Intelligent regulation and control method and regulation and control system of regulating valve

Abstract

The invention discloses an intelligent regulation and control method and system for a regulating valve, and the method comprises the following steps: obtaining the current operation condition data of the regulating valve, including fluid parameters and valve execution states; establishing a multi-physics field coupling model according to the operation condition data; dynamically correcting the multi-physics field coupling model by using an acoustic signal acquired by an acoustic sensor; a control instruction of the opening degree of the valve element is generated according to the corrected multi-physics field coupling model, and an actuator is driven to adjust the position of the valve element; in the valve element adjusting process, operation condition data and acoustic signals are collected in real time, and closed-loop feedback based on the condition-model-acoustic signals is formed; and when the acoustic characteristics in the acoustic signals exceed a preset threshold value, a slow opening and slow closing or segmented opening control strategy is triggered. According to the method, prediction errors can be remarkably reduced, the accuracy and reliability of a valve element opening degree control instruction are improved, and therefore the safety and stability of system operation are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of regulating valve control, in particular to an intelligent control method of a regulating valve and a control system thereof. BACKGROUND

[0002] As a key component in process control systems, regulating valves are widely used in petroleum and chemical industry, power, metallurgy and energy transportation, etc. The main function of regulating valves is to adjust fluid flow, pressure and temperature by changing the opening of the valve core, so as to meet the operation requirements under complex working conditions. The existing control method of regulating valves usually predicts the operating state of the valve based on fluid mechanics model or empirical formula, and adjusts the position of the valve core in combination with the actuator. However, in actual application, fluid parameters are affected by the coupling effect of multiple physical fields such as temperature change, pressure fluctuation and material deformation, and the traditional single physical field model often cannot fully reflect the actual working conditions, resulting in lag or deviation of the valve control command.

[0003] In addition, the traditional control method mainly relies on conventional sensors such as pressure and flow rate to obtain working condition data, and lacks sensitive identification means for abnormal disturbance of the flow field. When complex phenomena such as enhanced turbulence, local cavitation or shock wave occur during valve operation, the existing model prediction often has large errors and is difficult to correct in time, which may cause instability of valve core movement and even lead to system operation risk. SUMMARY

[0004] This section is intended to summarize some aspects of the embodiments of the present application and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the abstract and title of the specification to avoid obscuring the purpose of this section, abstract and title, and such simplifications or omissions cannot be used to limit the scope of the present application.

[0005] To solve the above technical problems, the present application provides the following technical solutions: an intelligent control method of a regulating valve, comprising the following steps: obtaining current operating condition data of the regulating valve, including fluid parameters and valve execution state; establishing a multi-physical field coupling model according to the operating condition data; using the acoustic signals collected by the acoustic sensor to dynamically correct the multi-physical field coupling model; generating a control command of the valve core opening according to the corrected multi-physical field coupling model, and driving the actuator to adjust the position of the valve core; During the adjustment of the valve core, real-time collection of operating condition data and acoustic signals is performed to form a closed-loop feedback based on working condition-model-acoustic signal; When the acoustic characteristics in the acoustic signal exceed a preset threshold, a slow opening and slow closing or segmented opening control strategy is triggered.

[0006] As a preferred embodiment of the intelligent control method for the regulating valve described in this invention, establishing the multiphysics coupling model includes the following steps: Based on the structure and operating condition data of the control valve, the type of physical field of the model is determined and the boundary conditions are defined. The type of physical field of the model includes fluid dynamics, heat conduction and valve material deformation. Construct fluid dynamics models, heat conduction models, and materials mechanics models respectively; The fluid dynamics model, heat conduction model, and materials mechanics model are coupled to form a multiphysics coupling equation system to simulate the effect of temperature on fluid viscosity, the effect of fluid pressure on the valve body, and the feedback effect of material deformation on the flow field. Among them, the fluid dynamics model outputs local flow velocity, pressure and turbulence characteristics; the heat conduction model outputs temperature distribution and heat flux density; and the materials mechanics model outputs structural deformation results.

[0007] In a preferred embodiment of the intelligent control method for the regulating valve described in this invention, the multiphysics coupling model is coupled in the following manner: The local flow velocity, pressure, and turbulence characteristics output by the fluid dynamics model are mapped to the boundary conditions of the heat conduction model to calculate the temperature distribution and heat flux density of the valve. The temperature distribution and heat flux density output from the heat conduction model are mapped together with the pressure load output from the fluid dynamics model and used as input loads to calculate the stress, strain and deformation distribution of the valve structure. The structural deformation results output from the mechanics of materials model are fed back to the fluid dynamics model to dynamically correct the flow field boundary conditions.

[0008] As a preferred embodiment of the intelligent control method for the regulating valve described in this invention, the method involves dynamically correcting the multiphysics coupling model using acoustic signals collected by an acoustic sensor, comprising the following steps: Acoustic signals from inside the valve and downstream pipeline are collected in real time using an array of acoustic sensors arranged downstream of the valve. The collected acoustic signals are feature extracted, and the acoustic features are obtained by inversion based on the correspondence between acoustic features and fluid physical quantities; The acoustic features are compared with the prediction output of the multiphysics coupling model at the same time or within the same time window. The observation residuals are calculated and mapped to the dynamic correction factor of the multiphysics coupling model. The dynamic correction factor is applied to the local pressure gradient, turbulent kinetic energy, and fluid viscosity coefficient of the multiphysics coupling model to achieve adaptive correction of the model. The fluid, heat conduction, and material mechanical properties are recalculated using a modified multiphysics coupling model to obtain updated valve condition predictions.

[0009] In a preferred embodiment of the intelligent control method for the regulating valve described in this invention, the calculation formula for the dynamic correction factor is as follows:

[0010] in: It is a dynamic correction factor used to correct local pressure gradients, turbulent kinetic energy, and viscosity coefficients; These are the spatial coordinates of the flow field, i.e., the positions inside the valve and in the downstream pipeline; For time; The instantaneous acoustic characteristic energy obtained from acoustic sensor observations; The corresponding physical quantity energy predicted by the multiphysics coupling model; The difference between acoustic observations and model predictions represents the observation residual. The relative residual is used to normalize acoustic bias. The hyperbolic tangent function is used to restrict the correction factor to a certain value. ; To correct the amplitude coefficient, control the observation residuals through The magnitude of the impact of mapping on the final correction factor; This is a nonlinear sensitivity coefficient that controls the intensity of the corrected response. It is a small constant.

