Pressure self-adaptive regulation and control method and system for executing mechanism

Abstract

The invention discloses a pressure self-adaptive regulation and control method and system for an execution mechanism, and belongs to the technical field of intelligent control, and the method comprises the steps: deploying a pressure sensor at the output end of the execution mechanism, collecting system pressure data in real time, and carrying out the signal conditioning of the system pressure data, and obtaining standard system pressure data; setting a target pressure value according to process requirements, and calculating a pressure deviation value and a deviation change rate; constructing a pressure controller, performing fuzzy composite calculation and parameter dynamic adjustment on the pressure deviation value and the deviation change rate based on the pressure controller, and generating a pressure control update parameter; and generating a mechanism pressure control signal according to the pressure control updating parameter, and driving the execution mechanism to perform pressure self-adaptive regulation and compensation closed-loop control based on the mechanism pressure control signal. The technical problems that in the prior art, pressure control depends on fixed parameters and is difficult to adapt to non-linear time-varying working conditions, so that control precision is low, response is slow, and the anti-interference capacity is weak are solved.

Description

Technical Field

[0001] This invention relates to the field of intelligent control technology, and more specifically to a method and system for adaptive pressure control of actuators. Background Technology

[0002] In fields such as industrial automation and intelligent manufacturing, actuators, as core power output components, are widely used in hydraulic and pneumatic control systems and precision machining equipment. Their pressure control accuracy directly determines the stability of equipment operation, work efficiency, and product processing quality. With the advancement of Industry 4.0, high-end manufacturing places higher demands on the pressure response speed, adaptive adjustment capability, and control stability of actuators. Pressure regulation technology has become one of the key links in enhancing the core competitiveness of equipment.

[0003] Existing pressure control methods for actuators mostly employ fixed-parameter control algorithms, which are susceptible to noise interference during signal processing and suffer from problems such as signal distortion and unstable switching transitions due to their simplistic gain adjustment. Furthermore, traditional controller rules are highly subjective in design, making it difficult to adapt to dynamic pressure changes under complex operating conditions. This results in insufficient control accuracy and weak anti-interference capabilities, failing to meet the demands of high-precision operations and hindering the application expansion of actuators in high-end manufacturing. Therefore, there is an urgent need for a pressure adaptive control method that integrates high-fidelity signal sensing, data-driven intelligent decision-making, and composite precise compensation to comprehensively improve the control performance of actuators in complex industrial environments. Summary of the Invention

[0004] This application provides a pressure adaptive control method and system for actuators, aiming to solve the technical problems in the prior art where pressure control relies on fixed parameters and is difficult to adapt to nonlinear time-varying working conditions, resulting in low control accuracy, slow response and weak anti-interference ability.

[0005] In view of the above problems, this application provides a method and system for adaptive pressure control of actuators.

[0006] The first aspect disclosed in this application provides a pressure adaptive control method for an actuator. The method includes: deploying a high-precision pressure sensor at the output end of the actuator; acquiring system pressure data in real time through the high-precision pressure sensor; performing signal conditioning on the system pressure data to obtain standard system pressure data; setting a target pressure value according to process requirements; calculating the pressure deviation and deviation change rate between the standard system pressure data and the target pressure value; constructing a pressure controller; performing fuzzy composite calculation and dynamic parameter adjustment on the pressure deviation and deviation change rate based on the pressure controller to generate pressure control update parameters; generating an actuator pressure control signal based on the pressure control update parameters; and driving the actuator to perform pressure adaptive control and compensation closed-loop control based on the actuator pressure control signal.

[0007] Another aspect of this application discloses a pressure adaptive control system for an actuator. This system includes: a pressure data acquisition module for deploying a high-precision pressure sensor at the output end of the actuator, acquiring system pressure data in real time through the high-precision pressure sensor, and performing signal conditioning on the system pressure data to obtain standard system pressure data; a deviation data acquisition module for setting a target pressure value according to process requirements, and calculating the pressure deviation value and deviation change rate between the standard system pressure data and the target pressure value; an update parameter acquisition module for constructing a pressure controller, performing fuzzy composite calculation and dynamic parameter adjustment on the pressure deviation value and deviation change rate based on the pressure controller, and generating pressure control update parameters; and a control signal generation module for generating an actuator pressure control signal based on the pressure control update parameters, and driving the actuator to perform pressure adaptive control and compensation closed-loop control based on the actuator pressure control signal.

