Intelligent regulation and control method of pressure reducing valve and regulation and control system thereof

Abstract

The invention discloses an intelligent regulation and control method for a pressure reducing valve, which comprises the following steps: acquiring sensor data in the operation process of the pressure reducing valve, constructing a digital twin basic physical model of the pressure reducing valve, and calculating by using the digital twin basic physical model to obtain a basic data calculation result of the current state of the pressure reducing valve; acquiring real-time acquisition data in a dynamic operation process based on the basic data calculation result, and preprocessing the real-time acquisition data to obtain processed standardized operation feature data; optimizing the digital twin basic physical model by using the standardized operation characteristic data, and constructing a predictive regulation and control model fusing physical modeling and data driving; and performing operation state prediction based on the prediction regulation and control model to obtain a voltage regulation prediction result in a future time period. Intelligent prediction of the voltage regulation state in the future is achieved, then the regulation instruction is given in advance, the control lag or over-regulation phenomenon is effectively restrained, and the response speed and stability of the whole system are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of pressure reducing valve control, and in particular to an intelligent control method and a control system for a pressure reducing valve. Background Art

[0002] In recent years, pressure reducing valves (PRVs), as key pressure regulating devices, have been widely used to maintain stable system operation and ensure the safety of downstream equipment in applications such as industrial automation, urban water supply, gas regulation, and energy management. Traditional PRV control methods often rely on mechanical feedback or simple PID control strategies, making them difficult to handle rapid changes and complex interference in system operation. With the advancement of sensing technology, modeling technology, and intelligent control, combining multi-source sensor data with virtual modeling to achieve digital mapping and control optimization of PRV operating states has become a key development direction in the field of pressure control.

[0003] However, existing pressure reducing valve control processes typically rely on preset physical models or empirical algorithms as the basis for pressure regulation. These methods fail to dynamically modify the model based on real-time operating data and lack a feedback mechanism to continuously correct modeling errors. This reliance on static models can easily lead to deviations between model output and actual operating conditions. This is especially true in scenarios with sudden load changes or frequent operating disturbances, where model predictions become disconnected from physical feedback. This in turn affects the accuracy and timeliness of pressure regulation commands, potentially leading to system overshoot or regulation lag. Summary of the Invention

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

[0005] In view of the problems existing in the above-mentioned existing intelligent control method of a pressure reducing valve and its control system, the present invention is proposed.

[0006] Therefore, the purpose of the present invention is to provide an intelligent control method and control system for a pressure reducing valve, which is suitable for solving the problem that the predicted value of the existing technology model is out of line with the physical feedback, thereby affecting the accuracy and timeliness of the pressure regulation instruction, and easily causing system overregulation or regulation lag.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions: In a first aspect, an embodiment of the present invention provides an intelligent control method for a pressure reducing valve, comprising: Acquire sensor data during the operation of the pressure reducing valve, build a digital twin basic physical model of the pressure reducing valve, and use the digital twin basic physical model to perform calculations to obtain basic data calculation results of the current state of the pressure reducing valve; Based on the calculation results of the basic data, real-time collected data in the dynamic operation process is obtained and pre-processed to obtain processed standardized operation characteristic data; Utilizing the standardized operational characteristic data to optimize the digital twin basic physical model, and constructing a prediction and control model that integrates physical modeling and data-driven modeling; Based on the predictive control model, the operating state is predicted to obtain the pressure regulation prediction result in the future period, and the predictive pressure regulation and virtual-real comparison abnormal diagnosis of the pressure reducing valve are realized.

[0008] As a preferred solution of the intelligent control method of a pressure reducing valve described in the present invention, the obtaining of sensor data during the operation of the pressure reducing valve refers to collecting the pressure reducing valve outlet pressure through a pressure sensor, collecting the instantaneous flow of the pipeline through a flow sensor, collecting the medium temperature through a temperature sensor, collecting the real-time opening position information of the valve core through a valve position sensor, and collecting the micro-vibration and posture change data of the pressure reducing valve during operation through an inertial measurement unit.

[0009] As a preferred solution of the intelligent control method of a pressure reducing valve described in the present invention, the construction of a digital twin basic physical model of the pressure reducing valve includes the following steps: By introducing the fluid continuity equation and the relationship between the flow resistance calculation, the outlet pressure estimate is obtained, thereby simulating the pressure output state of the pressure reducing valve under different working conditions. The specific formula is as follows: ; in, is the estimated outlet pressure, is the pressure reducing valve inlet pressure, is the dynamic viscosity of the working medium, is the flow velocity in the valve body, is the instantaneous flow cross-sectional area of ​​the valve port, is the adjustment coefficient; by introducing the actuator inertia modeling structure, the control response delay value is obtained to compensate for the time lag between the virtual control signal and the actual action. The specific formula is as follows: ; in, It is the control response delay value, which indicates the deviation between the control signal and the actual action. is the target valve core displacement value issued by the control system, is the actual valve core position, is the system inertia time constant, is the base of the natural logarithm, a digital constant. By introducing the virtual and real pressure residual feedback mechanism, the model residual correction value is obtained, thereby realizing the adaptive correction and error learning of the model. The specific formula is as follows: ; in, is the model residual correction value, is the actual outlet pressure value obtained by the sensor; According to the estimated value of outlet pressure , control response delay value and model residual correction values , build a digital twin basic physical model, and comprehensively evaluate the impact of control instructions, mechanical inertia and system errors on the decompression output state.

