Valve intelligent adjustment and health management method based on multi-parameter fusion and state machine decision

Abstract

The invention discloses an intelligent valve adjusting method based on multi-parameter dynamic perception. The method comprises the steps that firstly, upstream and downstream pressure, valve positions, valve rod stress, system set values and other multi-dimensional parameters of a valve are collected in real time; then dynamically judging whether the system is in a stable or adjusting state through calculation and comparison; and finally, executing a corresponding intelligent strategy according to the state: in a stable period, executing micro-disturbance self-learning to monitor the mechanical state of the valve; and in the adjusting period, a multi-stage dynamic adjusting strategy comprising three stages of rapid approaching, smooth deceleration and accurate locking is executed, and it is ensured that the valve rapidly, stably and accurately reaches the target station. The problems that a traditional valve control method is prone to overregulation and oscillation and slow in response are effectively solved, and the control precision, the system stability and the equipment reliability are improved.

Description

Technical Field

[0001] This invention relates to the field of industrial process control technology, specifically to a valve control method, and more particularly to an intelligent working method that achieves precise, stable and energy-saving regulation by sensing multiple operating parameters in real time and dynamically adjusting the valve's action logic. Background Technology

[0002] As a key control component in fluid transport systems, the regulating performance of valves directly affects the overall system's operating efficiency, safety, and stability. Traditional valve control methods, whether simple manual control or automatic control based on a single parameter (such as pressure or flow), have significant limitations.

[0003] First, traditional methods typically employ fixed control logic. For example, a PID controller calculates the valve opening command based on the deviation between the pressure value at a detection point and the setpoint. This single-parameter, fixed-logic control method struggles to achieve precise adjustment in systems with complex operating conditions or frequent load fluctuations. The system is prone to oscillations, causing continuous fluctuations in controlled parameters (such as pressure and flow rate), affecting the stable operation of downstream equipment. Second, when faced with sudden changes in large flow demands, traditional PID control is slow to respond or may cause excessive valve movement due to integral saturation, resulting in a "water hammer" effect that impacts pipelines and equipment. Furthermore, traditional control methods lack awareness and compensation for the valve's own state. When the valve experiences wear, jamming, or other performance degradation due to long-term operation, its control accuracy significantly decreases, potentially leading to malfunctions.

[0004] Therefore, there is an urgent need for a smart valve regulation method that can adapt to complex working conditions, respond quickly, and has self-adaptive capabilities. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a valve intelligent regulation method based on multi-parameter dynamic sensing. This method integrates and analyzes multiple key operating parameters and introduces a dynamic adjustment strategy, enabling the valve to complete regulation tasks intelligently, smoothly, and efficiently.

[0006] To achieve the above objectives, the technical solution adopted by this invention is: a valve intelligent regulation and health management method based on multi-parameter fusion and state machine decision-making. Its innovation lies in the fact that this method is executed by a valve intelligent controller and includes the following steps: S1. Data Acquisition and Preprocessing Steps: Periodically and synchronously acquire valve inlet pressure, outlet pressure, fluid temperature, real-time opening degree, valve stem axial force, and actuator drive signal, and perform filtering, time delay compensation, and validity verification on the acquired raw data; S2. State identification step: Based on the preprocessed data, calculate the dynamic differential pressure and instantaneous theoretical flow rate; by comparing the deviation between the set flow rate and the instantaneous theoretical flow rate with at least two preset thresholds, set a state flag bit representing the system's adjustment requirements; and by analyzing the relationship between the valve stem axial force, opening degree, and differential pressure, set a health flag bit representing the valve's mechanical health status. S3. Multi-mode intelligent adjustment step: Based on the status flag and health flag, dynamically select and execute the corresponding control strategy from a set of operating modes containing at least the following modes: S31. Steady-state health monitoring mode: When in steady state, micro-disturbance detection signals are periodically injected into the actuator, and mechanical fault diagnosis and early warning are performed based on the response of valve stem axial force and opening degree; S32, Dynamic Adjustment Mode: When in a state requiring significant adjustment, a multi-stage trajectory tracking control strategy is executed, including a rapid approach phase, a smooth deceleration phase, and a precise locking phase.

[0007] Furthermore, in the data acquisition and preprocessing steps, the actuator drive signal is the air chamber pressure of the pneumatic actuator or the motor current of the electric actuator; the validity verification includes removing data points that exceed the physically reasonable range and replacing them with the effective value of the previous cycle or the Kalman filter prediction value.

