Automated control valve optimization methods and systems for fluid control

Abstract

This invention belongs to the field of fluid control technology, specifically an automated control valve optimization method and system for fluid control. It includes a composite intelligent algorithm based on multivariate feedback information, employing a fusion of fuzzy PID control and model predictive control to dynamically adjust valve opening in real time. The composite intelligent algorithm constructs a system dynamic model using real-time collected multivariate data, combining the adaptability of fuzzy PID control to nonlinear elements and the decoupling capability of MPC for multivariate coupling to generate valve opening optimization commands. This invention uses a composite intelligent algorithm that integrates fuzzy PID control and model predictive control to collect real-time data such as flow rate and pressure to construct a system dynamic model. It uses fuzzy logic rules to adjust PID parameters online, compensating for system nonlinear errors, and uses the fuzzy PID output as the initial control quantity for MPC. Through a rolling optimization strategy, it decouples multivariate coupling relationships, adapting to changes in the fluid system and thereby improving the control accuracy of valve opening.

Description

Technical Field

[0001] This invention belongs to the field of fluid control technology, specifically an automated control valve optimization method and system for fluid control. Background Technology

[0002] Automated control valves are devices that use sensors, actuators, and other automated components to automatically control fluid parameters. They are widely used in industrial fluid, chemical, petroleum, and power industries. Their characteristics include: ease of use and safety; the ability to control on-site or remotely; the ability to control a single valve or to centrally control multiple valves; the ability to perform simple on / off control as well as regulating control; and the ability to achieve programmed control in the field of industrial fluid control when used with a computer.

[0003] Traditional automated valve control systems mostly employ a single control algorithm, such as traditional PID control. When dealing with complex systems like fluid control, which exhibit nonlinearity, time-varying characteristics, and uncertainty, achieving high-precision control is challenging. Because fluid parameters such as flow rate, pressure, and temperature change in real time, and the characteristics of the medium significantly impact control effectiveness, traditional algorithms cannot dynamically adjust based on this multi-variable feedback. This results in imprecise valve opening control, affecting the overall efficiency of the fluid system. Furthermore, traditional systems have shortcomings in leak detection. They typically use a single detection method with limited accuracy, making it difficult to quickly and accurately identify leaking bubbles and issue timely warnings. This can easily lead to fluid leakage accidents, increasing production costs and safety risks. Simultaneously, traditional systems have weak remote monitoring and data transmission capabilities, with significant data transmission delays, hindering real-time remote control and fault diagnosis, which is detrimental to intelligent management in industrial production.

[0004] Therefore, the present invention provides an automated control valve optimization method and system for fluid control. Summary of the Invention

[0005] In order to overcome the shortcomings of the prior art, at least one technical problem raised in the background art is solved.

[0006] The technical solution adopted by this invention to solve its technical problem is: the automated control valve optimization method for fluid control described in this invention, comprising:

[0007] Based on multivariate feedback information, a composite intelligent algorithm that integrates fuzzy PID control and model predictive control is used to dynamically adjust the valve opening in real time.

[0008] The multivariate feedback information includes flow rate, pressure, temperature, and medium characteristic parameters;

[0009] The composite intelligent algorithm constructs a dynamic model of the system by collecting multivariate data in real time, and combines the adaptability of fuzzy PID control to nonlinear elements and the decoupling capability of MPC for multivariate coupling to generate valve opening optimization instructions.

[0010] It integrates a high-precision electromagnetic flow meter and a visual leak detection module. The visual leak detection module generates detection bubbles through high-precision air pressure control and automatically identifies leaking bubbles using machine vision algorithms.

[0011] The optimization instructions are synchronously transmitted to the valve actuator and the remote monitoring platform to achieve adaptive adjustment of valve opening and real-time early warning of leakage faults.

[0012] A further improvement of this invention is that the composite intelligent algorithm integrating fuzzy PID control and MPC includes:

[0013] A system state-space model is constructed by collecting real-time flow, pressure, and temperature data.