[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages: By employing a complete technical solution that utilizes high-precision sensors for real-time sensing and signal conditioning, intelligent decision-making based on fuzzy composite calculation and dynamic parameter adjustment, and drives the actuator to form a compensation closed loop, this solution solves the technical problems of existing pressure control technologies, which rely on fixed parameters, are difficult to adapt to nonlinear time-varying conditions, and thus suffer from low control accuracy, slow response, and weak anti-interference capabilities. It achieves the technical effect of enabling the pressure system to autonomously adapt to complex changes and realize high-precision, robust, and intelligent control.

[0009] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, specific embodiments of this application are given below. Attached Figure Description

[0010] Figure 1 A flowchart illustrating a pressure adaptive control method for an actuator is provided for embodiments of this application.

[0011] Figure 2 A schematic diagram of a pressure adaptive control system for an actuator is provided for embodiments of this application.

[0012] Figure labeling: Pressure data acquisition module 11, Deviation data acquisition module 12, Update parameter acquisition module 13, Control signal generation module 14. Detailed Implementation

[0013] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0014] The overall concept of the technical solution provided in this application is as follows: This application provides a method and system for adaptive pressure control of an actuator. First, the actual pressure is acquired through high-precision sensing and signal chain. Then, the deviation from the target value and its changing trend are calculated and used as input. This input is fed into a fuzzy controller with parameter self-adjustment capabilities for intelligent decision-making, generating optimized control commands to ultimately drive the actuator. Simultaneously, a compensation closed loop is introduced to accurately correct residual errors, thereby achieving adaptive high-precision pressure control.

[0015] After introducing the basic principles of this application, various non-limiting embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0016] Example 1, as Figure 1 As shown in the embodiments of this application, a pressure adaptive control method for an actuator is provided, the method comprising: Step S100: Deploy a high-precision pressure sensor at the output end of the actuator, collect system pressure data in real time through the high-precision pressure sensor, and perform signal conditioning on the system pressure data to obtain standard system pressure data.

[0017] Specifically, the output end of an actuator refers to the physical location where pressure is applied or transmitted. Examples include the piston rod end of a hydraulic cylinder, the pressure head of a stamping press, and the nozzle of an injection molding machine, where pressure directly acts on the controlled object. A high-precision pressure sensor is a measuring device that converts physical pressure signals into high-precision electrical signals. Its core characteristics include high resolution, low nonlinearity error, and good long-term stability. Examples include sensors employing strain gauges or microelectromechanical systems (MEMS) technology. System pressure data is the raw electrical signal directly output by the pressure sensor, typically a weak voltage or current signal, directly reflecting the instantaneous pressure at the actuator's output end, but may contain noise, drift, or be affected by temperature.

[0018] Specifically, a high-precision pressure sensor is directly installed at the output of the pressure-acting actuator to collect raw system pressure data reflecting the actual system load in real time. This raw signal is transmitted to a signal conditioner for a series of signal conditioning processes: first, a dynamic gain amplifier automatically adjusts the amplification factor according to the signal magnitude to ensure that small signals are not distorted and large signals are not saturated; then, a low-pass filter, such as a Butterworth filter, filters out high-frequency electrical noise and mechanical vibration interference; next, temperature and nonlinearity compensation is performed to correct the sensor's own errors; finally, a high-resolution analog-to-digital converter converts the analog signal into a digital signal. Through this process, stable, accurate, and uniformly formatted standard system pressure data is output, providing a reliable basis for subsequent intelligent decision-making.