[0011] As a preferred embodiment of the intelligent control method for the regulating valve described in this invention, the method for generating a control command for the valve core opening based on the modified multiphysics coupling model specifically includes the following steps: The fluid parameters output by the multiphysics coupling model are compared with the target operating condition indicators, and the deviation is calculated. Based on the deviation, a preliminary valve core opening adjustment amount is generated using a control algorithm; The dynamic correction factor is applied to the initial valve core opening adjustment amount, and the initial valve core opening adjustment amount is weighted and corrected to obtain the final valve core opening adjustment amount. The final valve core opening adjustment is converted into a control command that the actuator can recognize.

[0012] As a preferred embodiment of the intelligent control method for the regulating valve described in this invention, triggering the slow-opening / slow-closing or segmented opening control strategy includes the following steps: The acoustic features in the real-time acquired acoustic signals are compared with preset thresholds. When any acoustic feature exceeds its corresponding threshold, a slow opening / closing or segmented opening control strategy is triggered. Based on the type of acoustic feature that exceeds the threshold, select the corresponding valve core motion adjustment mode. When the turbulence intensity exceeds the threshold, the slow opening and slow closing mode is triggered. When the cavitation intensity or shock wave parameter exceeds the threshold, the segmented opening mode is triggered. Based on the current valve core position, the fluid parameters output by the multiphysics coupling model, and the acoustic dynamic correction factor, the valve core opening adjustment amount is dynamically calculated. The calculated valve core opening adjustment is converted into an electrical signal that the actuator can recognize, driving the valve core to move; During the movement of the valve core, data is continuously collected and the dynamic correction factor is updated and the control command is iteratively corrected until the acoustic characteristics are below the threshold and the target working condition is reached.

[0013] As a preferred embodiment of the intelligent control method for the regulating valve described in this invention, the slow-opening and slow-closing mode involves gradually and smoothly adjusting the valve core opening to allow the valve core to move at a small amplitude and low speed; the segmented opening mode involves dividing the valve core opening adjustment process into multiple continuous intervals, and executing different control strategies in each interval according to the current operating conditions.

[0014] As a preferred embodiment of the intelligent control method and control system for the regulating valve described in this invention, the fluid parameters include, but are not limited to, medium temperature, medium pressure, flow rate, density, and viscosity; the valve execution state includes, but is not limited to, valve core position and actuator drive signal.

[0015] This invention also provides an intelligent control system for a regulating valve, applied to the aforementioned intelligent control method for the regulating valve, comprising: The operating condition acquisition module is used to acquire real-time operating condition data of the control valve, including fluid parameters and valve execution status; The multiphysics coupling modeling module is used to build a multiphysics coupling model including fluid dynamics, heat conduction and materials mechanics based on the operating condition data, and output the predicted parameters. The acoustic sensing and dynamic correction module is used to acquire acoustic signals, extract features from the acoustic signals, and calculate dynamic correction factors based on the comparison between acoustic features and model prediction results to correct the multiphysics coupling model. The control command generation module is used to generate control commands for valve core opening based on the deviation between the output of the corrected multiphysics coupling model and the target operating condition. The actuator drive module is used to convert control commands into electrical signals that the actuator can recognize, and drive the actuator to adjust the valve core position; The closed-loop feedback module is used to continuously collect operating condition data and acoustic signals during valve core adjustment and input them into the multi-physics coupling model to update the dynamic correction factor, thereby realizing closed-loop feedback between operating condition, model and acoustic signal. The adaptive control module is used to trigger a slow-opening / slow-closing or segmented opening control strategy when the acoustic features in the acoustic signal exceed a preset threshold.

[0016] The beneficial effects of this invention are: 1. This invention introduces acoustic signals collected by acoustic sensors to dynamically correct a multiphysics coupling model. By comparing acoustic features with model prediction results, a dynamic correction factor is calculated and applied to local pressure gradient, turbulent kinetic energy, and fluid viscosity coefficient, enabling the model to reflect actual fluid conditions in real time. Compared with modeling methods based solely on fluid parameters, this method can significantly reduce prediction errors, improve the accuracy and reliability of valve core opening control commands, thereby reducing the probability of valve core motion instability and enhancing the safety and stability of system operation.

[0017] 2. In this invention, when the acoustic features in the acoustic signal exceed a preset threshold, a slow-opening and slow-closing or segmented opening control strategy is triggered. The slow-opening and slow-closing mode is triggered when the turbulence intensity exceeds the threshold, and the segmented opening mode is triggered when the cavitation intensity or shock wave parameter exceeds the threshold. The control mode can be adaptively selected according to the disturbance type, thereby reducing fluid impact and energy loss caused by the rapid movement of the valve core and ensuring the safety and stability of the valve system.

[0018] 3. This invention employs a closed-loop feedback mechanism and incorporates a dynamic correction factor during the control command generation process to weight and correct the valve core opening adjustment. This enables the system to iteratively adjust based on real-time operating conditions, thereby dynamically optimizing the valve core's movement trajectory. This approach ensures precise valve opening adjustment and enables intelligent, adaptive adjustment of the control valve under complex operating conditions. Attached Figure Description

[0019] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein: Fig. 1 This is an overall flowchart of an intelligent control method for a regulating valve according to the present invention.

[0020] Fig. 2 This is a flowchart illustrating the dynamic correction of a multiphysics coupling model in an intelligent control method for a regulating valve according to the present invention.

[0021] Fig. 3 This is a coupling diagram of a multiphysics coupling model for an intelligent control method of a regulating valve according to the present invention. Detailed Implementation

[0022] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0023] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0024] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0025] Secondly, the present invention is described in detail with reference to the schematic diagrams. When detailing the embodiments of the present invention, for ease of explanation, the cross-sectional views illustrating the device structure may be partially enlarged, not according to the usual scale. Furthermore, the schematic diagrams are merely examples and should not limit the scope of protection of the present invention. In addition, actual fabrication should include three-dimensional spatial dimensions of length, width, and depth.

[0026] Example 1 Reference Figs. 1-3 The first embodiment of the present invention provides an intelligent control method for a regulating valve, comprising the following steps: S1: Obtain the current operating condition data of the control valve, including fluid parameters and valve execution status.

[0027] It should be noted that fluid parameters include, but are not limited to, medium temperature, medium pressure, flow velocity, density, and viscosity. These parameters are the core variables in the multiphysics coupling model: temperature directly affects the coefficient of thermal expansion and viscosity changes of the medium; pressure determines the pressure difference across the valve and the sensitivity of flow regulation; flow velocity reflects the fluid's transport capacity and is closely related to cavitation, noise, and energy loss; density and viscosity are used to correct for the fluid's inertial effects and energy transfer characteristics, thereby ensuring the accuracy and stability of the multiphysics coupling model.