[0014] PID parameters are adjusted online using fuzzy logic rules to compensate for system nonlinearity errors;

[0015] The fuzzy PID output is used as the initial control variable for MPC. The multivariable coupling relationship is decoupled through a rolling optimization strategy to generate valve opening optimization instructions.

[0016] The rolling optimization strategy takes minimizing the system tracking error and the rate of change of the control quantity as the objective function, and uses a quadratic programming algorithm to solve for the optimal control sequence.

[0017] A further improvement of this invention is that the high-precision electromagnetic flowmeter employs low-frequency rectangular wave excitation technology with an excitation frequency of 6.25 Hz. The excitation current is 200 ;

[0018] The air pressure control accuracy of the visual leak detection module is ±0.01 kPa, the bubble generation frequency is 5-20 bubbles / second, and the machine vision algorithm adopts... The target detection model has a training dataset containing ≥100,000 leaked bubble images and a detection speed of ≥30 frames / second.

[0019] A further improvement of this invention is that the remote monitoring platform integrates an Internet of Things (IoT) communication module, supporting... Multimode communication protocol, data transmission delay ≤100ms;

[0020] The platform adopts an edge computing architecture, which completes data preprocessing and preliminary analysis on local nodes, and only uploads key feature data to the cloud server.

[0021] The cloud server deploys a digital twin model that maps the physical state of the valve in real time, supporting virtual debugging and fault prediction.

[0022] A further improvement of the present invention is that the valve actuator adopts a modular design, including a motor drive module, a transmission module and a sealing module;

[0023] The motor drive module integrates a fast-response motor and an absolute encoder, with a motor start-up time ≤50ms and an encoder resolution ≥16 bits.

[0024] The transmission module uses a harmonic reducer with a transmission ratio of 1:100 and a transmission efficiency of ≥90%.

[0025] The sealing module adopts a metal bellows sealing structure with a leakage rate ≤1×10⁻⁶. -9 Pa·m³ / s, pressure resistance ≥40MPa.

[0026] A further improvement of this invention is that the medium characteristic parameters include medium density, viscosity, corrosivity, and solid content;

[0027] The composite intelligent algorithm dynamically adjusts the control strategy based on the medium characteristic parameters as follows:

[0028] When the viscosity of the medium changes by ≥10%, it automatically switches to viscosity compensation mode to compensate for the impact of viscosity on flow rate through feedforward control.

[0029] When the corrosivity of the medium exceeds the preset threshold, the sealing module self-test program is activated to detect the wear of the seals and generate maintenance recommendations.

[0030] A further improvement of this invention is that it also includes an energy efficiency optimization module based on reinforcement learning. The energy efficiency optimization module uses valve energy consumption and system efficiency as state variables and valve opening adjustment amount as action variables. It trains the energy efficiency optimization model using a deep deterministic strategy gradient algorithm. The model learns the optimal control strategy through interaction with the fluid system. Under the premise of meeting the flow control accuracy, it minimizes valve energy consumption and improves energy efficiency by no less than 15%.

[0031] A further improvement of this invention is that it also includes a virtual debugging module based on digital twins, wherein the virtual debugging module constructs a digital twin of the valve on a cloud server, and the digital twin integrates a fluid dynamics model, a mechanical structure model, and a control algorithm model;

[0032] The effectiveness of the control strategy is verified through simulation testing, and potential design flaws are identified in advance.

[0033] The virtual debugging module supports bidirectional data synchronization with the physical valves, enabling virtual-physical linkage debugging.

[0034] An automated control valve optimization system for fluid control, applied to the aforementioned automated control valve optimization method for fluid control, includes:

[0035] The intelligent control unit is used to execute and generate valve opening optimization commands;

[0036] A multi-parameter sensor group, including an electromagnetic flowmeter, pressure sensor, temperature sensor, and media characteristic detection module, is used to collect multivariate feedback information;

[0037] A visual leak detection unit is used to generate detection bubbles and identify leak faults;

[0038] Valve actuator, used to adjust valve opening according to optimization instructions;

[0039] The remote monitoring platform is used to receive and display valve operation data, and supports remote control and fault early warning.