[0008] Furthermore, in the state identification step, the logic for setting the state flag bit includes: If the deviation is less than or equal to the first threshold and the duration exceeds it, it is determined to be in a steady state; If the deviation is greater than the second threshold, it is immediately determined to be an adjustment state that requires significant adjustment; If the deviation is between the first threshold and the second threshold, it is determined to be in a fine-tuning state; wherein the second threshold is greater than the first threshold.

[0009] Furthermore, the homeostatic health monitoring mode specifically includes: Generate a micro-disturbance detection signal with an amplitude of less than 0.5% of the rated stroke and a duration of less than 300 milliseconds, with a period of not less than 30 seconds; Analyze the starting torque value of the valve stem axial force in response to the detection signal and the frictional force change curve during the movement process; If the starting torque value exceeds the baseline threshold established based on historical data, it is determined that there is a risk of jamming, and a jamming warning signal is generated. If the response of the opening shows an increased hysteresis or a non-monotonic change, it is determined that there is a risk of transmission clearance and a loosening warning signal is generated.

[0010] Furthermore, the multi-stage trajectory tracking control strategy in the dynamic adjustment mode is specifically as follows: Rapid approach phase: The goal is to control the valve to move towards the target opening at the maximum permissible speed. Smooth deceleration section: When the difference between the real-time opening and the target opening enters the first range, it switches to tracking a preset smooth speed curve, so that the valve's movement speed smoothly decreases to near zero; Precise locking segment: When the difference enters a second range that is less than the first range, the control target is switched from tracking the opening to stabilizing the downstream pressure, and the fluid system is determined to be in equilibrium by monitoring the rate of change of the dynamic pressure difference; when the rate of change is continuously lower than the threshold and maintained for a predetermined time, the adjustment is determined to be completed and the mode is exited.

[0011] Furthermore, in the precise locking section, an independently configured PID controller with parameters tuned to a small deviation range is used to stabilize and regulate the downstream pressure.

[0012] Furthermore, the method also includes a step: under steady-state conditions, the stored valve flow characteristic curve is slowly and adaptively corrected using long-term recorded flow setpoint, real-time opening, and dynamic differential pressure data to compensate for valve characteristic drift.

[0013] Furthermore, the health flag is used to dynamically correct the controller output intensity in other operating modes. When a jamming risk is detected, the amplitude or energy of the control output is temporarily increased.

[0014] The beneficial effects of this invention are: Multi-parameter fusion enables more precise control: By comprehensively analyzing multi-dimensional information such as upstream and downstream pressure, temperature, valve position, and force, it overcomes the one-sidedness of single-parameter control, making the valve's adjustment action more in line with actual working conditions.

[0015] Intelligent status recognition for faster response: The system is clearly divided into two states, "stable" and "regulation", and different strategies are adopted to ensure energy efficiency and equipment monitoring during the stable period, and to ensure rapid response during the regulation period.

[0016] Dynamic adjustment strategy for smoother operation: The innovative three-stage adjustment method of "rapid approach - smooth deceleration - precise locking" effectively avoids over-adjustment and system oscillation caused by inertia when the valve approaches the target position, greatly improves the stability of the system and eliminates the risk of "water hammer".

[0017] It has self-learning and early warning capabilities: Through the "micro-disturbance self-learning strategy", it can diagnose the mechanical health status of valves during normal operation, realize predictive maintenance, and improve the reliability and safety of the system. Detailed Implementation

[0018] The present invention will be further described below with reference to embodiments.

[0019] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0020] A valve intelligent control method based on multi-parameter dynamic sensing, the method includes the following steps: Step S1: Multi-parameter data acquisition and fusion Real-time acquisition of multi-dimensional parameters related to valve operation, including: Upstream parameters: fluid pressure P_in and temperature T_in at the valve inlet; Downstream parameter: Fluid pressure P_out at the valve outlet; Valve body parameters: real-time valve opening value K_v and axial force F on the valve stem; System requirement parameters: Flow setpoint Q_set from the central control system.

[0021] The above parameters are fused together to form a comprehensive state vector State = [P_in, T_in, P_out, K_v, F, Q_set].

[0022] Step S2: Dynamic differential pressure calculation and status assessment Based on the collected parameters, the dynamic pressure difference ΔP = P_in - P_out on both sides of the valve is calculated in real time.

[0023] Based on the dynamic pressure difference ΔP and the flow setpoint Q_set, combined with the valve flow characteristic curve preset in the controller, the theoretical ideal opening K_ideal under the current operating conditions is calculated.

[0024] The theoretical ideal valve opening K_ideal is compared with the real-time valve opening value K_v, and a status flag is defined: If |K_ideal - K_v|≤δ (δ is a preset small threshold), then the system is determined to be in a stable state; If |K_ideal - K_v|>δ, then the system is determined to be in a state of adjustment.