[0028] In practice, fluid parameters can be acquired using various conventional sensors. For example, medium temperature can be obtained using thermocouples, platinum resistance thermometers, or fiber optic temperature sensors; medium pressure can be obtained using piezoelectric pressure sensors, strain gauge pressure transmitters, or differential pressure sensors; flow velocity can be obtained using electromagnetic flow meters or ultrasonic flow meters; and medium density and viscosity can be obtained using online vibratory density meters or ultrasonic viscosity sensors.

[0029] It should also be noted that valve operation status includes, but is not limited to, valve spool position and actuator drive signal. Valve spool position can be obtained in real time via a potentiometer or motor encoder, and actuator drive signal can be directly acquired from the current or voltage signal output by the controller via an electrical interface. Valve operation status reflects the actual response of the valve and, in conjunction with fluid parameters, can be used to determine whether the actuator exhibits hysteresis, overshoot, or jamming, and provides support for achieving slow opening / closing or segmented opening control.

[0030] Furthermore, by simultaneously acquiring the aforementioned fluid parameters and valve execution status, a closed-loop feedback mechanism of "operating condition-model-acoustic signal" can be achieved. Compared with control methods that rely solely on fluid parameters or solely on valve execution status, the method of this invention can predict risks before fluid disturbances occur and achieve adaptive control by dynamically adjusting the valve core movement trajectory when disturbances occur, thereby improving the operational stability and intelligence level of the control valve.

[0031] S2: Based on the operating condition data, establish a multi-physics coupling model, and dynamically correct the multi-physics coupling model using acoustic signals collected by acoustic sensors.

[0032] Specifically, establishing a multiphysics coupling model includes the following steps: Based on the structure and operating condition data of the control valve, the type of physical field of the model is determined, and the boundary conditions are defined. The type of physical field of the model includes fluid dynamics, heat conduction, and valve material deformation. Construct fluid dynamics models, heat conduction models, and materials mechanics models respectively; By coupling the fluid dynamics model, the heat conduction model, and the materials mechanics model, a multiphysics coupling equation system is formed, thereby completing the establishment of the multiphysics coupling model and realizing the influence of temperature on fluid viscosity, the effect of fluid pressure on the valve body, and the feedback influence of material deformation on the flow field.

[0033] It should be noted that boundary conditions include, but are not limited to, inlet flow rate, outlet pressure, hot channel, fixed support conditions, motion constraints, no-slip boundary conditions, and hot channel. Specifically, inlet flow rate determines the fluid flow rate at the valve inlet, providing known conditions for fluid dynamics calculations; outlet pressure limits the valve outlet pressure for calculating the pressure difference across the valve; hot channel is a heat flux boundary set at the fluid-valve body contact surface, simulating the heat transfer effect of the fluid on the valve body; fixed support conditions are used as constraint nodes at the valve body connection to the pipeline, support surfaces, etc., with zero displacement; motion constraints apply to the valve core connected to the actuator via the valve stem, requiring constraints on its direction and range of motion; no-slip boundary conditions are typically applied to the valve body, valve core, and valve seat surfaces, meaning the fluid velocity is zero; and hot channel simulates the heat transfer effect of the fluid on the valve body. By setting these boundary conditions, the actual operating state of the valve can be accurately reflected, ensuring that the interaction between fluid, heat conduction, and structural deformation is reasonably calculated. Setting boundary conditions can also improve the model's prediction accuracy.

[0034] It should be noted that the structure of a control valve includes, but is not limited to, the valve body, valve core, valve stem, valve seat, and actuator. These structures not only have geometric structures but also material properties such as elastic modulus, Poisson's ratio, density, thermal conductivity, and coefficient of thermal expansion.

[0035] The valve body serves as the fluid-bearing and guiding structure, providing the internal flow channels and a fixed support surface for the valve core. The valve core is the key moving component for regulating flow; its positional changes directly determine the valve's diameter and flow characteristics. The valve stem connects the valve core to the actuator, enabling precise positioning and motion transmission of the valve core. The valve seat provides the sealing surface for the valve core, ensuring no fluid leakage when the valve is closed. The actuator includes electric, pneumatic, or hydraulic drive devices that generate the force or torque required for the valve core's movement and includes a position feedback device for real-time acquisition of the valve core's position. The geometry and material properties of these structures provide a direct basis for establishing a multiphysics coupling model. Specifically, the geometric dimensions and material properties of the valve body, valve core, and valve seat determine the boundary conditions and coupling relationships of the fluid dynamics field, heat conduction field, and material deformation field, while the actuator and valve stem provide constraints on the valve core's motion, ensuring that the model can reflect the fluid-structure interaction of the valve under real-world operating conditions.

[0036] In one specific implementation, the steps for establishing a fluid dynamics model include: First, based on the three-dimensional geometry of the control valve and actual operating data, a three-dimensional flow field computational domain is constructed for the valve's interior and downstream pipeline. Specifically, this includes: establishing geometric models of the valve body, valve core, and valve seat in 3D modeling software; determining the valve core position and valve seat clearance using real-time measurement data and mapping them into the computational domain to ensure the model reflects actual operating conditions; dividing the computational domain into appropriate mesh elements, including structured or unstructured meshes, and refining the mesh in flow-related regions such as the vicinity of the valve core and the valve seat clearance to ensure the accuracy and numerical convergence of the flow field calculations.

[0037] Then, within the computational domain, a fundamental fluid dynamics model is established using the continuity and momentum equations to describe fluid mass conservation and momentum transfer. These equations can be discretized using the finite volume method or the finite element method, depending on the computational requirements, to form a numerically solvable computational model.

[0038] Next, fluid parameters are substituted into the continuity and momentum equations to correct the equation coefficients. For example, viscosity affects the viscous term, and density affects the inertial term, thus reflecting the true physical properties of the fluid. When solving the momentum equation, since it contains turbulent stress terms, a turbulence model needs to be introduced for closure. Depending on the specific operating conditions, the RANS method can be used to obtain flow characteristics applicable to steady-state or weakly unsteady-state conditions through turbulent averaging calculations; alternatively, the LES method can be used to perform time-domain analysis on complex transient turbulence, thereby obtaining more accurate local flow field characteristics.

[0039] Next, set boundary conditions: set known flow rate or pressure boundary conditions at the valve inlet, and outlet pressure or flow rate boundary conditions at the valve outlet; use no-slip boundary conditions for the valve body, valve core, and valve seat surfaces; simultaneously, heat flux boundary conditions can be set to describe the heat exchange between the fluid and the valve body surface. Setting boundary conditions ensures that the model can reflect the actual flow constraints and operating conditions of the valve.

