regime threshold, the flow field enters a transitional or turbulent state, where inertial forces dominate and energy dissipation increases. In this state, the laminar pressure drop model is no longer applicable. The system then uses the existing Darcy Morse pressure drop model for processing.

[0111] In this physical model, the pressure drop across the channel is not only related to the channel's equivalent diameter but also subject to nonlinear modulation by the equivalent frictional resistance. Since the volumetric flow rate and pressure difference are known observational inputs, while the channel's equivalent diameter and equivalent frictional resistance are unknowns, direct solution would lead to an ill-conditioned inverse problem. Therefore, a joint solution strategy is needed to obtain these two parameters and capture the channel cross-sectional characteristics under turbulent conditions.

[0112] Step 204: Use at least one of the following strategies to close the equation:

[0113] Optionally, one of the two processing strategies can be dynamically selected or parallel computation can be used for cross-validation. The two strategies provide supplementary computational conditions from the perspective of time continuity or empirical statistical laws, so that the number of unknowns matches the number of independent equations, ensuring the convergence of the inversion algorithm and the uniqueness of the solution.

[0114] Step 204a: Use the equivalent friction at the previous sampling time to approximate the equivalent friction at the current time;

[0115] The initial equivalent friction value is preset at the first sampling time.

[0116] In one possible strategy, the time step of the high-frequency sampling control system is short, and within adjacent millisecond-level time windows, the roughness of the channel wall and the fluid flow state do not change abruptly. Based on this, the equivalent friction value calculated at the previous sampling moment is extracted from the storage unit and forcibly assigned to the equivalent friction variable in the current calculation cycle.

[0117] Through this process, the Darcy Mohr resistance drop model is simplified from a binary equation to a univariate equation containing only the equivalent diameter of the channel, and the current geometric parameters can be solved by root-finding operations.

[0118] Step 204b, or, an empirical constitutive relation between a pre-configured friction coefficient and the Reynolds number is introduced to aid in the solution.

[0119] Another possible strategy is to use the Brassius empirical formula or the Colbrook empirical formula. These formulas establish a deterministic mapping function relationship between equivalent friction and the Reynolds number.

[0120] Since the Reynolds number is itself a function of the channel's equivalent diameter, introducing this constitutive relation allows us to combine it with the Darcy Morse resistance voltage drop model to form a closed nonlinear equation system. Solving this system of equations using an iterative algorithm can also yield the channel's equivalent diameter.

[0121] It should be understood that this strategy is suitable for scenarios where the control step size is large or the flow pattern fluctuates drastically.

[0122] Step 205: Calculate the average flow velocity of the channel based on the equivalent diameter of the channel and the volumetric flow rate of the leakage channel.

[0123] After obtaining the equivalent diameter of the channel, the system performs a transformation of the kinematic state variables. Based on the principle of fluid dynamics continuity, the system's computational unit calculates the ratio by dividing the volumetric flow rate of the leakage channel by the cross-sectional area of ​​the channel calculated from the equivalent diameter.

[0124] Based on this, the ratio represents the average flow velocity within the channel cross-section. This velocity parameter can be used to calculate the kinematic time window of the material within the channel and the shear strength it experiences.

[0125] Step 206: Use the channel equivalent diameter as the real-time equivalent geometric parameter, the channel average flow velocity as the flow field characteristic parameter, and store the equivalent friction for subsequent iterative solution of time step.

[0126] Optionally, the equivalent channel diameter obtained from the flow splitting calculation can be directly assigned to the real-time equivalent geometric parameters, while the average channel velocity can be directly assigned to the flow field characteristic parameters. This allows the subsequent residence time estimation and phase transition time estimation processes to obtain the updated physical boundary conditions.

[0127] Step 207: The equivalent diameter of the channel is used as a latent variable that evolves over time, and the discrete-time recursive update model is used to correct it in real time.

[0128] In actual high-head hydraulic engineering projects, the inner wall of the seepage channel will erode and peel off under continuous high-velocity hydraulic scouring. At the same time, local deposition of grout may occur in the early stage of grouting, causing dynamic deformation of the channel size. If no correction is made, the error of the fixed inversion model will accumulate over time.

[0129] Based on this, the channel size is defined as a latent variable that changes over time. A discrete-time recursive update model from control theory is introduced to smooth and compensate the geometric parameters initially derived from the voltage drop model in the time dimension.

[0130] Step 208: The discrete-time recursive update model is based on the channel equivalent diameter at the previous moment, and the net compensation amount of scouring and deposition, which is jointly determined by the leakage channel volume flow rate and the pressure difference at both ends of the channel, is superimposed to obtain the corrected channel equivalent diameter.