[0025] Step S3: Selection and Execution of State-Based Intelligent Adjustment Strategy Based on the judgment result of step S2, different adjustment strategies are implemented: When in a steady state, implement a micro-perturbation self-learning strategy: A small, alternating positive and negative micropulse signal with a very small amplitude is applied to the valve control signal at intervals of long period T_long (e.g., 30 seconds), and the gradient ΔF / Δt of the change in the valve stem axial force F is monitored.

[0026] If ΔF / Δt exceeds the preset sensitivity threshold, it is determined that there is a potential risk of valve jamming, and the force of subsequent adjustment actions will be automatically increased by one level, and a warning signal will be issued.

[0027] This strategy allows for continuous monitoring of the valve's mechanical condition without compromising system stability.

[0028] When in a state of adjustment, a multi-stage dynamic adjustment strategy is implemented, which is further divided into: S3.2.1: Rapid Approach Phase With a large control gain, the valve is driven to move rapidly from the current opening degree K_v to the theoretical ideal opening degree K_ideal.

[0029] S3.2.2: Smooth deceleration phase When the difference between the valve opening and the target opening, |K_ideal - K_v|, enters the first preset range ε1, the control gain is switched to a smaller value, and a smooth function that decays exponentially with time is introduced to correct the control command, so that the valve action speed gradually decreases.

[0030] S3.2.3: Precise Locking Phase When the difference between the valve opening and the target opening, |K_ideal - K_v|, enters the second preset range ε2 (ε2 < ε1), the precise locking stage begins. During this stage: The tracking of the flow setpoint Q_set is paused, and instead, the PID adjustment is made with a very small amplitude to maintain the stability of the downstream pressure P_out.

[0031] Simultaneously, the rate of change of the dynamic differential pressure ΔP is monitored in real time. If the rate of change of ΔP approaches zero and remains so for more than a set time, it is determined that the system has reached a new equilibrium point, the current valve opening is locked, and the regulation state is exited, returning to step S1. The method of this invention is implemented by a program module embedded in a valve controller or a host computer. This controller must at least have the ability to receive various sensor signals described in step S1 above, and the ability to output control signals to drive the valve actuator. Example

[0032] Taking a pressure regulating valve at the outlet of a pumping station as an example, its task is to maintain the downstream pipeline pressure at a stable 5.0 MPa.

[0033] Initialization: The system is powered on and the controller loads preset parameters, such as δ=0.5%, ε1=3%, ε2=0.8%, T_long=30s, etc.

[0034] Step S1: The controller continuously collects the inlet pressure P_in, outlet pressure P_out, valve opening K_v, valve stem force F, and pressure setpoint P_set (5.0 MPa in this example, which is equivalent to a form of flow setpoint).

[0035] Step S2: Calculate ΔP and the deviation between the current outlet pressure and the set value in real time. Calculate the ideal valve opening K_ideal required to eliminate the deviation based on the deviation. Assuming the current K_v is 50%, K_ideal is calculated to be 52%, and |52%-50%|=2%>δ=0.5%, the system determines that it has entered the regulation state.

[0036] Step S3: Implement a multi-stage dynamic adjustment strategy.

[0037] Rapid approach phase: The controller uses a high proportional gain to output a control signal, causing the valve to open rapidly from 50% to 52%.

[0038] Smooth deceleration phase: When the valve opening reaches 51% (i.e., the difference from 52% is 1%, which is less than ε1=3%), the controller switches to a lower gain and initiates the smoothing function. The valve opening speed begins to decrease exponentially, slowly approaching 52%.

[0039] Precision Locking Phase: This phase begins when the valve opening reaches 51.8% (difference 0.2% < ε2 = 0.8%). The controller temporarily ignores the opening command and instead fine-tunes the valve to ensure P_out is precisely stabilized at 5.0 MPa. Simultaneously, changes in ΔP are monitored. After approximately 3 seconds, ΔP remains essentially unchanged, the system determines that adjustment is complete, locks the current opening (possibly 51.9%), and returns to a stable state.

[0040] During the subsequent long period of stable operation, the controller executes a micro-perturbation self-learning strategy every 30 seconds: it sends a tiny pulse to the control signal for 0.1 seconds with an amplitude of 0.1% of the valve opening and observes the response of the valve stem force F. If a slight change in F is detected, it may indicate jamming, and the controller will log the event and issue a maintenance warning.

[0041] Using the above methods, the pressure regulating valve can respond quickly, smoothly, and accurately to changes in downstream pressure, and also has a self-monitoring function, which is significantly better than traditional control methods.