[0010] As a preferred solution of the intelligent control method of a pressure reducing valve described in the present invention, the specific formula of the digital twin basic physical model is as follows: ; in, Output pressure value for the final digital twin basic physical model, To take into account the delayed prediction value, is the residual correction coefficient.

[0011] As a preferred solution of the intelligent control method of a pressure reducing valve of the present invention, the real-time collected data includes pre-valve pressure data, post-valve pressure data, valve core displacement data, flow data and valve body vibration intensity data; Acquiring real-time collected data during dynamic operation based on the basic data calculation results includes the following steps: A joint judgment is performed based on the after-valve pressure change and the valve core displacement change calculated in the basic data calculation result. If the after-valve pressure change value per unit time is less than a first pressure change threshold, and the valve core displacement change value is less than the first displacement change threshold, it is determined that the current state is a low-fluctuation state, and the above-mentioned real-time data is collected using the first sampling frequency f1; If the post-valve pressure change value within a unit time is between the first pressure change threshold and the second pressure change threshold, or the valve core displacement change value is between the first displacement change threshold and the second displacement change threshold, then the current state is determined to be a medium fluctuation state, and the above-mentioned real-time data is collected using the second sampling frequency f2; If the change value of the valve rear pressure in unit time is greater than or equal to the second pressure change threshold, or the change value of the valve core displacement is greater than or equal to the second displacement change threshold, the current state is determined to be a high fluctuation state, and the third sampling frequency f3 is used to collect the above real-time data, where f3>f2>f1.

[0012] As a preferred solution of the intelligent control method of a pressure reducing valve described in the present invention, the construction of a prediction and control model integrating physical modeling and data-driven includes the following steps: By introducing a delay correction function, the pressure prediction error caused by the control delay is adjusted on the time axis to obtain the response compensation pressure value, thereby improving the adaptability of the prediction model to the inertia characteristics of the control system. The specific formula is as follows: ; in, To respond to the compensation pressure value; by introducing the state offset prediction function to model the coupling effect of flow and temperature fluctuations, the environmental disturbance prediction term is obtained, thereby improving the robustness of the model under dynamic conditions. The specific formula is as follows: ; in, is the environmental disturbance prediction item, is the flow rate change rate per unit time, is the temperature change rate per unit time, and is the empirical fitting coefficient; By introducing the residual correction term and combining the above-mentioned response compensation pressure value and environmental disturbance prediction term to build a prediction and control model, the final predicted pressure regulation target pressure value is obtained, thereby realizing the prediction and adjustment of the future operating state. The specific formula is as follows: ; in, is the final predicted pressure regulation target pressure value, is the residual weight that is adaptively adjusted according to the historical residual trend, and the unit is dimensionless. is the residual correction term.

[0013] As a preferred embodiment of the intelligent control method for a pressure reducing valve according to the present invention, the method includes the following steps: performing an operating state prediction based on the predictive control model, obtaining a pressure regulation prediction result in a future period, and implementing predictive pressure regulation and virtual-real comparison abnormality diagnosis of the pressure reducing valve. Calculate the difference between the actual and virtual pressures based on the target pressure value output by the predictive control model and the actual pressure after the valve. , and combined with the current fluctuation state, select and execute the following judgment: If the initial current state is a low fluctuation state, and the virtual and real pressure difference When the error is less than the first error threshold, the system is judged to be operating normally and the voltage regulation instruction is executed according to the predictive control model; If the current state is initially determined to be medium or high fluctuation, or the difference between the real and virtual pressures If the error is greater than or equal to the first error threshold, it is preliminarily determined that the system has a control anomaly and the next step is executed; If the difference between the real and virtual pressures If the change in valve core displacement within the current sampling period is less than the second error threshold, and the valve core displacement change within the current sampling period is less than the valve core displacement threshold, and the valve body vibration intensity within the current period is less than the valve body vibration threshold, then the current state is determined to be a mild abnormality, the current prediction error and dynamic response data are recorded, and the output gain coefficient of the prediction model is adjusted; If the difference between the real and virtual pressures If the error is greater than or equal to the second error threshold, or if either the change in valve core displacement within the current sampling period or the vibration intensity of the valve body within the current period is greater than the corresponding threshold, the current state is determined to be a serious abnormal state, triggering the pressure regulation protection mechanism and terminating the current control execution process.

[0014] Secondly, to further address the problem that the model prediction value in the prior art is out of sync with the physical feedback, which in turn affects the accuracy and timeliness of the pressure regulation instructions and easily causes system overregulation or regulation lag, an embodiment of the present invention provides an intelligent control system for a pressure reducing valve, including: State data acquisition module: used to obtain sensor data during the operation of the pressure reducing valve, and build a digital twin basic physical model of the pressure reducing valve. The digital twin basic physical model is used to perform calculations to obtain the basic data calculation results of the current state of the pressure reducing valve; Data processing module: used to obtain real-time collected data during dynamic operation based on the basic data calculation results and perform preprocessing to obtain processed standardized operation characteristic data; Model optimization module: used to optimize the digital twin basic physical model using the standardized operating characteristic data, and build a prediction and control model that integrates physical modeling and data-driven; Control execution module: used to predict the operating status based on the predictive control model, obtain the pressure regulation prediction results in the future period, and realize the predictive pressure regulation and virtual-real comparison abnormal diagnosis of the pressure reducing valve.

[0015] In a third aspect, an embodiment of the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the intelligent control method of a pressure reducing valve as described in the first aspect of the present invention is implemented.