[0040] The intelligent control unit is connected to a multi-parameter sensor group, a visual leak detection unit, a valve actuator, and a remote monitoring platform to form a closed-loop control system.

[0041] A further improvement of this invention is that the intelligent control unit adopts an embedded architecture, integrating a high-performance microprocessor and an FPGA coprocessor;

[0042] The microprocessor runs a composite intelligent algorithm and an energy efficiency optimization model, while the FPGA coprocessor is responsible for data acquisition and preprocessing.

[0043] The intelligent control unit supports multi-protocol communication, including and It is compatible with existing industrial control systems;

[0044] The system also includes a local storage module that stores ≥30 days of operational data and supports data backtracking and fault analysis.

[0045] The beneficial effects of this invention are as follows:

[0046] 1. This invention employs a composite intelligent algorithm integrating fuzzy PID control and model predictive control. It dynamically adjusts valve opening in real time using multivariate feedback information. This algorithm constructs a dynamic system model by collecting data such as flow rate, pressure, and temperature in real time. It then uses fuzzy logic rules to adjust PID parameters online, compensating for system nonlinear errors. The fuzzy PID output is used as the initial control variable for MPC. A rolling optimization strategy decouples multivariate coupling relationships, enabling more precise adaptation to complex changes in the fluid system and improving the control accuracy of valve opening. This addresses the limitations of traditional algorithms in effectively handling multivariate and nonlinear problems. Secondly, it integrates a high-precision electromagnetic flowmeter and a visual leak detection module. The visual leak detection module generates detection bubbles through high-precision air pressure control and automatically identifies leak bubbles using machine vision algorithms. This multi-method detection approach significantly improves the accuracy and speed of leak detection, enabling timely detection and real-time warnings of leaks, effectively preventing fluid leakage accidents, and thus reducing production costs and safety risks.

[0047] 2. This invention establishes a remote monitoring platform integrating an IoT communication module and adopts an edge computing architecture. Data preprocessing and preliminary analysis are completed at local nodes, with only key feature data uploaded to the cloud server. This achieves efficient remote monitoring and low-latency data transmission, facilitating real-time remote control and fault diagnosis, and improving the intelligent management level of industrial production. The motor drive module in valve actuation integrates a fast-response motor and an absolute encoder, while the sealing module uses a metal bellows sealing structure. The modular design makes maintenance and replacement more convenient, reducing maintenance costs. Simultaneously, the high-performance design of each module ensures the reliable operation of the valve actuator. Attached Figure Description

[0048] The invention will now be further described with reference to the accompanying drawings.

[0049] Figure 1 This is a flowchart of the present invention. Detailed Implementation

[0050] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0051] Please see Figure 1 ,

[0052] Automated control valve optimization methods for fluid control include:

[0053] Based on multivariate feedback information, a composite intelligent algorithm that integrates fuzzy PID control and model predictive control is used to dynamically adjust the valve opening in real time.

[0054] It should be noted that fluid systems are characterized by nonlinearity, time-varying nature, and multivariable coupling, making precise control difficult with a single control algorithm. This composite intelligent algorithm combines the adaptability of fuzzy PID control to nonlinear elements with the decoupling capability of MPC for multivariable coupling. It can dynamically adjust valve opening based on real-time feedback information from multiple variables such as flow rate, pressure, temperature, and media characteristic parameters, improving control accuracy and stability. For example, in chemical production processes, the flow rate and pressure of fluids constantly change with the reaction, while the viscosity of the medium may also change due to temperature variations. Using this composite intelligent algorithm, valve opening can be precisely adjusted based on these real-time changes, ensuring stable fluid flow according to set parameters.