[0040] Finally, the above equations are solved using a numerical solver, and the local flow velocity, pressure and turbulence characteristics data of the valve interior and downstream pipeline are output. These output data will serve as input conditions for the subsequent coupled calculation of the heat conduction model and the material mechanics model, realizing multi-physics coupled simulation and providing basic data for intelligent valve core control.

[0041] In one specific implementation, the steps for establishing the heat conduction model include: First, based on the three-dimensional geometry of the control valve and actual operating data, a thermal computational domain is constructed for the valve body, valve core, and valve seat. The local flow velocity, pressure, and turbulence characteristics output from the fluid dynamics model are then mapped into this thermal computational domain to account for the convective heat transfer effect of the fluid on the valve. Appropriate mesh elements are then divided within the computational domain, with the mesh being fined in regions with significant changes in thermal gradient to ensure the accuracy and numerical convergence of the thermal field calculations.

[0042] Then, a basic model of heat conduction within the valve and its structure is established using the heat conduction equation, including options for steady-state and transient heat conduction calculations. The thermophysical parameters of the valve material, such as thermal conductivity, specific heat capacity, and coefficient of thermal expansion, are substituted into the heat conduction equation to calculate heat transfer and temperature changes, and the model is modified based on boundary conditions. For example, heat flux boundary conditions are applied to the valve body-fluid contact surface to simulate the heat exchange effect of the fluid on the valve; adiabatic or isothermal conditions are applied between the valve body and the fixed support surface; and heat flow constraints are applied to the connection between the valve stem and the actuator to ensure that the model reflects the actual thermal environment.

[0043] Next, the heat conduction equation is solved, and the temperature distribution and heat flux density data of each part of the valve are output, providing thermal load input conditions for the subsequent material mechanics model, and realizing fluid-thermal-structure coupled calculation.

[0044] It should be noted that the heat conduction equation is solved using numerical methods, such as the finite element method or the finite volume method, which can be used to solve the discretized heat conduction equation.

[0045] In one specific implementation, the steps for constructing the materials mechanics model are as follows: First, the calculation results of the fluid dynamics model are used as the boundary conditions input of the heat conduction model. The local flow velocity, pressure and turbulence characteristics output by the fluid dynamics model are used to describe the flow state of the fluid inside the valve and in the downstream pipeline, and are transferred to the heat conduction model as the driving force for convective heat transfer to calculate the temperature distribution and heat flux density of the valve body, valve core and valve seat.

[0046] Secondly, the temperature distribution and heat flux density obtained from the heat conduction model are mapped together with the pressure load output from the fluid dynamics model into the materials mechanics model as the input load under the combined action of heat and fluid. Based on this, the materials mechanics model calculates the stress, strain, and deformation distribution of the valve structure under actual working conditions.

[0047] Finally, the structural deformation results output by the material mechanics model are fed back into the fluid mechanics model to dynamically correct the flow field boundary conditions, realize the two-way coupling between fluid, heat conduction and structural deformation, and thus form a complete multi-physics coupling equation system, which truly reflects the fluid-structure-thermal interaction characteristics of the control valve under complex working conditions.

[0048] It should be noted that turbulence characteristics are derived quantities calculated by fluid dynamics models. Fluid dynamics models (such as RANS or LES) calculate local turbulence characteristics based on parameters such as velocity, pressure, density, and viscosity. Turbulence characteristics are used to describe local fluid disturbances and convection enhancement effects. Turbulence characteristics include, but are not limited to, turbulence intensity (here, turbulence intensity is a model output, which differs from the turbulence intensity in S3, where the S3 value is the actual observed value), turbulence kinetic energy, and eddy viscosity. Among them, turbulence intensity represents the ratio of fluid velocity fluctuations to the average velocity, used to describe the strength of local disturbances; turbulence kinetic energy reflects the magnitude of kinetic energy in turbulence; and eddy viscosity characterizes the enhancing effect of turbulence on fluid viscosity. Turbulence characteristics affect both convective heat transfer in fluid-thermal coupling and fluctuating pressure loads in fluid-structure coupling, making them indispensable parameters for multiphysics coupling.

[0049] In one specific implementation, the fluid dynamics model, heat conduction model, and materials mechanics model are coupled to form a multiphysics coupling equation system, thereby completing the establishment of the multiphysics coupling model. Specifically, the coupling includes the following: The local flow velocity, pressure, and turbulence characteristics output by the fluid dynamics model are mapped to the boundary conditions of the heat conduction model to describe the convective heat transfer between the fluid and the valve body, valve core, and valve seat, thereby calculating the temperature distribution and heat flux density of the valve. The temperature distribution and heat flux density output by the heat conduction model are mapped together with the pressure load output by the fluid dynamics model into the material mechanics model as the input load under the combined action of heat and fluid to calculate the stress, strain and deformation distribution of the valve structure. The structural deformation results output from the material mechanics model are fed back to the fluid mechanics model to dynamically correct the flow field boundary conditions, thereby realizing the bidirectional coupling between fluid, heat conduction and structural deformation, and thus forming a complete multiphysics coupling equation system.

[0050] It should be noted that, through the above coupling relationship, a multi-physics coupling equation system is formed, realizing the interaction modeling between fluid, heat conduction and structural mechanics, thereby ensuring that the established model can truly reflect the comprehensive response characteristics of the control valve under operating conditions.

[0051] Specifically, the multiphysics coupling model is dynamically corrected using acoustic signals collected by acoustic sensors, including the following steps: Acoustic signals from inside the valve and downstream pipeline are collected in real time using an array of acoustic sensors arranged downstream of the valve. The acoustic signals were sampled and their features were extracted. Based on the correspondence between the acoustic features and fluid physical quantities, acoustic features such as turbulence intensity, cavitation intensity and shock wave parameters were obtained by inversion. The acoustic features are compared with the prediction output of the multiphysics coupling model at the same time or within the same time window. The observation residuals are calculated and mapped to the dynamic correction factor of the multiphysics coupling model. The dynamic correction factor is applied to the local pressure gradient, turbulent kinetic energy, and fluid viscosity coefficient of the multiphysics coupling model to achieve adaptive correction of the model. The fluid, heat conduction, and material mechanical properties are recalculated using a modified multiphysics coupling model to obtain updated valve condition predictions.