[0016] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, any step of the intelligent control method of a pressure reducing valve as described in the first aspect of the present invention is implemented.

[0017] The beneficial effects of the present invention are as follows: By constructing a digital twin basic physical model that includes modules such as pressure output, control delay, and residual self-correction, it is possible to more realistically reflect the actual response behavior of the pressure reducing valve under different working conditions. In particular, in environments where the load frequently changes or disturbances occur suddenly, the model can dynamically correct deviations to ensure that the adjustment instructions are always closely consistent with the physical state. By introducing a delay correction function, a disturbance prediction term, and a residual correction mechanism, a predictive control model that integrates physical modeling and data-driven is constructed, enabling intelligent prediction of future pressure regulation states, and thus issuing adjustment instructions in advance, effectively suppressing control lag or overshoot, and improving the response speed and stability of the overall system. Through multi-dimensional analysis of virtual and real pressure comparisons and dynamic operating characteristics, abnormal states can be divided into mild and severe abnormalities, and response mechanisms such as model gain adjustment or protective shutdown can be adopted accordingly to improve the system's ability to cope with potential out-of-control risks. The data collection frequency can be adaptively adjusted according to the pressure fluctuation state, while ensuring data adequacy. The system load, energy consumption and computing pressure are reduced, and the system's deployment capabilities in edge devices or low-power scenarios are improved. It can be widely used in key equipment involving pressure regulation in urban water supply systems, petrochemical processes, thermal pipelines, and gas transportation systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. Those skilled in the art can also derive other drawings based on these drawings without inventive effort. Among them: Figure 1 This is a schematic diagram of the overall process of an intelligent control method for a pressure reducing valve proposed by the present invention; Figure 2 This is a schematic diagram of the abnormality judgment process of the intelligent control method of a pressure reducing valve proposed by the present invention. DETAILED DESCRIPTION

[0019] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0020] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0021] Secondly, the term "one embodiment" or "embodiment" 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 various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0022] Furthermore, the present invention is described in detail with reference to schematic diagrams. For ease of illustration, when describing the embodiments of the present invention, cross-sectional views illustrating device structures may be partially enlarged and not to scale. Furthermore, the schematic diagrams are merely illustrative and should not limit the scope of protection of the present invention. Furthermore, in actual production, the three-dimensional dimensions of length, width, and depth should be included.

[0023] Example 1 Reference Figure 1-Figure 2 , as an embodiment of the present invention, provides an intelligent control method for a pressure reducing valve.

[0024] In existing pressure reducing valve control processes, preset physical models or empirical algorithms are often used as the basis for pressure regulation. These methods fail to dynamically modify the model based on real-time operating data and lack a feedback mechanism to continuously correct modeling errors. This reliance on static models can easily lead to deviations between model output and actual operating conditions. This is especially true in scenarios with sudden load changes or frequent operating disturbances, where model predictions become disconnected from physical feedback. This in turn affects the accuracy and timeliness of pressure regulation commands, and can easily cause problems such as system overshoot or regulation lag.

[0025] The present application provides an effective solution to the above-mentioned problems. Next, a plurality of embodiments will be combined to explain in detail how to implement the intelligent control method of the pressure reducing valve.

[0026] Figure 1 The overall flow chart of an intelligent control method for a pressure reducing valve is shown, including: S1: Obtain sensor data during the operation of the pressure reducing valve, build a digital twin basic physical model of the pressure reducing valve, and use the digital twin basic physical model to perform calculations to obtain the basic data calculation results of the current state of the pressure reducing valve.

[0027] Preferably, obtaining sensor data during the operation of the pressure reducing valve refers to collecting the pressure reducing valve outlet pressure through a pressure sensor, collecting the instantaneous flow of the pipeline through a flow sensor, collecting the medium temperature through a temperature sensor, collecting the real-time opening position information of the valve core through a valve position sensor, and collecting the micro-vibration and posture change data of the pressure reducing valve during operation through an inertial measurement unit.

[0028] Furthermore, building a digital twin basic physical model of the pressure reducing valve includes the following steps: By introducing the fluid continuity equation and the relationship between the flow resistance calculation, the outlet pressure estimate is obtained, thereby simulating the pressure output state of the pressure reducing valve under different working conditions. The specific formula is as follows: ; in, is the estimated outlet pressure, is the pressure reducing valve inlet pressure, is the dynamic viscosity of the working medium, is the flow velocity in the valve body, is the instantaneous flow cross-sectional area of ​​the valve port, is the adjustment coefficient; By introducing the actuator inertia modeling structure, the control response delay value is obtained to compensate for the time lag between the virtual control signal and the actual action. The specific formula is as follows: ; in, It is the control response delay value, which indicates the deviation between the control signal and the actual action. is the target valve core displacement value issued by the control system, is the actual valve core position, is the system inertia time constant, is the base of natural logarithms, a numerical constant; By introducing the virtual and real pressure residual feedback mechanism, the model residual correction value is obtained, thereby realizing the adaptive correction and error learning of the model. The specific formula is as follows: ; in, is the model residual correction value, is the actual outlet pressure value obtained by the sensor; According to the estimated value of outlet pressure , control response delay value and model residual correction values , build a digital twin basic physical model, and comprehensively evaluate the impact of control instructions, mechanical inertia, and system errors on the decompression output state. The specific formula of the digital twin basic physical model is as follows: ; in, Output pressure value for the final digital twin basic physical model, To take into account the delayed prediction value, is the residual correction coefficient.