[0055] Multivariate feedback information includes flow rate, pressure, temperature, and medium characteristic parameters;

[0056] It should be noted that comprehensively considering various key parameters of the fluid system can more accurately reflect the actual operating state of the system, providing sufficient information support for composite intelligent algorithms, thereby achieving more precise control. For example, in oil transportation pipelines, flow rate determines the transport capacity, pressure affects transport efficiency and safety, temperature may affect the viscosity and fluidity of the fluid, and medium characteristic parameters such as solid content may affect pipeline wear. By comprehensively considering these parameters, valve opening can be better controlled, ensuring the stability and safety of oil transportation.

[0057] Composite intelligent algorithms that integrate fuzzy PID control and MPC include:

[0058] A system state-space model is constructed by collecting real-time flow, pressure, and temperature data.

[0059] PID parameters are adjusted online using fuzzy logic rules to compensate for system nonlinearity errors;

[0060] The fuzzy PID output is used as the initial control variable for MPC. The multivariable coupling relationship is decoupled through a rolling optimization strategy to generate valve opening optimization instructions.

[0061] The rolling optimization strategy takes minimizing the system tracking error and the rate of change of the control quantity as the objective function, and uses a quadratic programming algorithm to solve for the optimal control sequence.

[0062] The composite intelligent algorithm constructs a dynamic model of the system by collecting multivariate data in real time, and combines the adaptability of fuzzy PID control to nonlinear elements and the decoupling capability of MPC to multivariate coupling to generate valve opening optimization instructions.

[0063] It should be noted that constructing a dynamic system model allows for a better understanding of the system's operational patterns. Utilizing the advantages of fuzzy PID and MPC to generate optimized instructions can effectively address nonlinear and multivariate coupling issues in the system, improving the accuracy and response speed of valve opening control. For example, in the fluid control system of a wastewater treatment plant, parameters such as wastewater flow rate and water quality are constantly changing, and there are coupling relationships between various control links. By constructing a dynamic model and using composite intelligent algorithms to generate optimized instructions, valve opening can be controlled more precisely, improving wastewater treatment efficiency.

[0064] It integrates a high-precision electromagnetic flow meter and a visual leak detection module. The visual leak detection module generates detection bubbles through high-precision air pressure control and automatically identifies leaking bubbles using machine vision algorithms.

[0065] The high-precision electromagnetic flowmeter employs low-frequency rectangular wave excitation technology with an excitation frequency of 6.25 Hz. The excitation current is 200 ;

[0066] It should be noted that high-precision electromagnetic flowmeters can accurately measure fluid flow, while visual leak detection modules can promptly detect leaks. The combination of these two technologies improves the accuracy and reliability of fluid control, reducing losses and safety risks caused by fluid leaks. For example, in natural gas pipelines, high-precision electromagnetic flowmeters can accurately measure the flow rate of natural gas, ensuring accurate delivery. Visual leak detection modules can monitor pipeline leaks in real time, issuing an alarm immediately upon detecting leaking bubbles to prevent safety accidents caused by natural gas leaks.

[0067] The visual leak detection module has a pressure control accuracy of ±0.01 kPa and a bubble generation frequency of 5-20 bubbles / second. The machine vision algorithm employs... The target detection model has a training dataset containing ≥100,000 leaked bubble images and a detection speed of ≥30 frames / second.

[0068] The optimized command is transmitted synchronously to the valve actuator and remote monitoring platform to achieve adaptive adjustment of valve opening and real-time early warning of leakage faults.

[0069] It should be noted that adaptive valve opening adjustment can improve the system's automation level and operational efficiency. Real-time early warning of leakage faults allows for timely intervention to prevent accidents from escalating and ensure the safe and stable operation of the system. For example, in the steam system of a thermal power plant, by synchronously transmitting optimization commands, valves can automatically adjust their opening according to system needs, ensuring a stable steam supply. Furthermore, in the event of a leak, the remote monitoring platform can receive early warning information promptly, allowing staff to take swift action.