[0019] This step, through the deployment of high-precision sensors and specialized signal conditioning, ensures the authenticity, accuracy, and anti-interference capability of the pressure feedback signal. Transforming the raw physical signal, which is susceptible to contamination and contains noise and errors, into stable and reliable standardized data lays a precise sensing foundation for the entire adaptive control system, serving as a prerequisite for subsequent high-precision intelligent control.

[0020] Step S200: Set the target pressure value according to the process requirements, and calculate the pressure deviation and deviation change rate between the standard system pressure data and the target pressure value.

[0021] Specifically, process requirements refer to the specific, quantifiable requirements of a particular production process on the output pressure of the actuator. This is typically manifested as one or more key process parameters. For example, maintaining a constant cavity pressure during the "holding pressure stage" of injection molding, or achieving a precise pressing force curve that is initially fast and then slows down during press assembly. The target pressure value is the pressure value or pressure curve that the actuator's output is expected to reach and maintain, set according to the process requirements. It can be a constant value or a time-varying program, i.e., a set trajectory. For example, in vulcanizing machine operation, the target pressure value might be a complex curve that first rises to 15 MPa and then holds pressure in three stages. The pressure deviation value refers to the instantaneous difference between the standard system pressure data and the target pressure value at any given sampling moment. It is the most fundamental driving force of the control system, indicating "how much" the pressure differs. The rate of change of deviation is the speed at which the pressure deviation value changes over time, i.e., its derivative. It reflects whether the pressure deviation is increasing or decreasing, and how fast it is changing, indicating "how quickly the pressure is deteriorating."

[0022] Specifically, a target pressure value (P_target) is set according to process requirements, and this value is stored in the controller's registers or variables. Simultaneously, processed standard system pressure data (P_actual) from the upstream signal chain is read in real-time at a fixed sampling period. The pressure deviation value is then calculated. Further calculations are made to determine the rate of change of deviation by calculating the difference between the deviation values ​​of the current period and the previous period, and then dividing by the sampling period. To achieve this, that is In practical applications, to ensure the calculated rate of change is smooth and noise-resistant, a first-order low-pass filter is often applied to the difference result. Thus, we obtain two key decision inputs describing the relationship between the current system state and the target: the magnitude of the error, i.e., the pressure deviation value, and the trend of the error, i.e., the deviation rate of change.

[0023] This step transforms the static process target and the dynamic real-time system state into a pair of quantifiable characteristic quantities with clear control guidance through a precise, fast, and noise-resistant difference / differential calculation process. The pressure deviation value directly answers the question of "whether adjustment is needed" and is the foundation of proportional control; the deviation change rate proactively reveals "the direction and magnitude of the system's inertia" and is key to differential control and predictive decision-making. The accurate calculation of these two values ​​together provides the subsequent intelligent controller with a complete and quantitative "state profile" describing the system's dynamic behavior, serving as the fundamental logical input and decision-making basis for the control system to perform predictive and adaptive adjustments. Its calculation accuracy and real-time performance directly determine the response speed and control precision of the entire control system.

[0024] Step S300: Construct a pressure controller, and based on the pressure controller, perform fuzzy composite calculation and dynamic parameter adjustment on the pressure deviation value and deviation change rate to generate pressure control update parameters.

[0025] Specifically, the pressure controller is a data-driven, fuzzy logic-based composite controller. Its inputs are the pressure deviation value and the rate of change of the deviation, and its outputs are control parameters that drive the actuators, such as valve opening command increments and motor current setpoints. Fuzzy composite calculation refers to the core computational process within the controller. Using fuzzy mathematics theory, precise input quantities (deviation, rate of change) are converted into fuzzy linguistic variables, such as "negative large" and "positive small." Then, based on a set of predefined "IF-THEN" fuzzy rules, inference is performed, and the resulting fuzzy output is defuzzified into precise control quantities. The term "composite" emphasizes that its output is a joint decision result that integrates the magnitude and trend of the deviation.