[0029] In this embodiment, the dynamic viscosity of the working medium It can be measured based on the physical parameter table of the actual fluid used or on-site experiments, especially in high temperature or high pressure environments, using an online viscosity sensor for dynamic measurement to reflect the real-time changes in the physical properties of the medium during operation; adjustment coefficient By adjusting the proportional relationship between the pressure change before and after the valve core displacement, empirical fitting can be performed, and the optimization setting can be made according to the historical response curves of different types of pressure reducing valves under typical steady-state and fluctuating conditions; the system inertia time constant Determined by the response delay of the actuator under standard step input, it can be extracted through excitation response experiment combined with second-order inertial system fitting method; residual correction coefficient Learning is performed based on the residual sequence between the model-predicted pressure and the actual pressure, and dynamic adjustment is made using the exponentially weighted moving average method (EWMA) to continuously improve the model's prediction accuracy and adaptability. The above parameters can be set to initial values ​​based on on-site debugging experience and dynamically modified during subsequent operation. These parameters are not specifically limited in this embodiment.

[0030] Preferably, in the process of constructing the basic physical model of the digital twin, the combined influence of three key physical quantities, namely the outlet pressure estimation value, the control response delay value, and the model residual correction value, on the operating state of the pressure reducing valve is comprehensively considered to enhance the model's dynamic adaptability to complex working conditions. Among them, the outlet pressure estimation value introduces the fluid continuity equation and the flow resistance characteristic function, and adopts a piecewise nonlinear function to adapt to various pressure boundaries. The control response delay value is expressed by the time delay convolution function in combination with the system inertia characteristics to accurately reflect the lag effect between the control signal and the mechanical execution. The model residual correction value introduces an exponential decay weight function to continuously absorb historical errors and iteratively correct the model parameters. Through the combined effect of the above three indicators, the constructed model can not only accurately map the dynamic characteristics of the pressure reducing process, but also has good generalization ability and error self-compensation ability, providing stable physical support for the subsequent construction of a fusion prediction and control model.

[0031] S2: Based on the basic data calculation results, the real-time collected data during the dynamic operation process is obtained and preprocessed to obtain the processed standardized operation characteristic data.

[0032] Preferably, the real-time collected data includes valve front pressure data, valve rear pressure data, valve core displacement data, flow data, and valve body vibration intensity data. The real-time collected data is obtained as follows: The pressure data before and after the valve are collected in real time through high-precision pressure sensors installed on both sides of the pipeline. The pressure sensor is a strain gauge sensor with water hammer impact resistance. The valve core displacement data is measured by an inductive displacement sensor installed inside the valve body, which can continuously sense the displacement change trajectory of the valve core during the adjustment process. The flow data is collected by an electromagnetic flowmeter arranged in the middle section of the pipeline. The electromagnetic flowmeter can adapt to the conductive characteristics of different media and has anti-interference ability. The valve body vibration intensity data is obtained based on the inertial measurement unit (IMU). The IMU module integrates accelerometer and gyroscope components to record the tiny structural vibration and posture change information of the pressure reducing valve during operation in real time.

[0033] Acquiring real-time collected data during dynamic operation based on basic data calculation results includes the following steps: A joint judgment is performed based on the post-valve pressure change and the valve core displacement change calculated in the basic data calculation result. If the post-valve pressure change value per unit time is less than the first pressure change threshold, and the valve core displacement change value is less than the first displacement change threshold, then the current state is determined to be a low-fluctuation state, and the above-mentioned real-time data is collected using the first sampling frequency f1; If the post-valve pressure change value within a unit time is between the first pressure change threshold and the second pressure change threshold, or the valve core displacement change value is between the first displacement change threshold and the second displacement change threshold, then the current state is determined to be a medium fluctuation state, and the above-mentioned real-time data is collected using the second sampling frequency f2; If the change value of the valve rear pressure in unit time is greater than or equal to the second pressure change threshold, or the change value of the valve core displacement is greater than or equal to the second displacement change threshold, the current state is determined to be a high fluctuation state, and the third sampling frequency f3 is used to collect the above real-time data, where f3>f2>f1.

[0034] It should be noted that the setting of the first pressure change threshold, the second pressure change threshold, the first displacement change threshold, and the second displacement change threshold was first performed based on the collection and statistical analysis of historical measured data of the pressure reducing valve under typical operating conditions. The common ranges of the valve downstream pressure change and valve core displacement change per unit time and their fluctuation amplitude distribution were calculated. The reasonable fluctuation cutoff point was determined by combining the pressure reducing valve's adjustment response speed and error tolerance under different fluctuation levels, and the corresponding pressure and displacement thresholds were set to clarify the judgment criteria for low, medium, and high fluctuations. Furthermore, to improve the system's response efficiency and data processing capabilities, a comparative analysis of data integrity and model prediction accuracy at different sampling frequencies was conducted to select the optimal sampling configuration combination.

[0035] Specifically, the preprocessing process involves time synchronization and unit normalization of the collected pre-valve pressure, post-valve pressure, valve core displacement, flow rate, and vibration intensity data. A sliding average filter algorithm is then used to smooth short-term, high-frequency fluctuations in the sampled data. This algorithm, a common time-domain filtering method, effectively suppresses random noise in sensor signals. For vibration intensity data, the present invention introduces a multi-period weighted smoothing strategy as an application of the existing exponentially weighted average method. By assigning different weights to data in multiple consecutive time windows, the system enhances the recognition of subtle trends, thereby improving the ability to capture precursors to operational anomalies and the robustness of the model input data.