[0131] Accordingly, incremental compensation calculations are performed on the inversion results, as follows:

[0132] D _eq_t =D _eq_t_minus_1 +k _D ×g _function ;

[0133] Among them, D _eq_t D is the corrected equivalent diameter of the channel at the current moment. _eq_t_minus_1 k is the channel equivalent diameter calculated and saved in the previous time step. _D The preset damping coefficient, g, is used to characterize the erosion resistance of the soil and rock material in the channel. _function It is the net compensation amount for scouring and deposition effects, which is determined by the volumetric flow rate of the leakage channel and the pressure difference between the two ends of the channel.

[0134] In one implementation, g _function The output value and the fluctuation trend of volumetric flow rate have a corresponding mapping relationship, that is:

[0135] When the volumetric flow rate increases dramatically but the pressure differential does not increase proportionally, g _function The output is a positive value, indicating that the cross-section of the channel is increased due to the expansion and scouring of the channel.

[0136] When the volumetric flow rate shows a relatively rapid decreasing trend, g _function A negative output value indicates that deposition blockage has occurred within the channel, leading to a reduction in cross-section.

[0137] Based on this, the calculated D _eq_t Replace the original inversion values ​​with the input values ​​of the next level of calculation unit.

[0138] In another implementation, g _functionThis can be achieved using a linear combination function constructed from the weighted difference between the rate of change of volumetric flow rate and the rate of change of pressure difference. The appropriate function form can be selected based on the actual soil and rock material and hydraulic scour characteristics of the channel to reflect the net effect of scour and deposition.

[0139] The above helps the inversion results follow the evolution trajectory of the channel structure, avoiding the derailment of closed-loop control caused by physical deformation of the channel.

[0140] In a further embodiment, after completing the inversion of leakage channel parameters, the residence time and phase transition time of the slurry are calculated based on the channel geometry, flow field velocity, and thermodynamic environment, respectively, to construct physical criteria for control decisions. The corresponding optional implementation methods are as follows:

[0141] Step 301: Extract the equivalent length of the leakage channel from the real-time equivalent geometric parameters, and extract the average flow velocity of the channel from the flow field characteristic parameters;

[0142] In this step, the system's computation module performs data slicing and extraction on the physical state array. Specifically, the system separates the equivalent channel length, representing the spatial distance of fluid motion, and the average channel velocity, representing the fluid propulsion intensity.

[0143] The average flow velocity in the channel directly determines the motion hysteresis characteristics of fluid micro-particles traversing the aforementioned spatial range.

[0144] Alternatively, the equivalent length of the leakage channel can be extracted from the initial configuration data.

[0145] Alternatively, the channel equivalent length can be extracted from the channel nominal parameters in the initial configuration data.

[0146] Step 302: Determine the dwell time window based on the quotient of the channel equivalent length and the channel average flow velocity.

[0147] In practice, the system uses division operations to process the extracted physical quantities.

[0148] In some embodiments, τ _res =L _eq / v;

[0149] Where, τ _res For the residence time window, L _eq denoted as , where is the equivalent length of the channel obtained from the inversion, and v is the average flow velocity of the channel obtained from the inversion.

[0150] Furthermore, the calculated quotient is directly assigned to the residence time window variable.

[0151] Step 303: Extract the phase transition temperature from the pre-configured initial configuration data and calculate the supercooling by combining it with the fluid water temperature in the fluid state observation sequence;

[0152] The curing rate of phase change materials depends on thermodynamic boundary conditions.

[0153] In this step, the phase transition temperature, which is pre-stored in the configuration unit, is obtained; the phase transition temperature is determined by the material formulation.

[0154] Read the fluid water temperature fed back by the real-time water temperature sensor;

[0155] The difference between the phase change temperature and the fluid water temperature is calculated and defined as the supercooling, which reflects the intensity of the thermodynamic potential difference between the water and the phase change material in the leakage channel. It is an environmental factor that affects the rate of phase change solidification.

[0156] Step 304: Extract the channel average velocity from the flow field characteristic parameters and the channel equivalent diameter from the real-time equivalent geometric parameters, and calculate the shear rate characterization value based on the ratio between the two.

[0157] In high-velocity dynamic water environments, the solidification of phase change materials is not only controlled by heat conduction but also affected by the strong shear forces of the fluid. Shear forces can disrupt crystal nucleus formation and accelerate the loss of unsolidified material. Therefore, the average flow velocity and equivalent diameter of the channel can be extracted as inputs for evaluating the shear effect.

[0158] Step 305: Divide the average flow velocity of the channel directly by the equivalent diameter of the channel, and use the resulting quotient as the shear rate characterization value of the slurry inside the seepage channel, thereby replacing the flow field shear tensor calculation.

[0159] In three-dimensional flow fields, calculating the shear tensor requires significant computational resources, making it difficult to meet the demands of real-time high-frequency control. A dimensionality reduction approximation strategy can be employed, directly calculating the ratio of the average flow velocity in the channel to the equivalent diameter of the channel to generate a scalar, which maintains a positive correlation with the maximum shear stress near the pipe wall.

[0160] It should be understood that this scalar can be used as a characterization of the shear rate.