[0026] Specifically, historical pressure control datasets of the actuator under various operating conditions are collected, fuzzy sets of input / output variables are defined, and appropriate membership functions are selected to fuzzify the precise quantities. Next, data mining methods such as K-means clustering are used to automatically extract and optimize a fuzzy rule base describing the mapping relationship between input and output from the historical data. The controller model is verified and optimized through simulation or actual testing. During online operation, based on this pressure controller, fuzzy composite calculations are performed on the real-time calculated precise pressure deviation value and deviation change rate in each control cycle: first, fuzzification is performed, then inference is conducted through the fuzzy rule base to obtain a fuzzy output, and finally, defuzzification is performed to obtain a basic control quantity. Simultaneously, the controller parameters are dynamically adjusted according to the current deviation and change rate patterns, generating new pressure control update parameters, which are then passed to the execution stage.

[0027] This step, by introducing data-driven fuzzy logic and an online parameter adaptation mechanism, upgrades the traditional controller, which relies on precise models and fixed parameters, into an intelligent agent capable of understanding and applying expert experience, learning from data, and self-adjusting online. Fuzzy composite computing enables the controller to make reasonable and smooth decisions, much like a human operator, in complex and imprecise states where deviations are large and rapidly decreasing, and it excels at handling nonlinearity and uncertainty.

[0028] Step S400: Generate a mechanism pressure control signal based on the pressure control update parameters, and drive the actuator to perform pressure adaptive regulation and compensation closed-loop control based on the mechanism pressure control signal.

[0029] Specifically, the mechanism pressure control signal refers to the physical signal that can be directly recognized and executed by the hardware drive unit of the actuator. It is a standard industrial signal that matches the physical meaning of the pressure control update parameters. Pressure adaptive regulation refers to the pressure adjustment process itself performed by the actuator under the drive of the mechanism pressure control signal. Compensation closed-loop control specifically refers to the auxiliary control loop introduced in this scheme, in addition to the main fuzzy control loop, for accurately correcting residual deviations. It is typically composed of a PID controller, which rapidly and accurately fine-tunes the main control signal based on the final pressure control feedback parameter, i.e., the deviation between the actual pressure and the target pressure, forming an inner closed loop to ensure the system's steady-state accuracy and disturbance rejection.

[0030] Specifically, a pressure control signal is generated based on the pressure control update parameters. This signal drives the actuator; that is, the signal is sent to a proportional valve amplifier or servo driver to move the proportional valve spool or rotate the servo motor, thereby changing the flow rate of the hydraulic system or the position of the actuator to achieve adaptive pressure control. Simultaneously, a compensation closed-loop control is executed. A high-precision pressure sensor continuously detects the actual pressure and compares it with the target value to obtain the real-time deviation. The PID controller calculates this deviation to generate a compensation signal, which is superimposed on the output signal of the main fuzzy controller to form the pressure control update parameters, correcting dynamic or steady-state errors in real time.

[0031] This step ensures both lossless and precise issuance of control commands and high control performance. By accurately converting digitized update parameters into physical signals recognizable by the actuators, it ensures that the intentions of intelligent decisions are faithfully executed.

[0032] Furthermore, obtaining standard system pressure data includes: deploying a signal conditioner, connecting the input end of the signal conditioner to the high-precision pressure sensor, and connecting the output end to a data preprocessing module; dynamically adjusting the gain of the system pressure data based on the signal conditioner to obtain usable system pressure data; transmitting the usable system pressure data to the data preprocessing module, wherein the data preprocessing module includes a data filtering unit and a normalization processing unit; and using the data preprocessing module to perform signal conditioning on the usable system pressure data to obtain standard system pressure data.

[0033] Specifically, a signal conditioner is a hardware circuit or programmable analog front-end module whose core function is to perform preliminary, analog-circuit-based conditioning, particularly amplification, of the raw analog electrical signal output from a sensor. Directly connected to the sensor, it is a crucial interface before the signal enters the digital system. Dynamic gain adjustment means that the signal conditioner's amplification factor (gain) is not fixed but automatically and in real-time adjusted according to the amplitude of the input signal. This ensures that both weak pressure signals and large-amplitude signals can be adjusted to the optimal range of the subsequent analog-to-digital converter (ADC) to fully utilize the ADC's resolution and prevent small signals from being drowned out by noise or large signals from being clipped.