[0052] It should be noted that turbulence intensity can be calculated by the ratio of high-frequency broadband noise energy to background energy, and is used to characterize the strength of velocity fluctuations relative to the average flow velocity in the flow field; cavitation intensity can be obtained by the statistical distribution of high-frequency pulse peaks in the acoustic signal, and is used to reflect the intensity of bubble formation, growth, and collapse processes in the fluid; shock wave parameters can be obtained by identifying short-duration high-amplitude pulses in the acoustic signal and calculating their amplitude, duration, and repetition frequency, and are used to describe the impact effect caused by cavitation collapse or high-speed fluid shock waves. By simultaneously extracting the above three acoustic features, the main physical mechanisms of flow disturbances inside the valve can be covered: turbulence corresponds to energy dissipation and flow instability, cavitation corresponds to nonlinear noise sources caused by gas-liquid phase transitions, and shock waves correspond to local high-pressure events and possible structural impacts, thereby ensuring the comprehensiveness and accuracy of the correction to the multiphysics coupling model and improving the reliability of the model in predicting the valve's operating state.

[0053] In one specific implementation, the acoustic sensor array includes multiple microphones, piezo-acoustic sensors, or underwater sound level meters, arranged in a circumferential or axial multi-point distribution to ensure sufficient sensing of local turbulence, cavitation, and shock waves. The acquired raw acoustic signals are then subjected to noise suppression, timing alignment, and preprocessing to form an acoustic data sequence that can be used for analysis.

[0054] In one specific implementation, feature extraction includes calculating acoustic spectral energy, energy components in different frequency bands, pulse peak statistics, envelope spectral characteristics, and acoustic cross-correlation or phase difference information. These acoustic features can be further used to estimate the turbulence intensity, cavitation index, and shock wave parameters downstream of the valve through prior physical relationships or training-derived mapping relationships, forming a real-time acoustic feature vector.

[0055] In one specific implementation, the observation residuals are mapped to a dynamic correction factor of a multiphysics coupling model. This mapping can be achieved through methods such as physical empirical formulas, extended Kalman filtering, particle filtering, or machine learning regression models to ensure that the correction factor accurately reflects the relationship between acoustic features and fluid physical quantities.

[0056] In one specific implementation, the corrected values ​​of local pressure gradient, turbulent kinetic energy, and fluid viscosity coefficient are: correction factor × original fluid parameters calculated by the multiphysics coupling model; for example: corrected local pressure gradient = correction factor × original local pressure gradient.

[0057] In one specific implementation, the formula for calculating the dynamic correction factor is:

[0058] in: It is a dynamic correction factor, dimensionless, used to correct local pressure gradients, turbulent kinetic energy, and viscosity coefficients. These are the spatial coordinates of the flow field, i.e., the positions inside the valve and in the downstream pipeline; Time, in seconds; The instantaneous acoustic characteristic energy obtained from acoustic sensor observations, such as an estimated value of turbulence intensity, in units of [Pa]. 2 Or, after standardization, it is dimensionless; This refers to the energy of the corresponding physical quantity predicted by a multiphysics coupling model, such as pressure or turbulent energy, with units of [missing information]. ; The difference between acoustic observations and model predictions represents the observation residual, expressed in units of 1 / 2 and 2 / 3. ; The relative residual is used to normalize acoustic bias and is dimensionless. The hyperbolic tangent function is used to restrict the correction factor to a certain value. Dimensionless; To correct the amplitude coefficient, control the observation residuals through The magnitude of the impact on the final correction factor after mapping. ; The nonlinear sensitivity coefficient controls the intensity of the corrected response. >0; Use small constants to avoid zero denominators. The value is usually taken as Or adjust according to the dimensions.

[0059] Calculating the dynamic correction factor using the above formula enables a smooth, stable, and controllable mapping of acoustic observation residuals. On one hand, the formula provides an approximately linear response when the residuals are small, ensuring the sensitivity of the correction. On the other hand, it automatically enters the saturation range when the residuals are large (meaning...). (range), to avoid over-correction and numerical divergence; at the same time, The continuous differentiability of the function ensures the smoothness of the correction process and the stability of the numerical solution. Compared with traditional correction methods such as linear amplification, exponential amplification, or piecewise limiting, the formula of this invention is not only simple to calculate and has strong real-time performance, but also has stronger robustness and physical rationality, thereby significantly improving the dynamic correction capability and prediction accuracy of multiphysics coupled models.

[0060] In one embodiment of the present invention, the amplitude coefficient is corrected. Set the sensitivity coefficient to 0.3. Take 1.0 (when the acoustic residuals are dimensionlessly processed so that the typical residual values ​​fall within approximately...). (Interval time); when the residuals are not normalized and their typical order of magnitude is At that time, it can be Set as More generally, it can be Let in the interval Inside, and will Let in the interval (Normalized residuals) or Within (unnormalized residuals). In practical applications, optimal parameter values ​​can be determined through offline calibration or online adaptive algorithms.

[0061] The corrected formulas for calculating the local pressure gradient, turbulent kinetic energy, and fluid viscosity coefficient are as follows: ; ; ; in, , and These represent the corrected local pressure gradient, turbulent kinetic energy, and fluid viscosity coefficient, respectively, in Pa / m and m. 2 / s 2 and Pa s; , and The original local pressure gradient, turbulent kinetic energy, and fluid viscosity coefficient are obtained from the calculation of the multiphysics coupling model.

[0062] In summary, dynamic correction factor By comparing the acoustic characteristics such as turbulence intensity, cavitation intensity, and shock wave parameters collected by an acoustic sensor array located downstream of the valve with the prediction output of the multiphysics coupling model at the same location and time window in real time, and after normalization and hyperbolic tangent mapping, correction quantities are formed for adaptive adjustment of local pressure gradient, turbulent kinetic energy, and fluid viscosity coefficient. This enables the fluid dynamics, heat conduction, and materials mechanics models to perform dynamic coupling calculations under the actual operating conditions of the valve, realizing the self-correction and accurate prediction of the multiphysics coupling model.

[0063] For example: Suppose the local flow field of the control valve under actual operating conditions is as follows: a certain point Local pressure gradient of fluid =5000 Pa / m; Local turbulent kinetic energy =0.2m 2 / s 2 ; Fluid viscosity coefficient =0.001Pa s; The turbulence or cavitation intensity energy observed by the acoustic sensor at the same point is:

[0064] Multiphysics coupling model prediction values ; Take the correction amplitude coefficient Sensitivity coefficient =0.0001.

[0065] Calculate the dynamic correction factor :

[0066] Note: Due to the small observation bias, the correction factor is close to 1, and local physical quantities are slightly adjusted.

[0067] The corrected local pressure gradient, turbulent kinetic energy, and viscosity coefficient are as follows:

[0068]

[0069]

[0070] Note: At this point, the difference between the model output and the observation is small, and the correction is slight. If the difference between the observation and the actual observation is large, such as in the case of turbulence anomalies, the correction factor can reach 1.1 or 0.9, achieving significant adaptive correction.