[0070] The remote monitoring platform integrates an IoT communication module, supporting... Multimode communication protocol, data transmission delay ≤100ms;

[0071] The platform adopts an edge computing architecture, completing data preprocessing and preliminary analysis on local nodes, and only uploading key feature data to the cloud server;

[0072] A digital twin model is deployed on a cloud server to map the physical state of the valve in real time, supporting virtual debugging and fault prediction.

[0073] Furthermore, the valve actuator adopts a modular design, including a motor drive module, a transmission module, and a sealing module;

[0074] The motor drive module integrates a fast-response motor and an absolute encoder, with a motor start-up time of ≤50ms and an encoder resolution of ≥16 bits;

[0075] The transmission module uses a harmonic reducer with a transmission ratio of 1:100 and a transmission efficiency of ≥90%.

[0076] The sealing module adopts a metal bellows sealing structure with a leakage rate ≤1×10 -9 Pa·m³ / s, pressure resistance ≥40MPa.

[0077] Medium characteristic parameters include medium density, viscosity, corrosivity, and solids content;

[0078] The composite intelligent algorithm dynamically adjusts the control strategy based on the medium characteristic parameters as follows:

[0079] When the viscosity of the medium changes by ≥10%, it automatically switches to viscosity compensation mode to compensate for the impact of viscosity on flow rate through feedforward control.

[0080] When the corrosivity of the medium exceeds the preset threshold, the sealing module self-test program is activated to detect the wear of the seals and generate maintenance recommendations.

[0081] It also includes an energy efficiency optimization module based on reinforcement learning. The energy efficiency optimization module uses valve energy consumption and system efficiency as state variables and valve opening adjustment amount as action variables. It uses a deep deterministic strategy gradient algorithm to train the energy efficiency optimization model. The model learns the optimal control strategy through interaction with the fluid system. Under the premise of meeting the flow control accuracy, it minimizes valve energy consumption and improves energy efficiency by no less than 15%, effectively solving the problem of low energy efficiency in traditional systems.

[0082] It also includes a virtual debugging module based on digital twins, which builds a digital twin of the valve on a cloud server. The digital twin integrates fluid dynamics model, mechanical structure model and control algorithm model.

[0083] The effectiveness of the control strategy is verified through simulation testing, and potential design flaws are identified in advance.

[0084] The virtual debugging module supports bidirectional data synchronization with physical valves, enabling virtual-physical linkage debugging.

[0085] Example 1

[0086] This embodiment provides an automated control valve optimization method for fluid control, comprising the following steps:

[0087] S1. Preparation Phase:

[0088] Install a multi-parameter sensor group, including an electromagnetic flow meter, pressure sensor, temperature sensor and media characteristic detection module, to collect multivariate feedback information of the fluid system in real time;

[0089] Install a visual leak detection unit, generate detection bubbles through a high-precision air pressure control device, and use a camera with machine vision algorithms to identify the bubbles;

[0090] Assemble the valve actuator, and install the motor drive module, transmission module and sealing module according to the modular design requirements to ensure that the performance of each module meets the design standards;

[0091] Build a remote monitoring platform, integrate IoT communication modules, and configure support. Devices using multi-mode communication protocols, and the establishment of edge computing nodes and cloud servers;

[0092] The system is equipped with an intelligent control unit, which adopts an embedded architecture and integrates a high-performance microprocessor and an FPGA coprocessor. The composite intelligent algorithm and energy efficiency optimization model are deployed to the microprocessor.

[0093] S2, Processing Stage:

[0094] The multi-parameter sensor group collects data such as flow rate, pressure, temperature and medium characteristics in real time and transmits them to the intelligent control unit;

[0095] The composite intelligent algorithm in the intelligent control unit constructs a dynamic model of the system based on the collected data, adjusts the PID parameters online using fuzzy logic rules, and uses the fuzzy PID output as the initial control quantity of the MPC. It then generates valve opening optimization instructions through a rolling optimization strategy.