[0161] Step 306: Input the supercooling and shear rate characterization values ​​into the empirical curing model with dimensional physical constraints, and perform coupled calculations with the pre-calibrated coefficients in the characteristic model parameters to obtain the phase change curing time window.

[0162] In other words, multiphysics data is used to quantitatively extrapolate the actual duration of material phase transitions. During the calculation, the values ​​of supercooling and shear rate are obtained, and pre-calibrated coefficients acquired in the offline phase are used.

[0163] Specifically, in one embodiment τ _pc =θ _0 +θ _1 ×max(ΔT,ε)+θ_2 ×γ;

[0164] Where, τ _pc Corresponding to the phase change curing time window, θ _0 Corresponding to the basic curing time constant, θ _1 ε is the subcooling correction factor, ΔT corresponds to the calculated subcooling, ε corresponds to a small positive constant that ensures the subcooling input is not lower than the minimum effective value, and θ is the subcooling correction factor. _2 The corresponding shear correction coefficient, γ, is the calculated shear rate characterization value.

[0165] Based on this, the system outputs the phase change curing time window under the current multi-field coupling environment.

[0166] Step 307: Divide the residence time window by the phase change curing time window to calculate the capture index, which characterizes the degree of physical-time matching.

[0167] The capture index is compared with a preset capture index threshold to determine the result;

[0168] When the capture index is greater than or equal to the capture index threshold, it is confirmed that the current environment is feasible for flow rate control, and the capture index is used as a control criterion index to trigger and guide the generation of adaptive control vectors.

[0169] In this step, the system executes a division command, using the residence time window as the numerator and the phase change curing time window as the denominator, to calculate the ratio between the two. The result is the capture index, which reflects the relative probability that the slurry is trapped on the channel wall.

[0170] After obtaining the real-time capture index, a decision needs to be made in conjunction with safety constraints. The controller's internally preset capture index threshold is read, and relational calculations and comparisons are performed.

[0171] In some scenarios, the preset capture index threshold can be set to a safety factor greater than 1.0.

[0172] Furthermore, when the capture index is less than the capture index threshold, the capture index is used as a control criterion and input into the generation step of the adaptive control vector to drive parameter regulation.

[0173] For example, within a certain calculation cycle, the online inversion method is used to obtain the relevant characteristic parameters of the equivalent length of the leakage channel and the average flow velocity of the channel, and the corresponding residence time window index is obtained based on the residence time calculation logic;

[0174] The environmental supercooling and shear rate related characterization parameters were obtained, and the phase change curing time window index was calculated by combining the pre-calibrated model coefficients with the empirical curing model.

[0175] Based on the residence time window and the phase change curing time window, the capture index is calculated and compared with the preset capture index threshold.

[0176] If the capture index fails to reach the preset threshold, it is determined that the current environment cannot guarantee the retention and solidification of the phase change material, posing a risk of the slurry being washed away. Therefore, it is not considered feasible for flow rate control. Instead, the capture index under this negative deviation state is used as a control criterion and input into the system control generation module. This deviation will directly guide the generation of control vectors that shorten the phase change time or extend the residence time, closing the entire control loop.

[0177] Conversely, if the calculated capture index meets or exceeds the capture index threshold requirement, it is deemed feasible to maintain or proceed with the grouting operation.

[0178] Based on the above embodiments, the specific implementation process of the rheological retention determination method under extreme shear force environment is described, namely:

[0179] Step 401, the method also includes a feasibility assessment of flow rate control in response to extreme shear stress environments. When a time window ratio is not used, this is specifically achieved through the following methods:

[0180] In this step, parallel logical decision branches are allocated within the system controller. Under certain extreme engineering conditions, such as a sudden increase in pressure difference at both ends of the leakage channel causing the flow velocity to exceed the applicable upper limit of the empirical time window model, or when the latent heat of crystallization of the phase change material is strongly disturbed by the boundary fluid and causes an out-of-model delay, a single time window matching model may produce calculation errors.

[0181] To improve the system's engineering applicability, an optional decision-making mechanism was configured. This mechanism directly performs mechanical equilibrium calculations from the shear-yield dimension of fluid mechanics and materials rheology, avoiding the parameter drift risk of thermodynamic time window models under extreme conditions, and adding a physical dimension to the seepage prevention decision.

[0182] Step 402: Calculate the dynamic water shear rate inside the leakage channel based on the flow field characteristic parameters;

[0183] Optionally, the average flow velocity and equivalent diameter of the channel are extracted. Based on the fluid dynamics pipe flow model, the velocity gradient at the fluid boundary is calculated using the above parameters and taken as the dynamic water shear rate.

[0184] In specific calculations, a linear transformation relationship can be used. Multiplying the average flow velocity in the channel by a predetermined flow coefficient and then dividing by the equivalent diameter of the channel yields a characteristic scalar quantity reflecting the hydraulic drag strength. This value quantifies the kinematic origin of the destructive effect of the fluid on the grouting material.