[0071] In summary, step S2, by introducing real-time acoustic signals collected by acoustic sensors for dynamic correction based on the multiphysics coupling model, can improve the consistency between the model and the actual working conditions, thereby avoiding the accumulation of deviations between model predictions and the actual state, improving the accuracy and robustness of valve state prediction and control, and providing more reliable data support for subsequent valve core opening adjustment.

[0072] S3: Generate control commands for valve core opening based on the corrected multiphysics coupling model, and drive the actuator to adjust the valve core position; Specifically, the control command for the valve core opening is generated based on the modified multiphysics coupling model, which includes the following steps: The fluid parameters output by the multiphysics coupling model, such as pressure, flow rate, and turbulence characteristics, are compared with the target operating condition parameters to calculate the deviation; for example, the deviation may include pressure deviation. Flow deviation wait.

[0073] Based on the deviation, a preliminary valve core opening adjustment amount is generated using a control algorithm. ; The aforementioned dynamic correction factor for the acoustic signal is applied to the initial valve core opening adjustment amount, for After weighted correction, the final valve core opening adjustment amount is obtained. ; Adjust the final valve core opening amount The control commands are converted into control instructions that the actuator can recognize, such as voltage, current or pulse signals. The control commands are sent in real time through the actuator interface to drive the valve core to move along a predetermined direction and amplitude, thereby achieving the target valve opening adjustment.

[0074] In one specific implementation, the final valve core opening adjustment amount The calculation formula is:

[0075] in, This is an adjustment coefficient used to control the proportion of the dynamic correction factor's influence on the valve core opening, ensuring safety and stability. The value can generally be set to 0.05-0.5, and the specific value can be determined through experimentation based on the valve type, fluid characteristics and actual control requirements.

[0076] In one specific implementation, the control algorithm can be a proportional-integral-derivative (PID) control, fuzzy control, or model predictive control (MPC) to ensure the accuracy and response speed of the valve core adjustment. For example, when using fuzzy control, the deviation and its rate of change are used as fuzzy inputs, mapped to a fuzzy rule base through membership functions, and after fuzzy inference and defuzzification, the corresponding opening adjustment is obtained. .

[0077] In summary, when generating the valve core opening control command in step S3, the fluid parameters output by the multiphysics coupling model are first compared with the target operating condition index to obtain the deviation. Then, the initial adjustment amount is calculated by the control algorithm, and after weighted correction by the acoustic signal dynamic correction factor, it is converted into an electrical signal that the actuator can recognize and send out. This realizes the real-time response of the control command to the operating condition deviation, effectively improves the accuracy and stability of valve core opening adjustment, and avoids valve overshoot, oscillation or jamming caused by model error or fluid disturbance.

[0078] S4: During the valve core adjustment process, operating condition data and acoustic signals are collected in real time to form a closed-loop feedback based on operating condition-model-acoustic signal.

[0079] In one specific implementation, during the valve core adjustment process, real-time acquisition of operating condition data and acoustic signals forms a closed-loop feedback based on operating condition, model, and acoustic signal. This means that while the actuator drives the valve core to move according to control commands, the system continuously acquires fluid parameters such as temperature, pressure, flow rate, density, and viscosity, as well as valve execution status such as valve core position and drive signal. Simultaneously, it acquires acoustic signals collected by an acoustic sensor array arranged in the downstream pipeline of the valve. The operating condition data and acoustic signals are jointly input into a multiphysics coupling model to correct the model prediction results and generate new correction factors, thereby realizing a closed-loop iterative update of operating condition, model, and acoustic signal.

[0080] For example, when an acoustic sensor array positioned downstream of the valve detects a high-frequency acoustic characteristic peak related to turbulence intensity, the multiphysics coupling model can identify abnormal fluid disturbances based on this acoustic characteristic and determine that the current trajectory of the valve core may cause instability in the local flow field. Then, by calculating a dynamic correction factor, this correction factor is applied to the control command for the valve core opening, thereby correcting the valve core's trajectory in real time and avoiding cavitation or shock wave effects caused by flow field turbulence. Through this closed-loop feedback mechanism based on operating data and acoustic signals, the present invention can continuously and adaptively correct the prediction deviation of the multiphysics coupling model during the dynamic process of valve operation, thereby significantly improving the real-time performance and stability of the valve control system and realizing intelligent and adaptive regulation of the control valve under complex operating conditions.

[0081] S5: When the acoustic characteristics in the acoustic signal exceed the preset threshold, a slow opening / closing or segmented opening control strategy is triggered to reduce fluid disturbance.

[0082] Specifically, triggering slow opening / closing or segmented opening control includes the following steps: The acoustic features in the real-time acquired acoustic signals are compared with preset thresholds. When any acoustic feature exceeds its corresponding threshold, the system determines that there is a potential disturbance risk in the current flow field and triggers a slow opening and closing or segmented opening control strategy. Select the corresponding valve core movement adjustment mode based on the type of acoustic feature that exceeds the threshold; Based on the current valve core position, the fluid parameters output by the multiphysics coupling model, and the acoustic dynamic correction factor, the valve core opening adjustment amount is dynamically calculated. The calculated valve core opening adjustment is converted into an electrical signal that the actuator can recognize, and then sent in real time through the actuator interface to drive the valve core movement. During the valve core movement, fluid condition data and acoustic signals are continuously collected and input into the multiphysics coupling model to update the dynamic correction factor and iteratively correct the control command. When the acoustic characteristics are below the threshold and the target operating condition is reached, exit the slow opening / closing or segmented opening control mode.

[0083] It should be noted that, in this invention, the preset threshold refers to the upper limit of the acoustic characteristics used to judge the risk of fluid disturbance. The value can be determined according to the valve design conditions, fluid characteristics and safe operation requirements. For example, the critical values ​​of turbulence intensity, cavitation intensity or shock wave parameters under different operating conditions can be tested experimentally and set in combination with historical operating data and engineering experience to a safe range that can respond to abnormal disturbances in a timely manner and avoid false triggering, thereby ensuring the reliability and stability of the valve control system under complex flow field conditions.