[0096] The valve actuator adjusts the valve opening according to the optimization instructions to achieve precise fluid control;

[0097] The visual leak detection unit monitors the fluid system in real time. Once a leaking bubble is detected, it immediately sends a signal to the intelligent control unit. The intelligent control unit then transmits the leak fault information to the remote monitoring platform to achieve real-time early warning.

[0098] The remote monitoring platform receives and displays valve operation data, supports remote control and fault diagnosis, and allows operators to remotely adjust valve opening or perform system maintenance through the platform.

[0099] The energy efficiency optimization module monitors valve energy consumption and system efficiency in real time. Based on the operating conditions, it uses a deep deterministic strategy gradient algorithm to train the energy efficiency optimization model and dynamically adjusts the valve opening to minimize valve energy consumption while meeting flow control accuracy requirements.

[0100] S3, Dynamic Strategy Adjustment:

[0101] When the characteristics of the medium change, such as a change in medium viscosity exceeding 10%, the composite intelligent algorithm automatically switches to viscosity compensation mode and adjusts the valve opening through feedforward control to compensate for the impact of viscosity on flow rate.

[0102] When the corrosivity of the medium exceeds the preset threshold, the intelligent control unit starts the sealing module self-test program to detect the wear of the seals and generate maintenance suggestions based on the test results, reminding the operator to maintain or replace the seals.

[0103] Example 2

[0104] This embodiment provides an automated control valve optimization system for fluid control, including: an intelligent control unit, a multi-parameter sensor group, a visual leak detection unit, a valve actuator, and a remote monitoring platform;

[0105] The intelligent control unit adopts an embedded architecture, making it compact, efficient, and low-power, adaptable to complex industrial environments. The integrated high-performance microprocessor possesses powerful computing capabilities, enabling rapid execution of composite intelligent algorithms and energy efficiency optimization models to ensure real-time and accurate optimization of valve opening. The FPGA coprocessor, with its parallel processing capabilities, efficiently handles data acquisition and preprocessing, reducing the burden on the microprocessor and improving the overall system response speed.

[0106] Support includes and Multi-protocol communication enables seamless integration with different types of industrial equipment and management systems, ensuring compatibility with various existing industrial control systems and facilitating system integration and expansion. For example, when connecting to a factory's MES (Manufacturing Execution System), stable data transmission and interaction can be achieved through the OPC UA protocol.

[0107] The multi-parameter sensor group includes an electromagnetic flow meter, a pressure sensor, a temperature sensor, and a media characteristic detection module;

[0108] Electromagnetic flow meters utilize advanced electromagnetic induction principles to accurately measure fluid flow rates, unaffected by changes in fluid density, viscosity, temperature, pressure, and conductivity. They offer high measurement accuracy and fast response, providing precise flow feedback information to intelligent control units.

[0109] Pressure sensors are used to monitor pressure changes in fluid systems in real time. They employ high-precision sensing elements to detect minute pressure fluctuations and convert the pressure signal into an electrical signal, which is then transmitted to the intelligent control unit.

[0110] Temperature sensors are used to measure the temperature of fluids. They have high sensitivity and stability, and can operate accurately over a wide temperature range. They provide temperature parameters to the system so that the effects of temperature on fluid characteristics and control performance can be taken into account.

[0111] The media characteristic detection module can detect characteristic parameters of the medium such as density, viscosity, corrosivity, and solids content. By combining multiple detection technologies, such as optical detection and electrochemical detection, a comprehensive understanding of the medium's characteristics can be obtained, providing a basis for the dynamic adjustment of control strategies by composite intelligent algorithms.