[0185] Step 403: Obtain the pre-stored dynamic yield stress parameters of the phase change material;

[0186] In other words, the dynamic yield stress parameters of the phase change material are read from the offline storage unit.

[0187] When the external shear stress applied by the water flow is lower than this critical value, the phase change material exhibits solid-like mechanical retardation characteristics and can maintain its physical form to resist the impact of the water flow.

[0188] When the external shear stress exceeds this critical value, the material undergoes shear thinning and flow stripping.

[0189] In some alternative implementations, taking into account the effect of temperature on rheological properties, the dynamic yield stress parameter can be represented as a temperature-dependent lookup table. Further, the real-time measured fluid water temperature or injection temperature is obtained, and the matching dynamic yield stress value is extracted from the lookup table using a linear interpolation algorithm.

[0190] Step 404: Calculate the retention criterion index based on the relationship between the hydraulic drag effect applied to the injected fluid by the dynamic water shear rate and the shear resistance provided by the dynamic yield stress.

[0191] In this step, the destructive force and resistance force can be quantitatively combined to construct a dimensionless index for determining the stability of the sealing body. The calculated hydrodynamic shear rate is converted into equivalent shear stress and used as the denominator; the extracted dynamic yield stress parameter is used as the numerator. The specific calculation process is implemented by the following formula:

[0192] RI _yield =τ _y / (μ _w ×γ _w );

[0193] Among them, RI _yield τ is the retention criterion index obtained through calculation. _y The dynamic yield stress parameters of the phase change material are obtained to characterize the material's shear resistance. _w γ is the pre-configured dynamic viscosity of the leakage fluid. _w The calculated dynamic water shear rate is given. The denominator in the formula represents the hydraulic drag effect exerted by the dynamic water shear rate on the injected fluid. By solving for the quotient of the two, the data abstraction of the physical-mechanical game state is completed.

[0194] Step 405: Output the retention criterion index, and use it as the control criterion index in the main control process.

[0195] After calculating the retention criterion index, the decision logic unit is activated. The system retrieves the preset retention safety threshold stored in the internal register.

[0196] For example, for engineering redundancy, the preset retention security threshold is usually set to a security constant greater than 1.0, such as 1.5.

[0197] Furthermore, the calculated retention criterion index is compared with the threshold. When the retention criterion index is greater than or equal to 1.5, it is confirmed that the yield resistance of the current phase change material itself is sufficient to overcome the drag shear force of the high-velocity water flow, thus the retention condition is met.

[0198] Once this condition is met, the criterion index output will be retained and used as the control criterion index in the main control flow. The system control generation module will then generate instructions based on this mechanical dimension index, thereby maintaining the data continuity of the closed-loop control flow even without time-matching calculations.

[0199] In other embodiments, a method for prioritizing the generation of multidimensional control commands after acquiring control criterion indicators is described.

[0200] The adaptive control vector contains at least the following control components:

[0201] The injection temperature component used to adjust the initial thermodynamic state of the slurry;

[0202] Injection rate component used to control the release rhythm of slurry space;

[0203] Nucleation triggering intensity component used for real-time intervention in the slurry phase change process;

[0204] The nucleation triggering strength component can be achieved by controlling the output flow rate of the nucleating agent metering pump installed on the grouting pipeline.

[0205] In this step, the single grouting action is decoupled and reduced in dimension to generate a multi-dimensional adjustment command set composed of multiple orthogonal physical dimensions. Among them, the injection temperature component is adjusted by controlling the output power of the heating device to regulate the initial thermodynamic state of the grout; for example, the initial heating temperature of the phase change material is set to 150℃ to increase the flow heat capacity of the grout in a cold water environment.

[0206] The injection rate component, by controlling the volume displacement rate of the pumping equipment, determines the mass of material entering the leakage channel per unit time.

[0207] The nucleation trigger intensity component is used to change the kinetic activation energy of phase change material crystallization by adjusting the mixing ratio of external nucleating agents or catalysts.

[0208] It should be understood that this step establishes complete control over the thermodynamics and kinematics of the grouting process through the coordinated output of the three components.

[0209] In some alternative implementations, the adaptive control vector may also include a geometrical spatial component for controlling the physical distance between the grout outlet of the grouting pipe and the inlet of the seepage channel. For example, the system adjusts the grout outlet distance via a servo motor. When this distance is set to 3 cm, the grout can be drawn into the channel with a better flow pattern, avoiding surface cooling and floating due to excessive distance or being washed out by the water flow due to insufficient distance.

[0210] Furthermore, an adaptive control vector is generated based on the control criterion index, including a priority control strategy, namely:

[0211] When the control criterion index does not reach the preset flow rate control feasibility threshold, the adaptive control vector is adjusted first to shorten the phase change curing time window.