[0042] The above description is only used to illustrate the technical solution of the present invention and is not intended to limit it. Any other modifications or equivalent substitutions made by those skilled in the art to the technical solution of the present invention, as long as they do not depart from the spirit and scope of the technical solution of the present invention, should be covered within the scope of the claims of the present invention.

Claims

1. A method for intelligent valve regulation and health management based on multi-parameter fusion and state machine decision-making, characterized in that, This method is executed by the valve intelligent controller and includes the following steps: S1. Data Acquisition and Preprocessing Steps: Periodically and synchronously acquire valve inlet pressure, outlet pressure, fluid temperature, real-time opening degree, valve stem axial force, and actuator drive signal, and perform filtering, time delay compensation, and validity verification on the acquired raw data; S2. State identification step: Based on the preprocessed data, calculate the dynamic pressure difference and instantaneous theoretical flow rate; By comparing the deviation between the set flow rate and the instantaneous theoretical flow rate with at least two preset thresholds, a status flag representing the system's adjustment requirements is set; and by analyzing the relationship between the valve stem axial force, opening degree, and differential pressure, a health flag representing the valve's mechanical health status is set. S3. Multi-mode intelligent adjustment step: Based on the status flag and health flag, dynamically select and execute the corresponding control strategy from a set of operating modes containing at least the following modes: S31. Steady-state health monitoring mode: When in steady state, micro-disturbance detection signals are periodically injected into the actuator, and mechanical fault diagnosis and early warning are performed based on the response of valve stem axial force and opening degree; S32, Dynamic Adjustment Mode: When in a state requiring significant adjustment, a multi-stage trajectory tracking control strategy is executed, including a rapid approach phase, a smooth deceleration phase, and a precise locking phase.

2. The method according to claim 1, characterized in that, In the data acquisition and preprocessing steps, the actuator drive signal is the air chamber pressure of the pneumatic actuator or the motor current of the electric actuator; the validity verification includes removing data points that exceed the physical reasonable range and replacing them with the effective value of the previous cycle or the Kalman filter prediction value.

3. The method according to claim 1, characterized in that, The logic for setting the state flag bit in the state identification step includes: If the deviation is less than or equal to the first threshold and the duration exceeds it, it is determined to be in a steady state; If the deviation is greater than the second threshold, it is immediately determined to be an adjustment state that requires significant adjustment; If the deviation is between the first threshold and the second threshold, it is determined to be in a fine-tuning state; wherein the second threshold is greater than the first threshold.

4. The method according to claim 1, characterized in that, The aforementioned steady-state health monitoring mode specifically includes: Generate a micro-disturbance detection signal with an amplitude of less than 0.5% of the rated stroke and a duration of less than 300 milliseconds, with a period of not less than 30 seconds; Analyze the starting torque value of the valve stem axial force in response to the detection signal and the frictional force change curve during the movement process; If the starting torque value exceeds the baseline threshold established based on historical data, it is determined that there is a risk of jamming, and a jamming warning signal is generated. If the response of the opening shows an increased hysteresis or a non-monotonic change, it is determined that there is a risk of transmission clearance and a loosening warning signal is generated.

5. The method according to claim 1, characterized in that, The multi-stage trajectory tracking control strategy in the dynamic adjustment mode is specifically as follows: Rapid approach phase: The goal is to control the valve to move towards the target opening at the maximum permissible speed. Smooth deceleration section: When the difference between the real-time opening and the target opening enters the first range, it switches to tracking a preset smooth speed curve, so that the valve's movement speed smoothly decreases to near zero; Precise locking segment: When the difference enters a second range that is less than the first range, the control target is switched from tracking the opening to stabilizing the downstream pressure, and the fluid system is determined to be in equilibrium by monitoring the rate of change of the dynamic pressure difference; when the rate of change is continuously lower than the threshold and maintained for a predetermined time, the adjustment is determined to be completed and the mode is exited.

6. The method according to claim 5, characterized in that, In the precise locking section, an independently configured PID controller with parameters tuned to a small deviation range is used to stabilize and regulate the downstream pressure.

7. The method according to claim 1, characterized in that, The method further includes a step: under steady state, the stored valve flow characteristic curve is slowly and adaptively corrected using long-term recorded flow setpoint, real-time opening and dynamic differential pressure data to compensate for valve characteristic drift.

8. The method according to claim 1, characterized in that, The health flag is used to dynamically adjust the controller output intensity in other operating modes. When a jamming risk is detected, the amplitude or energy of the control output is temporarily increased.