[0084] It should be noted that "slow opening and slow closing" refers to the valve control system gradually and smoothly adjusting the valve core opening when the acoustic characteristics in the acoustic signal exceed a preset threshold. This allows the valve core to move along a predetermined direction at a small amplitude and low speed, thereby reducing the risk of transient fluid disturbances, pressure shocks, and cavitation. For example, when the turbulence intensity exceeds the threshold, the valve core gradually moves from the current opening to the target opening. Each adjustment amplitude can be calculated based on the fluid deviation and dynamic correction factor, ensuring smooth valve movement and avoiding flow field turbulence caused by rapid opening and closing. "Segmented opening" refers to dividing the valve core opening adjustment process into multiple continuous intervals, each with a different control strategy or adjustment amplitude, to achieve precise control for different flow or pressure conditions. For example, during the process of the valve core moving from the initial opening to the target opening, the entire stroke can be divided into several segments. Each segment calculates the corresponding opening adjustment amount based on the fluid parameters output by the multiphysics coupling model, the acoustic dynamic correction factor, and the current state of the valve core, and executes these segments sequentially. This achieves segmented adjustment of the valve opening, keeping fluid disturbances and local pressure fluctuations within a controllable range.

[0085] By combining slow opening and closing with segmented opening control, this invention can smooth the movement of the valve core and optimize it in stages when abnormal disturbances occur in the valve's operating conditions, avoiding overshoot, oscillation, or local cavitation, and improving the control stability and reliability of the regulating valve under complex flow field conditions.

[0086] In one specific implementation, the valve core movement adjustment mode includes: If the turbulence intensity abnormally exceeds the threshold, the slow-opening and slow-closing mode should be selected to reduce local velocity fluctuations. If the cavitation intensity or shock wave parameters abnormally exceed the threshold, select the segmented opening mode to release the pressure difference in stages and reduce the cavitation impact.

[0087] In one specific implementation, the steps for calculating the valve core opening adjustment based on the current valve core position, the fluid parameters output by the multiphysics coupling model, and the acoustic dynamic correction factor are as follows: First, the fluid parameters output by the multiphysics coupling model are compared with the target operating condition indicators to calculate various deviations, such as pressure deviation and flow deviation, as input references for opening adjustment. Then, the deviations are input into a control algorithm, such as proportional-integral-derivative (PID) control, fuzzy control, or model predictive control (MPC), to calculate the preliminary valve core opening adjustment based on the current valve core position and motion state, ensuring that the adjustment enables the valve to respond quickly and gradually approach the target operating condition. Next, the dynamic correction factor obtained by inverting the acoustic features collected by the acoustic sensor is applied to the preliminary valve core opening adjustment, and the preliminary adjustment is corrected by weighting to form the final valve core opening adjustment, in order to compensate for the deviation between the multiphysics coupling model and the actual operating condition, ensuring that the valve core motion trajectory can adapt to flow field disturbances in real time.

[0088] Step S5 above collects fluid condition data and acoustic signals in real time during valve core adjustment, and dynamically calculates the valve core opening adjustment based on the fluid parameters output by the multi-physics coupling model and the acoustic dynamic correction factor. At the same time, it combines the slow opening and slow closing and segmented opening control strategies to smooth and optimize the valve core movement in stages. The system can respond to flow field disturbances in real time when the acoustic characteristics exceed the preset threshold, reduce transient pressure fluctuations, turbulence intensity and cavitation impact, realize intelligent adaptive adjustment of the valve under complex working conditions, improve the accuracy, real-time performance and stability of valve core opening control, and effectively avoid valve overshoot, oscillation or local cavitation.

[0089] In summary, this invention introduces acoustic signals collected by acoustic sensors to dynamically correct a multiphysics coupling model. It calculates a dynamic correction factor by comparing acoustic features with model predictions and applies it to local pressure gradients, turbulent kinetic energy, and fluid viscosity coefficients, enabling the model to reflect actual fluid conditions in real time. Compared to modeling methods based solely on fluid parameters, this method significantly reduces prediction errors, improves the accuracy and reliability of valve core opening control commands, thereby reducing the probability of valve core instability and enhancing system safety and stability. In this invention, when acoustic features in the acoustic signal exceed a preset threshold, a slow-opening / slow-closing or segmented opening control strategy is triggered. Turbulence intensity exceeding a threshold triggers a slow-opening / slow-closing mode, while cavitation intensity or shock wave parameters exceeding a threshold trigger a segmented opening mode. This allows for adaptive selection of the control method based on the disturbance type, thereby reducing fluid impact and energy loss caused by rapid valve core movement and ensuring the safety and stability of the valve system. This invention employs a closed-loop feedback mechanism and incorporates a dynamic correction factor during control command generation to weightedly adjust the valve core opening. This enables the system to iteratively adjust based on real-time operating conditions, thereby dynamically optimizing the valve core's trajectory. This approach ensures precise valve opening adjustment and enables intelligent, adaptive control of the regulating valve under complex operating conditions.

[0090] Example 2 is a second embodiment of the present invention. This embodiment provides an intelligent control system for a regulating valve, applied to the above-mentioned intelligent control method for the regulating valve, including: The operating condition acquisition module is used to acquire real-time operating condition data of the control valve, including fluid parameters and valve execution status; The multiphysics coupling modeling module is used to build a multiphysics coupling model including fluid dynamics, heat conduction and materials mechanics based on the operating condition data, and output the predicted parameters. The acoustic sensing and dynamic correction module is used to acquire acoustic signals, extract features from the acoustic signals, and calculate dynamic correction factors based on the comparison between acoustic features and model prediction results to correct the multiphysics coupling model. The control command generation module is used to generate control commands for valve core opening based on the deviation between the output of the corrected multiphysics coupling model and the target operating condition. The actuator drive module is used to convert control commands into electrical signals that the actuator can recognize, and drive the actuator to adjust the valve core position; The closed-loop feedback module is used to continuously collect operating condition data and acoustic signals during valve core adjustment and input them into the multi-physics coupling model to update the dynamic correction factor, thereby realizing closed-loop feedback between operating condition, model and acoustic signal. The adaptive control module is used to trigger a slow-opening / slow-closing or segmented opening control strategy when the acoustic features in the acoustic signal exceed a preset threshold.

[0091] 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, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for intelligent control of a regulating valve, characterized in that, include: Acquire the current operating status data of the control valve, including fluid parameters and valve execution status; Based on the aforementioned operating condition data, a multiphysics coupling model is established; The multiphysics coupling model is dynamically corrected using acoustic signals collected by acoustic sensors; control commands for valve core opening are generated based on the corrected multiphysics coupling model, and the actuator is driven to adjust the valve core position. During the valve core adjustment process, operating condition data and acoustic signals are collected in real time to form a closed-loop feedback based on operating condition-model-acoustic signal; When the acoustic features in the acoustic signal exceed a preset threshold, a slow-opening / slow-closing or segmented opening control strategy is triggered.