[0036] Preferably, in order to take into account both the efficiency of data resource utilization and the quality of model input, the present invention proposes a three-level sampling strategy based on the fluctuation state, adopting a low sampling frequency in a low fluctuation state to reduce the system load, enabling a medium sampling frequency in a medium fluctuation state to balance real-time performance and accuracy, and switching to high-frequency sampling in a high fluctuation state to capture sudden dynamic characteristics, thereby achieving an optimal balance between data acquisition efficiency and system prediction sensitivity.

[0037] S3: Use standardized operating characteristic data to optimize the basic physical model of the digital twin and build a predictive control model that integrates physical modeling and data-driven.

[0038] Building a predictive control model that integrates physical modeling and data-driven modeling includes the following steps: By introducing a delay correction function, the pressure prediction error caused by the control delay is adjusted on the time axis to obtain the response compensation pressure value, thereby improving the adaptability of the prediction model to the inertia characteristics of the control system. The specific formula is as follows: ; in, To respond to compensated pressure value; By introducing a state offset prediction function to model the coupled effects of flow and temperature fluctuations, the environmental disturbance prediction term is obtained, thereby improving the robustness of the model under dynamic conditions. The specific formula is as follows: ; in, is the environmental disturbance prediction item, is the flow rate change rate per unit time, is the temperature change rate per unit time, and is the empirical fitting coefficient; By introducing the residual correction term and combining the above-mentioned response compensation pressure value and environmental disturbance prediction term to build a prediction and control model, the final predicted pressure regulation target pressure value is obtained, thereby realizing the prediction and adjustment of the future operating state. The specific formula is as follows: ; in, is the final predicted pressure regulation target pressure value, is the residual weight that is adaptively adjusted according to the historical residual trend, and the unit is dimensionless. is the residual correction term.

[0039] In the embodiment of the present application, the residual weight The fluctuation trend of the virtual and real pressure residuals in the historical operating data of the pressure reducing valve can be statistically analyzed, and the persistence and change rate of the residuals in different time periods can be combined to make adaptive settings, giving priority to the similarity between the current working conditions and the historical operating conditions to enhance the adaptability of the residual correction to different fluctuation scenarios; the empirical fitting coefficient and Used to quantize the residual correction term The correction amplitude for the overall predicted pressure value is derived from regression fitting of extensive experimental data, ensuring the model's high fitting accuracy across a wide range of flow-pressure coupling conditions. After initial setting, these two parameters can also be iteratively optimized based on field operational feedback to further enhance prediction stability and response sensitivity.

[0040] Preferably, in the present invention, by constructing a prediction and control model that integrates physical modeling and data-driven, the coupling effects of multiple factors such as control delay, operating disturbance and model error on the pressure regulation effect are fully considered, thereby improving the accuracy of pressure regulation prediction and the response sensitivity of the system. The response compensation pressure value is constructed using a delay function to enhance the adaptability of the model to control lag; the environmental disturbance term uses a fractional structure to express the influence of the interaction between flow and temperature on pressure fluctuations, thereby enhancing the robustness of the model to complex operating environments; the residual correction term integrates the historical error trend and the current prediction deviation to construct a dynamic correction mechanism to avoid prediction distortion caused by error accumulation. The final prediction and control model integrates the above-mentioned adjustment factors with a nonlinear superposition structure, dynamically adjusts the predicted target pressure value, and makes it more in line with actual operating changes. It is superior to traditional static modeling strategies and significantly improves the system's prediction and regulation effects and abnormal precursor recognition capabilities.

[0041] S4: Based on the predictive control model, the operating status is predicted to obtain the pressure regulation prediction results in the future period, and the predictive pressure regulation and virtual-real comparison abnormal diagnosis of the pressure reducing valve are realized.

[0042] Figure 2 This is a schematic diagram of an abnormality judgment process of an intelligent control method for a pressure reducing valve proposed by the present invention; Based on the predictive control model, the operating status is predicted to obtain the pressure regulation prediction results in the future period, and the predictive pressure regulation and virtual-real comparison abnormal diagnosis of the pressure reducing valve are realized, including the following steps: Calculate the difference between the actual and virtual pressures based on the target pressure value output by the predictive control model and the actual pressure after the valve. , and combined with the current fluctuation state, select and execute the following judgment: If the initial current state is a low fluctuation state, and the virtual and real pressure difference When the error is less than the first error threshold, the system is judged to be operating normally and the voltage regulation instruction is executed according to the predictive control model; It should be noted that executing a pressure regulation command refers to the control system using the target pressure value output by the predictive control model, combined with the current valve core displacement state and pressure control requirements, to precisely control the valve core movement amplitude and rate through the drive device to complete the corresponding opening adjustment operation. In actual operation, the control system monitors the difference between the current valve core displacement value and the model's recommended opening in real time, and uses this difference as a basis to issue control pulses or regulation commands to the electric actuator or pneumatic actuator, ensuring that the valve core dynamically responds and adjusts according to the path planned by the predictive model, thereby achieving precise control of downstream pressure.