[0212] When the adjusted phase change curing time window reaches the preset physical lower limit of the material and the control criterion index still does not meet the requirements, a flow control command is generated to temporarily reduce the average flow rate of the channel, and the residence time window is extended until the control criterion index meets the preset conditions.

[0213] Here, the physical lower limit of the material is the minimum curing time threshold, and the flow rate control feasibility threshold corresponds to the capture index threshold.

[0214] Accordingly, the system executes control actions based on the control criterion index as the feedback error source. Asymmetric control priorities are further configured. The injection temperature component and the nucleation trigger intensity component are prioritized, thereby shortening the phase change curing time window by altering the material's thermodynamic sensitivity.

[0215] When the phase change curing time window cannot be further compressed due to the inherent properties of the material, the system activates the auxiliary control layer. It outputs flow control commands to activate the flow channel shut-off device or pressure reducing valve, forcibly reducing the average flow velocity in the channel through mechanical damping, and extending the residence time window by utilizing the inverse mapping relationship between velocity and residence time.

[0216] This asymmetric strategy reduces disturbance to external water conservancy facilities while ensuring the success rate of flow velocity control.

[0217] Based on this, an adaptive control vector is generated according to the control criterion index, and it also includes a conservative degradation strategy based on inversion uncertainty, namely:

[0218] Based on the jump magnitude of real-time equivalent geometric parameters within adjacent time steps, the confidence level of the current inversion result is evaluated, and an inversion consistency index is generated.

[0219] When the inversion consistency index is lower than the preset consistency security threshold, the conservative control mode is automatically triggered.

[0220] In conservative control mode, the injection rate component in the adaptive control vector is subject to step size decay limit, and the control command verification time is extended simultaneously to avoid slurry shedding due to aggressive injection when the channel state is uncertain.

[0221] In some scenarios, under the scouring of high-velocity flowing water, there is a physical possibility that the seepage channel may suddenly collapse or expand. The system calculates the time derivative of the real-time equivalent geometric parameters over multiple consecutive time steps to obtain the amplitude of the jump within adjacent time steps.

[0222] Furthermore, the system calculates the ratio of the preset reference constant to the jump amplitude to obtain the inversion consistency index. When this inversion consistency index is lower than the preset consistency safety threshold, it indicates that the current signal-to-noise ratio is low and the inversion result exhibits severe oscillations. At this point, the system strips the standard closed-loop control logic and switches to conservative control mode. The update step size of the adaptive control vector is attenuated by 0.5 times, and the acquisition confirmation period is increased. This is used to avoid high-speed erroneous grouting induced by incorrect inversion input.

[0223] Furthermore, the actuator is driven by an adaptive control vector to adjust the injection parameters until a preset flow rate control termination condition is met, including:

[0224] During the process of adjusting the injection parameters of the drive actuator, the latest fluid state observation sequence is continuously collected;

[0225] The latest fluid state observation sequence is used as dynamic input to trigger the recalculation of real-time equivalent geometric parameters, residence time window, phase change solidification time window and control criterion index at high frequency;

[0226] The adaptive control vector is dynamically corrected based on the recalculated and updated control criterion index to form a closed-loop control for process data reinjection until the fluid state observation sequence meets the flow rate control termination condition that the flow rate in the characterization channel is reduced to below the preset target flow rate threshold.

[0227] In this step, while outputting commands, the system controls the analog-to-digital converter to continuously acquire flow rate, differential pressure, and water temperature information from the sensors. The physical field changes caused by the control output results of the previous moment are used as the current fluid state observation sequence and re-input into the system inversion and time estimation module.

[0228] Through high-frequency iteration in the state space, the control error is continuously approximated and converged until the detected quantity reaches the preset flow rate control termination condition. This closed data chain constitutes the adaptive optimization kernel in seepage prevention engineering.

[0229] According to one aspect of this application, the preset flow rate control termination condition includes a steady-state verification and diagnostic mechanism, specifically:

[0230] After ceasing the execution of the adaptive control vector, the fluid state observation sequence is continuously acquired;

[0231] The average channel velocity obtained by inverting the fluid state observation sequence is compared with the preset target velocity threshold to calculate the velocity attenuation index.

[0232] When the average flow velocity in the channel decreases below the target flow velocity threshold, and the duration of this state meets the preset steady-state holding time threshold, the flow velocity control target is confirmed to have been achieved.

[0233] When the average flow velocity in the channel does not decrease below the target flow velocity threshold or the duration is insufficient, the failure pattern attribution logic is triggered. Based on the historical deviation trajectory between the residence time window and the phase change curing time window, a failure reason label is generated, and the equivalent geometric parameters at the current cutoff time are extracted and output as reprocessing suggestion parameters.