2. The intelligent control method for the regulating valve as described in claim 1, characterized in that: Establishing the multiphysics coupling model includes the following steps: Based on the structure and operating condition data of the control valve, the type of physical field of the model is determined and the boundary conditions are defined. The type of physical field of the model includes fluid dynamics, heat conduction and valve material deformation. Construct fluid dynamics models, heat conduction models, and materials mechanics models respectively; The fluid dynamics model, heat conduction model, and materials mechanics model are coupled to form a multiphysics coupling equation system to simulate the effect of temperature on fluid viscosity, the effect of fluid pressure on the valve body, and the feedback effect of material deformation on the flow field. Among them, the fluid dynamics model outputs local flow velocity, pressure and turbulence characteristics; the heat conduction model outputs temperature distribution and heat flux density; and the materials mechanics model outputs structural deformation results.

3. The intelligent control method for the regulating valve as described in claim 2, characterized in that: Multiphysics coupling models are coupled in the following ways: The local flow velocity, pressure, and turbulence characteristics output by the fluid dynamics model are mapped to the boundary conditions of the heat conduction model to calculate the temperature distribution and heat flux density of the valve. The temperature distribution and heat flux density output from the heat conduction model are mapped together with the pressure load output from the fluid dynamics model and used as input loads to calculate the stress, strain and deformation distribution of the valve structure. The structural deformation results output from the mechanics of materials model are fed back to the fluid dynamics model to dynamically correct the flow field boundary conditions.

4. The intelligent control method for the regulating valve as described in claim 2, characterized in that: The multiphysics coupling model is dynamically corrected using acoustic signals collected by acoustic sensors, including the following steps: Acoustic signals from inside the valve and downstream pipeline are collected in real time using an array of acoustic sensors arranged downstream of the valve. The collected acoustic signals are feature extracted, and the acoustic features are obtained by inversion based on the correspondence between acoustic features and fluid physical quantities; The acoustic features are compared with the prediction output of the multiphysics coupling model at the same time or within the same time window. The observation residuals are calculated and mapped to the dynamic correction factor of the multiphysics coupling model. The dynamic correction factor is applied to the local pressure gradient, turbulent kinetic energy, and fluid viscosity coefficient of the multiphysics coupling model to achieve adaptive correction of the model. The fluid, heat conduction, and material mechanical properties are recalculated using a modified multiphysics coupling model to obtain updated valve condition predictions.

5. The intelligent control method for the regulating valve as described in claim 4, characterized in that: The formula for calculating the dynamic correction factor is as follows: in: It is a dynamic correction factor used to correct local pressure gradients, turbulent kinetic energy, and viscosity coefficients; These are the spatial coordinates of the flow field, i.e., the positions inside the valve and in the downstream pipeline; For time; The instantaneous acoustic characteristic energy obtained from acoustic sensor observations; The corresponding physical quantity energy predicted by the multiphysics coupling model; The difference between acoustic observations and model predictions represents the observation residual. The relative residual is used to normalize acoustic bias. The hyperbolic tangent function is used to restrict the correction factor to a certain value. ; To correct the amplitude coefficient, control the observation residuals through The magnitude of the impact of mapping on the final correction factor; This is a nonlinear sensitivity coefficient that controls the intensity of the corrected response. It is a small constant.

6. The intelligent control method for the regulating valve as described in claim 4, characterized in that: The control command for valve core opening is generated based on the modified multiphysics coupling model, specifically including the following steps: The fluid parameters output by the multiphysics coupling model are compared with the target operating condition indicators, and the deviation is calculated. Based on the deviation, a preliminary valve core opening adjustment amount is generated using a control algorithm; The dynamic correction factor is applied to the initial valve core opening adjustment amount, and the initial valve core opening adjustment amount is weighted and corrected to obtain the final valve core opening adjustment amount. The final valve core opening adjustment is converted into a control command that the actuator can recognize.

7. The intelligent control method for the regulating valve as described in claim 1, characterized in that: Triggering the slow-opening / slow-closing or segmented opening control strategy includes the following steps: The acoustic features in the real-time acquired acoustic signals are compared with preset thresholds. When any acoustic feature exceeds its corresponding threshold, a slow opening / closing or segmented opening control strategy is triggered. Based on the type of acoustic feature that exceeds the threshold, select the corresponding valve core motion adjustment mode. When the turbulence intensity exceeds the threshold, the slow opening and slow closing mode is triggered. When the cavitation intensity or shock wave parameter exceeds the threshold, the segmented opening mode is triggered. Based on the current valve core position, the fluid parameters output by the multiphysics coupling model, and the acoustic dynamic correction factor, the valve core opening adjustment amount is dynamically calculated. The calculated valve core opening adjustment is converted into an electrical signal that the actuator can recognize, driving the valve core to move; During the movement of the valve core, data is continuously collected and the dynamic correction factor is updated and the control command is iteratively corrected until the acoustic characteristics are below the threshold and the target working condition is reached.

8. The intelligent control method for the regulating valve as described in claim 7, characterized in that: The slow-opening and slow-closing mode adjusts the valve core opening gradually and smoothly, allowing the valve core to move at a small amplitude and low speed. The segmented opening mode divides the valve core opening adjustment process into multiple continuous intervals, and executes different control strategies in each interval according to the current operating conditions.

9. The intelligent control method for the regulating valve as described in claim 1, characterized in that: The fluid parameters include, but are not limited to, medium temperature, medium pressure, flow rate, density, and viscosity; the valve execution status includes, but is not limited to, valve core position and actuator drive signal.

10. An intelligent control system for a regulating valve, applied to the intelligent control method for the regulating valve as described in any one of claims 1-9, characterized in that, include: The operating condition acquisition module is used to acquire real-time operating condition data of the control valve, including fluid parameters and valve execution status; The multiphysics coupling modeling module is used to build a multiphysics coupling model including fluid dynamics, heat conduction and materials mechanics based on the operating condition data, and output the predicted parameters. The acoustic sensing and dynamic correction module is used to acquire acoustic signals, extract features from the acoustic signals, and calculate dynamic correction factors based on the comparison between acoustic features and model prediction results to correct the multiphysics coupling model. The control command generation module is used to generate control commands for valve core opening based on the deviation between the output of the corrected multiphysics coupling model and the target operating condition. The actuator drive module is used to convert control commands into electrical signals that the actuator can recognize, and drive the actuator to adjust the valve core position; The closed-loop feedback module is used to continuously collect operating condition data and acoustic signals during valve core adjustment and input them into the multi-physics coupling model to update the dynamic correction factor, thereby realizing closed-loop feedback between operating condition, model and acoustic signal. The adaptive control module is used to trigger a slow-opening / slow-closing or segmented opening control strategy when the acoustic features in the acoustic signal exceed a preset threshold.

Images