[0112] The visual leak detection unit generates detection bubbles through a high-precision air pressure control system with an accuracy of ±0.01 kPa, ensuring the stability and consistency of bubble generation. At the same time, it utilizes deep learning-based machine vision algorithms, such as the YOLOv8 object detection model, to quickly and accurately identify leaking bubbles. This model has been trained on a large number of leaking bubble images, with a training dataset containing ≥100,000 images and a detection speed of ≥30 frames / second. It can effectively distinguish between normal bubbles and leaking bubbles, and promptly detect leak faults.

[0113] The valve actuator adopts a modular design, including a motor drive module, a transmission module, and a sealing module, facilitating installation, maintenance, and replacement. The motor drive module integrates a fast-response motor and an absolute encoder, with a motor start-up time ≤50ms and an encoder resolution ≥16 bits, enabling rapid and precise valve control. The transmission module uses a harmonic reducer with a transmission ratio of 1:100 and a transmission efficiency ≥90%, ensuring high efficiency and stability in power transmission. The sealing module employs a metal bellows sealing structure with a leakage rate ≤1×10⁻⁶. - 9 Pa·m³ / s, pressure resistance ≥40MPa, effectively prevents fluid leakage and ensures safe operation of the system;

[0114] The remote monitoring platform receives valve operating data in real time and displays it in intuitive charts and curves, allowing operators to easily understand the valve's operating status. It supports remote control, enabling operators to remotely adjust valve openings and intervene in the fluid system. Furthermore, it features a fault warning function; when the system detects an anomaly, it promptly issues an alarm to notify relevant personnel for handling.

[0115] Equipped with a large-capacity storage device, it can store ≥30 days of operating data and supports data backtracking and fault analysis. By analyzing historical data, the system's operating patterns and potential problems can be discovered, providing a basis for system optimization and improvement.

[0116] The terms "front," "back," "left," "right," "top," and "bottom" all refer to the figures in the accompanying drawings. Figure 1 Based on the perspective of the observer, the side of the device facing the observer is defined as the front, the left side of the observer is defined as the left, and so on.

[0117] In the description of this invention, it should be understood that the terms "center", "longitudinal", "lateral", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limiting the scope of protection of this invention.

[0118] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. An automated control valve optimization method for fluid control, characterized in that, include: Based on multivariate feedback information, a composite intelligent algorithm that integrates fuzzy PID control and model predictive control is used to dynamically adjust the valve opening in real time. The multivariate feedback information includes flow rate, pressure, temperature, and medium characteristic parameters; The composite intelligent algorithm constructs a dynamic model of the system by collecting multivariate data in real time, and combines the adaptability of fuzzy PID control to nonlinear elements and the decoupling capability of MPC for multivariate coupling to generate valve opening optimization instructions. It integrates a high-precision electromagnetic flow meter and a visual leak detection module. The visual leak detection module generates detection bubbles through high-precision air pressure control and automatically identifies leaking bubbles using machine vision algorithms. The optimization instructions are synchronously transmitted to the valve actuator and the remote monitoring platform to achieve adaptive adjustment of valve opening and real-time early warning of leakage faults.

2. The method for optimizing automated control valves for fluid control according to claim 1, characterized in that: The composite intelligent algorithm integrating fuzzy PID control and MPC includes: A system state-space model is constructed by collecting real-time flow, pressure, and temperature data. PID parameters are adjusted online using fuzzy logic rules to compensate for system nonlinearity errors; The fuzzy PID output is used as the initial control variable for MPC. The multivariable coupling relationship is decoupled through a rolling optimization strategy to generate valve opening optimization instructions. The rolling optimization strategy takes minimizing the system tracking error and the rate of change of the control quantity as the objective function, and uses a quadratic programming algorithm to solve for the optimal control sequence.

3. The method for optimizing automated control valves for fluid control according to claim 1, characterized in that: The high-precision electromagnetic flowmeter employs low-frequency rectangular wave excitation technology with an excitation frequency of 6.25 Hz. The excitation current is 200 ; The air pressure control accuracy of the visual leak detection module is ±0.01 kPa, the bubble generation frequency is 5-20 bubbles / second, and the machine vision algorithm adopts... The target detection model has a training dataset containing ≥100,000 leaked bubble images and a detection speed of ≥30 frames / second.