[0234] Accordingly, the system makes a termination decision based on the channel average flow velocity obtained from the inversion, and the specific process is as follows:

[0235] Continuous inversion of the current channel average velocity v _current Compare it with the preset target flow rate threshold v _target Compare;

[0236] The judgment logic is: when v _current ≤v _target For example, v _target When the flow rate is set to 2 m / s and the low flow rate is maintained for ≥600 seconds, it is determined that the phase change block in the leakage channel has a stable flow rate control capability, and the flow rate control target is confirmed to be achieved.

[0237] When the above conditions are not met, a diagnostic classification algorithm is executed; that is, the time-series database is retrieved to analyze the historical deviation trajectory between the residence time window and the phase change solidification time window. For example, if the historical trajectory shows that the phase change time lags behind the residence time, a second type of failure label is generated, indicating that the slurry was flushed out before the phase change; if the trajectory shows that the phase change time is abnormally short of the residence time, a first type of failure label is generated, indicating that the slurry is blocked at the channel inlet. The current equivalent geometric parameters are further extracted and output as reprocessing suggestion parameters.

[0238] This invention employs a flow-adaptive inversion mechanism based on external observation sequences to transform underground seepage channel parameters into precisely calculable equivalent geometric and flow field characteristic parameters. Furthermore, to address the model failure problem caused by dynamic channel deformation due to water scouring, a discrete-time recursive update compensation mechanism with the equivalent diameter as a latent variable is introduced. This suppresses error accumulation during long-term continuous grouting. When the inversion signal-to-noise ratio is low, a conservative degradation defense strategy is employed to enhance the system's engineering robustness under harsh hydrodynamic conditions.

[0239] Based on the above embodiments, an offline calibration method for obtaining empirical model parameters of phase transition time windows is described, including the construction of a physical simulation platform and the method for obtaining a multiphysics coupling calibration dataset. Specifically, this includes:

[0240] Step 601: Obtain solidification observation records of phase change materials under different water temperature boundary conditions and fluid shear environments, and construct a cleaning calibration dataset;

[0241] Since it is difficult to directly visualize and conduct large-scale destructive tests on the internal seepage channels of high-head dams, the underlying physical data is obtained in advance by relying on an indoor closed-loop seepage sealing simulation test platform.

[0242] Specifically, the physical simulation platform consists of an inlet tank, an outlet tank, a quartz tube, a water storage tank, a pipeline centrifugal pump, valves, and an electromagnetic flow meter. The inlet tank uses a pull-out adjustable backplate, which, together with the pipeline centrifugal pump, controls the upstream water level. A transparent quartz tube is used to simulate the leakage channel.

[0243] For example, a quartz tube with an inner diameter of 0.05 m and a length of 1.2 m can be used for calibration tests. In terms of fluid dynamic boundary settings, an electromagnetic flowmeter is installed in the full pipe section, and the flowmeter reading is controlled to 14.14 cubic meters per hour by adjusting the valve opening of the centrifugal pump in the pipeline, thereby maintaining a predetermined high-velocity flow field characteristic with an average flow velocity of 2.0 m / s inside the quartz tube.

[0244] After constructing the test environment, multiple rounds of iterative grouting and release were performed. The phase change material was specifically formulated using a paraffin-based material. During the injection process, different thermodynamic and geometric injection parameters were traversed to obtain the curing characteristics under different operating conditions. For example, the initial temperature of the phase change grout was increased to 150°C by adjusting the heating device, extending its flow time in a cold water environment and preventing premature solidification at the grouting pipe outlet. Simultaneously, the geometric distance between the grouting pipe outlet and the seepage channel inlet was dynamically adjusted.

[0245] When the distance is set to 5.0 cm, the grout undergoes surface cooling and rises before entering the channel, preventing it from entering the leakage channel. When the distance is set to 1.0 cm, the grout enters the channel quickly, but the sealing body lacks stability and is easily washed away as a whole. When the distance is set to 3.0 cm, the grout enters the channel with the water flow and accumulates at the outlet, triggering spontaneous sealing. Record the water temperature, flow rate, injection parameters, and corresponding physical phenomena for each test batch.

[0246] Furthermore, physical phenomena are transformed into failure type identification tags. Specifically, these include the following four types of tags:

[0247] Type I labels indicate that the slurry floated to the surface or underwent a premature phase change before entering the seepage channel;

[0248] Type II labels indicate that grout entered the seepage channel but was carried through the channel by the water flow and was flushed out as a whole;

[0249] Type III label indicates that the slurry undergoes a phase change and forms a plug, but the entire plug is washed away by the water flow;

[0250] Type IV labels indicate that the phase change mainly occurs outside the inlet of the leakage channel or at non-target locations.

[0251] Based on this, raw records containing multidimensional parameters and failure labels are collected. After removing abnormal samples with sensor packet loss and missing records, a cleaned and calibrated dataset is obtained for subsequent algorithm training.

[0252] Step 602: Based on the cleaning calibration dataset, perform parameter identification and fitting, construct an empirical model of phase transition time window, and extract the pre-constructed characteristic model parameters from it.