[0034] Specifically, at the hardware level, a signal conditioner is deployed. For example, a circuit board consisting of a programmable gain amplifier (PGA) and a precision operational amplifier is directly connected to the output cable of a high-precision pressure sensor, while its output is connected to the analog input channel of a data acquisition card. The core task of the signal conditioner is dynamic gain adjustment: its internal PGA automatically selects an optimal amplification factor based on the amplitude of the raw signal sampled in real time, thereby amplifying and normalizing the system pressure data into usable system pressure data. Subsequently, this analog signal is converted from analog to digital by the data acquisition card and transmitted to the data preprocessing module via a communication interface such as USB or Ethernet. In this module, the data filtering unit first smooths and denoises the digital signal. Then, the normalization processing unit, based on the known sensor range and current gain setting, uses a linear transformation formula to obtain the standard system pressure data from the filtered digital voltage.

[0035] This step, through the construction of a two-tier processing architecture of analog dynamic gain adjustment and digital filtering and normalization, elevates the raw pressure signal from "usable" to "standard, clean, and directly applicable for decision-making." Dynamic gain adjustment at the analog front end maximizes the system's dynamic range and signal-to-noise ratio, forming the hardware foundation for ensuring measurement accuracy. Meanwhile, filtering and normalization at the digital end eliminate interference and unify dimensions, providing subsequent controllers with high-fidelity, unambiguous, and uniformly formatted pressure feedback values. This clearly defined division of labor and the integration of hardware and software significantly improves the accuracy, stability, and reliability of the entire system's sensing capabilities.

[0036] Furthermore, obtaining usable system pressure data includes: acquiring a pressure signal distribution threshold based on the pressure application requirements of the actuator; dividing the pressure signal distribution threshold into intervals and analyzing the gain strategy to construct a pressure signal partition gain strategy; the signal conditioner performing dynamic gain analysis on the system pressure data based on the pressure signal partition gain strategy to obtain a signal dynamic gain coefficient; and adjusting the gain amplification of the system pressure data based on the signal dynamic gain coefficient to obtain usable system pressure data.

[0037] Specifically, pressure application requirements refer to the specific performance indicators and requirements that a particular process places on the pressure control of the actuator, including but not limited to the target pressure range, control accuracy, response speed, and pressure change rate. For example, precision press fitting requires micro-force control of 0-10KN, while heavy-duty stamping may involve a wide range of force control of 0-1000KN.

[0038] Specifically, pressure signal distribution thresholds are obtained based on the pressure application requirements of the actuator. These thresholds are then divided into zones and analyzed using gain strategies: for example, 0-5N is designated as a "precision control zone," configured with high gain, such as 1000x, to amplify weak signals; 5-20N is designated as an "overload protection zone," configured with medium gain, such as 200x; and above 20N is designated as a "safe zone," configured with low gain, such as 50x. This constructs a pressure signal zone gain strategy. In real-time operation, the signal conditioner, i.e., a programmable gain amplifier controlled by the MCU, continuously performs dynamic gain analysis on the system pressure data. The MCU monitors the original signal amplitude through a low-speed ADC, determines its zone, and retrieves the corresponding dynamic gain coefficient from the strategy table in real time. This result is then sent to the PGA chip via a digital interface, such as SPI. The PGA then adjusts the gain of the system pressure data based on this dynamic gain coefficient, outputting stable and usable system pressure data. This ensures that signals at any pressure level can enter the subsequent ADC for digitization with optimal resolution.

[0039] This step introduces a pressure signal partitioning gain strategy based on prior knowledge, achieving intelligent and adaptive signal amplification. Its core effect is to maximize the effective resolution of analog-to-digital conversion across the entire pressure range, avoiding the contradiction of "insufficient resolution for small signals and easy saturation for large signals" in traditional fixed-gain methods during wide dynamic range measurements.