4. The method for optimizing automated control valves for fluid control according to claim 1, characterized in that: The remote monitoring platform integrates an IoT communication module, supporting... Multimode communication protocol, data transmission delay ≤100ms; The platform adopts an edge computing architecture, which completes data preprocessing and preliminary analysis on local nodes, and only uploads key feature data to the cloud server. The cloud server deploys a digital twin model that maps the physical state of the valve in real time, supporting virtual debugging and fault prediction.

5. The method for optimizing automated control valves for fluid control according to claim 1, characterized in that: The valve actuator adopts a modular design, including a motor drive module, a transmission module and a sealing module; The motor drive module integrates a fast-response motor and an absolute encoder, with a motor start-up time ≤50ms and an encoder resolution ≥16 bits. The transmission module uses a harmonic reducer with a transmission ratio of 1:100 and a transmission efficiency of ≥90%. The sealing module adopts a metal bellows sealing structure with a leakage rate ≤1×10⁻⁶. -9 Pa·m³ / s, pressure resistance ≥40MPa.

6. The method for optimizing automated control valves for fluid control according to claim 1, characterized in that: The medium characteristic parameters include medium density, viscosity, corrosivity, and solids content; The composite intelligent algorithm dynamically adjusts the control strategy based on the medium characteristic parameters as follows: When the viscosity of the medium changes by ≥10%, it automatically switches to viscosity compensation mode to compensate for the impact of viscosity on flow rate through feedforward control. When the corrosivity of the medium exceeds the preset threshold, the sealing module self-test program is activated to detect the wear of the seals and generate maintenance recommendations.

7. The method for optimizing automated control valves for fluid control according to claim 1, characterized in that: It also includes an energy efficiency optimization module based on reinforcement learning, which uses valve energy consumption and system efficiency as state variables and valve opening adjustment amount as action variables, and uses a deep deterministic strategy gradient algorithm to train the energy efficiency optimization model. The model learns the optimal control strategy through interaction with the fluid system, minimizing valve energy consumption while meeting flow control accuracy requirements.

8. The method for optimizing automated control valves for fluid control according to claim 1, characterized in that: It also includes a virtual debugging module based on digital twins, which constructs a digital twin of the valve on a cloud server. The digital twin integrates a fluid dynamics model, a mechanical structure model, and a control algorithm model. The effectiveness of the control strategy is verified through simulation testing, and potential design flaws are identified in advance. The virtual debugging module supports bidirectional data synchronization with the physical valves, enabling virtual-physical linkage debugging.

9. An automated control valve optimization system for fluid control, applied to the automated control valve optimization method for fluid control as described in any one of claims 1-8, characterized in that, include: The intelligent control unit is used to execute and generate valve opening optimization commands; A multi-parameter sensor group, including an electromagnetic flowmeter, pressure sensor, temperature sensor, and media characteristic detection module, is used to collect multivariate feedback information; A visual leak detection unit is used to generate detection bubbles and identify leak faults; Valve actuator, used to adjust valve opening according to optimization instructions; The remote monitoring platform is used to receive and display valve operation data, and supports remote control and fault early warning. The intelligent control unit is connected to a multi-parameter sensor group, a visual leak detection unit, a valve actuator, and a remote monitoring platform to form a closed-loop control system.

10. The automated control valve optimization system for fluid control according to claim 9, characterized in that: The intelligent control unit adopts an embedded architecture, integrating a high-performance microprocessor and an FPGA coprocessor. The microprocessor runs a composite intelligent algorithm and an energy efficiency optimization model, while the FPGA coprocessor is responsible for data acquisition and preprocessing. The intelligent control unit supports multi-protocol communication, including and It is compatible with existing industrial control systems; The system also includes a local storage module that stores ≥30 days of operational data and supports data backtracking and fault analysis.

Images