[0253] After obtaining the cleaning and calibration dataset, an offline mathematical modeling process is performed. Using curing time as the target variable and supercooling and shear rate as input independent variables, the least squares method or multiple nonlinear regression algorithm is applied to perform surface fitting on the sample set.

[0254] During this process, τ _pc =θ _0 +θ _1 ×max(ΔT,ε)+θ _2 ×γ;

[0255] Where, τ _pc The phase change curing time window is given by ΔT, where ΔT is the supercooling, ε is a constant, γ is the shear rate, and θ is the shear rate. _0 θ _1 With θ _2 These are all model coefficients to be solved.

[0256] In some embodiments, an iterative optimization algorithm is used to find the coefficient combination that minimizes the sum of squared residuals. After identification, the converged θ is extracted. _0 θ _1 With θ _2 Constant values ​​are defined as parameters of a pre-constructed characteristic model. This reorganization of constants solidifies the phase transition time evolution of paraffin-based phase change materials under dynamic water scouring conditions.

[0257] Based on this, the parameters of this characteristic model are written into the non-volatile storage unit of the online adaptive controller, serving as the physical prior configuration for executing the dynamic matching closed-loop control of online residence time and phase transition time. Through the physical environment construction and offline identification process, the trial-and-error experience of materials at the laboratory scale can be transformed into a basis for guiding the high-frequency adaptive controller.

[0258] This application constructs a dynamic matching mechanism between the residence time window and the phase change curing time window. By coupling thermodynamic phase change behavior with hydrodynamic boundary conditions, an adaptive closed-loop control is implemented that prioritizes shortening the curing time and assists in reducing the flow velocity, reversing the blind trial-and-error mode of traditional fixed-parameter grouting. This allows the phase change material to solidify and remain in the target area before being stripped away by strong shear water flow, forming a barrier within the seepage channel. This reduces the water flow velocity within the channel from a high velocity range exceeding the capabilities of traditional grouting methods to a low velocity range (below 2 m / s) suitable for traditional grouting methods, creating hydrodynamic preconditions for subsequent seepage prevention treatment using conventional grouting materials. For smaller seepage channels or cracks, the phase change grout can also directly seal them during this process. This invention improves the emergency response capability and long-term structural safety of hydraulic engineering seepage prevention systems.

[0259] Based on the aforementioned embodiments, an executable control rule for a priority control strategy is provided, such as a tiered control table, i.e.:

[0260] When the capture index is less than 0.5, it is judged to be in a severely insufficient state. The injection temperature component is set to the maximum allowable value, the nucleation trigger intensity component is set to the maximum output, and the flow control command is started.

[0261] When the capture index is between 0.5 and 0.8, the nucleation trigger intensity component is increased by one level by incrementing the preset temperature step based on the current injection temperature;

[0262] When the capture index is between 0.8 and a preset threshold, only the nucleation trigger intensity component is adjusted.

[0263] In another possible implementation, the generation of the adaptive control vector can also adopt the following step-by-step control logic:

[0264] When the capture index does not reach the preset threshold, the system adjusts the injection temperature component according to the capture index deviation. The larger the deviation, the larger the adjustment. The injection temperature increment is proportional to the capture index deviation, and the proportionality coefficient is the preset temperature control gain.

[0265] If the phase change curing time window is not shortened sufficiently after temperature adjustment, the nucleation triggering intensity component will be increased according to the preset nucleation intensity increment step.

[0266] If the above two adjustments still fail to achieve the target capture index, the system outputs a flow control command to reduce the injection rate component.

[0267] In one alternative implementation, the temperature control gain can be determined through engineering experiments based on the temperature-sensitive characteristics of the specific phase change material.

[0268] In other embodiments, step 206 may also be set as:

[0269] The equivalent diameter of the channel and the pre-configured equivalent length of the channel in the initial configuration data are used together as the real-time equivalent geometric parameters, and the average flow velocity of the channel is used as the flow field characteristic parameter.

[0270] Specifically, the equivalent channel diameter calculated from the flow split is assembled together with the pre-configured equivalent channel length and assigned to the real-time equivalent geometric parameters. Simultaneously, the average channel velocity is directly assigned to the flow field characteristic parameters. The equivalent channel length originates from the nominal channel parameters in the initial configuration data and is considered a known static geometric boundary in this implementation.

[0271] The preferred embodiments of the present invention have been described in detail above. However, the present invention is not limited to the specific details in the above embodiments. Within the scope of the technical concept of the present invention, various equivalent transformations can be made to the technical solutions of the present invention, and these equivalent transformations all fall within the protection scope of the present invention.