[0043] Specifically, the execution process of the pressure regulation instruction includes the following steps: first, the system obtains the pressure regulation target pressure value calculated by the prediction model, and inversely infers the current required valve core opening target value based on the valve flow characteristic curve and the pressure response curve; then, the controller compares the target opening value with the current valve core position collected in real time to calculate the error signal; finally, based on the error signal, the system generates a control signal through a preset proportional-integral-derivative (PID) controller or other adaptive regulator to drive the pressure reducing valve actuator to adjust the valve core position, and finally achieve dynamic matching of the pressure regulation target.

[0044] If the current state is initially determined to be medium or high fluctuation, or the difference between the real and virtual pressures If the error is greater than or equal to the first error threshold, it is preliminarily determined that the system has a control anomaly and the next step is executed; If the difference between the real and virtual pressures If the value of the prediction error and dynamic response data is less than the second error threshold, the change amplitude of the valve core displacement in the current sampling period is less than the valve core displacement threshold, and the valve body vibration intensity in the current period is less than the valve body vibration threshold, it is determined that the current state is a mild abnormality, the current prediction error and dynamic response data are recorded, and the output gain coefficient of the prediction model is adjusted.

[0045] In this embodiment, when the system determines it is in a mildly abnormal state, it automatically records prediction errors and dynamic response data to improve the predictive model's adaptability under actual operating conditions. Specifically, the system records the deviation between the target pressure value output by the predictive control model and the corresponding measured downstream pressure during the current sampling period. It also simultaneously collects key dynamic response parameters such as valve core displacement, instantaneous flow rate, and valve body vibration during the same period, establishing an abnormal data recording unit that contains multi-source operational information.

[0046] Based on the recorded prediction error and dynamic response data, the system will further adjust the output gain coefficient of the prediction model to improve the response sensitivity and stability of subsequent prediction results. This adjustment process refers to previous historical operating data and combines the persistence characteristics of the current abnormal state to dynamically correct the mapping relationship between the model output and the actual voltage regulation behavior, enabling the model to achieve more accurate voltage regulation predictions in the current operating scenario. To avoid fluctuation interference caused by frequent adjustments, the system also sets a minimum adjustment interval and gain adjustment amplitude limit. The adjustment process is only initiated after the abnormal state has persisted for a certain period of time to ensure the robustness of the model update.

[0047] If the difference between the real and virtual pressures If the error is greater than or equal to the second error threshold, or if either the change in valve core displacement within the current sampling period or the vibration intensity of the valve body within the current period is greater than the corresponding threshold, the current state is determined to be a serious abnormal state, triggering the pressure regulation protection mechanism and terminating the current control execution process.

[0048] The voltage regulation protection mechanism includes a multi-level response strategy. First, it immediately suspends the issuance of the current voltage regulation command generated by the predictive control model to prevent potential abnormal operations from causing further disturbances to the system. Second, it sends a fault alarm signal to the upper control system to indicate the current risks of control deviation, abnormal response or hardware failure.

[0049] On this basis, the system terminates the current control execution process by ceasing continuous prediction of the pressure regulation target value, freezing all sensor data collection tasks within the current sampling cycle, recording all process parameter data of the abnormal state, and automatically switching to standby mode or manual intervention mode, awaiting subsequent manual confirmation or a host computer system restart command. In the event of a serious abnormality, the system can also jointly activate redundant pressure regulation modules or bypass safety valves to ensure that the pressure regulation function of the main control circuit is not completely disabled, ensuring the safe operation of the entire system.

[0050] The triggering condition of this protection mechanism is not only based on the difference between the virtual and actual pressures exceeding the second error threshold, but also combines the joint judgment of key indicators such as the valve core displacement change amplitude and the valve body vibration intensity to ensure that the protection strategy is comprehensive and sensitive, and can quickly respond to sudden abnormalities under complex working conditions, significantly improving the stability and fault response capabilities of the pressure reducing system.

[0051] It should be noted that, for the setting of fluctuation thresholds and sampling frequencies for judging the state of the pressure reducing valve, firstly, based on the actual operating data of the pressure reducing valve, by analyzing the statistical distribution characteristics of the pressure change after the valve and the valve core displacement change under different working conditions, combined with the system response speed and accuracy requirements, the first pressure change threshold, the second pressure change threshold and the corresponding first displacement change threshold and the second displacement change threshold are optimized and determined; secondly, for the setting of sampling frequencies f1, f2 and f3, the real-time performance of data acquisition and the system computing load are comprehensively considered, and the minimum necessary sampling frequency under different fluctuation states is calibrated through experiments to ensure data integrity under high fluctuation conditions, while avoiding resource waste under low fluctuation conditions.

[0052] For example, consider a natural gas distribution station where an intelligent pressure reducing valve control system employing the present invention detects periodic fluctuations in downstream pressure during operation. The system first determines the current state of medium fluctuation based on the downstream pressure change per unit time (between a first and second pressure change threshold) and the valve core displacement change (less than the first displacement change threshold), automatically switching the sampling frequency to f2. The digital twin model then predicts an upward trend in the fluctuation amplitude. When the downstream pressure change exceeds the second pressure change threshold, the system immediately upgrades to a high fluctuation state and activates the f3 sampling frequency. At this point, the predictive control model detects a continued increase in the residual difference between the actual pressure and the virtual model, triggering an abnormality diagnosis mechanism and automatically generating maintenance recommendations. Operations and maintenance personnel inspect the valve core and discover minor wear. After timely replacement, the system resumes stable operation, avoiding the risk of pressure loss due to response delays.

[0053] It should be noted that traditional pressure reducing valve control methods usually use a fixed sampling frequency and a single threshold alarm mechanism, which makes it difficult to adapt to dynamic changes in working conditions. This embodiment achieves adaptive adjustment of the data acquisition strategy by establishing a three-level fluctuation state judgment system: energy-saving mode (f1) is adopted under low fluctuation conditions, accuracy and efficiency are balanced under medium fluctuation conditions (f2), and high-precision monitoring is enabled under high fluctuation conditions (f3). This dynamic sampling mechanism, combined with the real-time optimization of the digital twin model, not only ensures data completeness under critical working conditions, but also significantly reduces the average computing load of the system. At the same time, by combining physical modeling with data-driven, the predictive control model can identify potential anomalies earlier (such as the small vibration characteristics generated by the early stage of valve core wear), which greatly improves the reliability of the control method compared to the traditional threshold alarm method.

[0054] In summary, by building a digital twin basic physical model that includes modules such as pressure output, control delay, and residual self-correction, the actual response behavior of the pressure reducing valve under different operating conditions can be more realistically reflected. Especially in environments with frequent load changes or sudden disturbances, the model can dynamically correct deviations to ensure that the adjustment instructions are always closely consistent with the physical state. By introducing a delay correction function, a disturbance prediction term, and a residual correction mechanism, a predictive control model that integrates physical modeling and data-driven is constructed to achieve intelligent prediction of future pressure regulation states, thereby issuing adjustment instructions in advance, effectively suppressing control lag or overshoot, and improving the response speed and stability of the overall system. Through multi-dimensional analysis of virtual and real pressure comparisons and dynamic operating characteristics, abnormal states can be divided into mild and severe abnormalities, and response mechanisms such as model gain adjustment or protective shutdown can be adopted accordingly to improve the system's ability to cope with potential out-of-control risks. The data collection frequency can be adaptively adjusted according to the pressure fluctuation state, while ensuring data adequacy. The system load, energy consumption and computing pressure are reduced, and the system's deployment capabilities in edge devices or low-power scenarios are improved. It can be widely used in key equipment involving pressure regulation in urban water supply systems, petrochemical processes, thermal pipelines, and gas transportation systems.

[0055] Example 2, an embodiment of the present invention, provides an intelligent control system for a pressure reducing valve, comprising: State data acquisition module: used to obtain sensor data during the operation of the pressure reducing valve, build a digital twin basic physical model of the pressure reducing valve, and use the digital twin basic physical model to perform calculations to obtain the basic data calculation results of the current state of the pressure reducing valve; Data processing module: used to obtain real-time collected data during dynamic operation based on basic data calculation results and perform preprocessing to obtain processed standardized operation characteristic data; Model optimization module: used to optimize the basic physical model of the digital twin using standardized operating characteristic data, and build a prediction and control model that integrates physical modeling and data-driven; Control execution module: used to predict the operating status based on the predictive control model, obtain the pressure regulation prediction results in the future period, and realize the predictive pressure regulation and virtual-real comparison abnormal diagnosis of the pressure reducing valve.

[0056] Example 3 is an embodiment of the present invention, which is different from the previous embodiment in that: If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the various embodiments of the method of the present invention. The aforementioned storage medium includes various media that can store program code, such as USB flash drives, mobile hard drives, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical disks.

[0057] The logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing the logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device). For purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0058] More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk cartridge (magnetic devices), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disc read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting, or processing it in another suitable manner as necessary, and then storing it in a computer memory.

[0059] It should be understood that various components of the present invention may be implemented using hardware, software, firmware, or a combination thereof. In the above-described embodiments, multiple steps or methods may be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof may be used: a discrete logic circuit having logic gate circuits for implementing logic functions on data signals, an application-specific integrated circuit having suitable combinational logic gate circuits, a programmable gate array (PGA), a field-programmable gate array (FPGA), etc.

[0060] 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 the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. An intelligent control method for a pressure reducing valve, characterized in that: include: Acquire sensor data during the operation of the pressure reducing valve, build a digital twin basic physical model of the pressure reducing valve, and use the digital twin basic physical model to perform calculations to obtain basic data calculation results of the current state of the pressure reducing valve; Based on the calculation results of the basic data, real-time collected data in the dynamic operation process is obtained and pre-processed to obtain processed standardized operation characteristic data; Utilizing the standardized operational characteristic data to optimize the digital twin basic physical model, and constructing a prediction and control model that integrates physical modeling and data-driven modeling; Based on the predictive control model, the operating state is predicted to obtain the pressure regulation prediction result in the future period, and the predictive pressure regulation and virtual-real comparison abnormal diagnosis of the pressure reducing valve are realized.

2. The intelligent control method for a pressure reducing valve according to claim 1, characterized in that: Acquiring sensor data during the operation of the pressure reducing valve refers to collecting the pressure of the pressure reducing valve outlet through a pressure sensor, collecting the instantaneous flow of the pipeline through a flow sensor, collecting the medium temperature through a temperature sensor, collecting the real-time opening position information of the valve core through a valve position sensor, and collecting the micro-vibration and posture change data of the pressure reducing valve during operation through an inertial measurement unit.