Claims

1. A method for controlling the leakage velocity of phase change materials at high flow rates, characterized in that, include: Obtain the initial configuration data of the target leakage channel, as well as the characteristic model parameters of the phase change material; Real-time acquisition of fluid state observation sequences in the leakage channel; Based on the fluid state observation sequence, the real-time equivalent geometric parameters and flow field characteristic parameters of the leakage channel are inverted; Based on this, the residence time window of the injected grout in the seepage channel can be estimated; By combining the characteristic model parameters of phase change materials with the fluid state observation sequence, the phase change solidification time window of the injected slurry is estimated; The control criteria index is determined based on the ratio between the residence time window and the phase change curing time window; An adaptive control vector is generated based on the control criterion index. The actuator is then driven to adjust the injection parameters based on the adaptive control vector until the flow rate control termination condition is met.

2. The method according to claim 1, characterized in that, The fluid state observation sequence includes real-time acquisition of the volumetric flow rate of the leakage channel, the pressure difference between the two ends of the channel, and the fluid temperature; The initial configuration data includes the channel nominal parameters and the phase transition temperature of the phase change material.

3. The method according to claim 1, characterized in that, The real-time equivalent geometric parameters and flow field characteristic parameters of the inverted leakage channels include: The Reynolds number of the current fluid is calculated based on the fluid state observation sequence, and then compared with the flow state discrimination threshold to obtain the flow state discrimination result; When the flow regime discrimination result indicates a laminar flow state, the equivalent diameter of the channel is obtained by inversion based on the laminar flow pressure drop model, the pressure difference at both ends of the channel and the volumetric flow rate of the leakage channel. When the flow regime discrimination result indicates a non-laminar flow state, the equivalent diameter and equivalent friction of the channel are obtained by jointly solving the Darcy Morse pressure drop model. The average flow velocity of the channel is calculated based on the equivalent diameter of the channel and the volumetric flow rate of the leakage channel. The equivalent diameter of the channel is used as the real-time equivalent geometric parameter, and the average flow velocity of the channel is used as the flow field characteristic parameter.

4. The method according to claim 3, characterized in that, In obtaining the equivalent diameter and equivalent friction of the channel based on the Darcy Morse resistance-voltage drop model through joint solution, at least one of the following strategies is employed to achieve equation closure: The equivalent friction at the previous sampling time is used to approximate the equivalent friction at the current time. Alternatively, the relationship between the friction coefficient and the Reynolds number can be introduced and substituted into the Darcy friction pressure drop model for simultaneous solution.

5. The method according to claim 1, characterized in that, Based on real-time equivalent geometric parameters and flow field characteristic parameters, the residence time window of the injected grout in the seepage channel is estimated, including: The equivalent length of the leakage channel is extracted from the real-time equivalent geometric parameters, and the average flow velocity of the channel is extracted from the flow field characteristic parameters. The residence time window is determined based on the quotient of the channel's equivalent length and the channel's average flow velocity.

6. The method according to claim 1, characterized in that, The control criteria indicators are determined based on the ratio between the residence time window and the phase change curing time window, including: Divide the residence time window by the phase change curing time window to calculate the capture index, which characterizes the degree of physical-time matching. The capture index is compared with the capture index threshold to determine the result. When the capture index is greater than or equal to the capture index threshold, it is confirmed that the current environment is feasible for flow rate control, and the capture index is used as the control criterion.

7. The method according to claim 1, characterized in that, An adaptive control vector must contain at least the following control components: The injection temperature component used to adjust the initial thermodynamic state of the slurry; Injection rate component used to control the release rhythm of slurry space; Nucleation triggering intensity component used for real-time intervention in the slurry phase change process.

8. The method according to claim 1, characterized in that, Adaptive control vectors are generated based on control criteria indicators, including priority control strategies: When the control criterion index does not reach the feasibility threshold for flow rate control, the adaptive control vector is adjusted first to shorten the phase change curing time window. When the adjusted phase change curing time window reaches the physical lower limit of the material and the control criterion index still does not meet the requirements, a flow control command is generated to temporarily reduce the average flow rate of the channel, and the residence time window is extended until the control criterion index meets the preset conditions.

9. The method according to claim 1, characterized in that, The actuator is driven by an adaptive control vector to adjust the injection parameters until the flow rate control termination condition is met, including: During the process of adjusting the injection parameters of the drive actuator, the latest fluid state observation sequence is continuously collected; The latest fluid state observation sequence is used as dynamic input to trigger the recalculation of real-time equivalent geometric parameters, residence time window and phase change solidification time window at high frequency; The adaptive control vector is dynamically corrected based on the recalculated and updated control criterion index, forming a closed-loop control for process data reinjection until the fluid state observation sequence meets the velocity control termination condition, which represents the channel velocity decreasing below the target velocity threshold.

10. The method according to claim 1, characterized in that, Before acquiring the initial configuration data for the target leakage channels and the characteristic model parameters of the phase change material, pre-calibration for offline model building is also included, namely: Acquire solidification observation records of phase change materials under different water temperature boundary conditions and fluid shear environments, and construct a cleaning calibration dataset; Based on the cleaning calibration dataset, the parameters were identified and fitted, and an empirical model of phase transition time window was constructed to extract the characteristic model parameters.

Images