3. The intelligent control method for a pressure reducing valve according to claim 1, characterized in that: The construction of the digital twin basic physical model of the pressure reducing valve includes the following steps: By introducing the fluid continuity equation and the relationship between flow resistance calculation, the outlet pressure estimation value is obtained, thereby simulating the pressure output state of the pressure reducing valve under different working conditions; By introducing the actuator inertia modeling structure, the control response delay value is obtained to compensate for the time lag between the virtual control signal and the actual action; By introducing the virtual and real pressure residual feedback mechanism, the model residual correction value is obtained, thereby realizing the adaptive correction and error learning of the model; Based on the obtained outlet pressure estimation value, control response delay value and model residual correction value, a digital twin basic physical model is constructed to comprehensively evaluate the impact of control instructions, mechanical inertia and system errors on the decompression output state.

4. The intelligent control method for a pressure reducing valve according to claim 3, characterized in that: The specific formula of the digital twin basic physical model is as follows: ; in, Output pressure value for the final digital twin basic physical model, To take into account the delayed prediction value, is the residual correction coefficient, is the estimated outlet pressure, is the model residual correction value, To control the response delay value.

5. The intelligent control method for a pressure reducing valve according to claim 4, characterized in that: The real-time collected data includes valve front pressure data, valve rear pressure data, valve core displacement data, flow data and valve body vibration intensity data; Acquiring real-time collected data during dynamic operation based on the basic data calculation results includes the following steps: A joint judgment is performed based on the post-valve pressure change and the valve core displacement change calculated in the basic data calculation result. If the post-valve pressure change value per unit time is less than the first pressure change threshold, and the valve core displacement change value is less than the first displacement change threshold, it is preliminarily determined that the current state is a low-fluctuation state, and the first sampling frequency f1 is used to collect the above-mentioned real-time data; If the post-valve pressure change value within a unit time is between the first pressure change threshold and the second pressure change threshold, or the valve core displacement change value is between the first displacement change threshold and the second displacement change threshold, it is preliminarily determined that the current state is a medium fluctuation state, and the second sampling frequency f2 is used to collect the above-mentioned real-time data; If the change value of the valve rear pressure in unit time is greater than or equal to the second pressure change threshold, or the change value of the valve core displacement is greater than or equal to the second displacement change threshold, it is preliminarily determined that the current state is a high fluctuation state, and the third sampling frequency f3 is used to collect the above real-time data, where f3>f2>f1.

6. The intelligent control method for a pressure reducing valve according to claim 5, characterized in that: The construction of the prediction and control model integrating physical modeling and data-driven includes the following steps: By introducing a delay correction function, the time axis is adjusted for the pressure prediction error caused by the control delay to obtain the response compensation pressure value, thereby improving the adaptability of the prediction model to the inertia characteristics of the control system; By introducing a state offset prediction function to model the coupled effects of flow and temperature fluctuations, an environmental disturbance prediction term is obtained, thereby improving the robustness of the model under dynamic conditions. By introducing the residual correction term and combining the above-mentioned response compensation pressure value and environmental disturbance prediction term to construct a prediction and control model, the final predicted pressure regulation target pressure value is obtained, thereby realizing the prediction and adjustment of the future operating status.

7. The intelligent control method for a pressure reducing valve according to claim 6, characterized in that: The operation state prediction based on the predictive control model is performed to obtain the pressure regulation prediction result in the future period, and the predictive pressure regulation and virtual-real comparison abnormal diagnosis of the pressure reducing valve are realized, including the following steps: Calculate the difference between the actual and virtual pressures based on the target pressure value output by the predictive control model and the actual pressure after the valve. , and combined with the current fluctuation state, select and execute the following judgment: If the initial current state is a low fluctuation state, and the virtual and real pressure difference When the error is less than the first error threshold, the system is judged to be operating normally and the voltage regulation instruction is executed according to the predictive control model; If the current state is initially determined to be medium or high fluctuation, or the difference between the real and virtual pressures If the error is greater than or equal to the first error threshold, it is preliminarily determined that the system has a control anomaly and the next step is executed; If the difference between the real and virtual pressures If the change in valve core displacement within the current sampling period is less than the second error threshold, and the valve core displacement change within the current sampling period is less than the valve core displacement threshold, and the valve body vibration intensity within the current period is less than the valve body vibration threshold, then the current state is determined to be a mild abnormality, the current prediction error and dynamic response data are recorded, and the output gain coefficient of the prediction model is adjusted; If the difference between the real and virtual pressures If the error is greater than or equal to the second error threshold, or if either the change in valve core displacement within the current sampling period or the vibration intensity of the valve body within the current period is greater than the corresponding threshold, the current state is determined to be a serious abnormal state, triggering the pressure regulation protection mechanism and terminating the current control execution process.

8. An intelligent control system for a pressure reducing valve, based on the intelligent control method according to any one of claims 1 to 7, characterized in that: include: State data acquisition module: used to obtain sensor data during the operation of the pressure reducing valve, and build a digital twin basic physical model of the pressure reducing valve. The digital twin basic physical model is used to perform calculations to obtain the basic data calculation results of the current state of the pressure reducing valve; Data processing module: used to obtain real-time collected data during dynamic operation based on the basic data calculation results and perform preprocessing to obtain processed standardized operation characteristic data; Model optimization module: used to optimize the digital twin basic physical model using the standardized operating characteristic data, and build a prediction and control model that integrates physical modeling and data-driven; Control execution module: used to predict the operating status based on the predictive control model, obtain the pressure regulation prediction results in the future period, and realize the predictive pressure regulation and virtual-real comparison abnormal diagnosis of the pressure reducing valve.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the intelligent control method for a pressure reducing valve according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the intelligent control method of a pressure reducing valve according to any one of claims 1 to 7 are implemented.

Images