[0040] Furthermore, before performing gain amplification adjustment on the system pressure data based on the signal dynamic gain coefficient, the process includes: when the signal conditioner is in the gain switching transition phase, monitoring the switching time of the signal conditioner to obtain the signal switching delay duration; calculating the compensation coefficient for the signal switching delay duration according to the amplitude continuity compensation target to determine the signal switching compensation coefficient; performing gain compensation on the system pressure data based on the signal switching compensation coefficient, and performing gain amplification adjustment on the system pressure data after the compensation transition based on the signal dynamic gain coefficient.

[0041] Specifically, the gain switching transition phase refers to the brief period of instability experienced by the signal conditioner after receiving a gain switching command, as it transitions from its current stable gain value to the target stable gain value. During this time, the amplifier's internal circuitry is reconfiguring, and the output signal may exhibit glitches, overshoot, or brief periods of inactivity. The signal switching compensation coefficient is a time-varying parameter used to temporarily correct the input or intermediate signal during the transition phase. It is typically a gain or bias value that varies with time.

[0042] Specifically, when pressure changes necessitate a gain range switch, the signal conditioner's controller issues a switching command. A timer is then started to monitor the switching time, measuring the signal switching delay from the command's issuance until the output signal voltage fully stabilizes within the linear region corresponding to the new gain. Based on this delay and the two gain values ​​before and after the change, compensation coefficients are calculated according to the amplitude continuity compensation objective. A common method is the "previous value hold and ramp transition" algorithm: within 50 microseconds after the switch begins, an attenuation coefficient linearly changing from 1.0 to 0.0 is calculated, along with an enhancement coefficient linearly changing from 0.0 to 1.0. Gain compensation is then performed based on the signal switching compensation coefficients: within 50 microseconds, the original signal is weighted and mixed using the old gain and the calculated attenuation coefficient, and the new gain and the calculated enhancement coefficient, respectively, or an additional analog multiplier is used to smooth the signal through interpolation. The system pressure data after the compensation transition is continuous in amplitude. Finally, standard gain amplification adjustment is applied based on a predetermined new signal dynamic gain coefficient, resulting in smooth, abrupt, usable system pressure data.

[0043] The specific implementation steps of the "previous value hold and ramp transition" algorithm are as follows: Within the accurately measured signal switching delay time T, periodically perform high-frequency operation, typically much higher than the signal bandwidth. Discretization processing; in each processing cycle k (k ranges from 0 to N, where The algorithm calculates two time-varying weighting coefficients: the attenuation coefficient of the old gain channel. The coefficient decreases linearly from 1 to 0; the enhancement coefficient of the new gain channel. This coefficient increases linearly from 0 to 1; simultaneously, the system buffers the input original system pressure data P_raw, delaying it by one cycle to align the timing. In each cycle k, the final compensated output signal P_compensated(k) is calculated by the following formula:

[0044] Where G_old and G_new are the dynamic gain coefficients of the signal before and after the switching, respectively. This process is equivalent to smoothly and linearly transitioning from a historical signal amplified entirely by the old gain to a real-time signal amplified entirely by the new gain within a time period T, thus mathematically and physically ensuring the continuity of the output signal amplitude change.

[0045] This step, by introducing transient monitoring and active compensation mechanisms for gain switching, shifts the system's focus from "whether gain can be switched" to "how to switch gain without disturbance," fundamentally eliminating additional interference introduced into the feedback loop due to the regulator's own actions. This ensures that the pressure feedback signal maintains its historical continuity and instantaneous accuracy when gain adjustments are caused by changes in operating conditions, preventing the misinterpretation of circuit switching noise as pressure surges and its transmission to the controller. Consequently, it significantly improves the stability, control quality, and reliability of the entire control system under wide dynamic range and rapidly changing operating conditions. This is an indispensable detail for achieving high-precision adaptive control.

[0046] Furthermore, constructing a pressure controller includes: collecting historical pressure control datasets of the actuator; preprocessing the historical pressure control datasets to obtain a standard pressure control dataset; defining input variables and output variables, wherein the input variables include pressure deviation data and deviation change rate data, and the output variables include pressure control data; selecting a membership function, and using the membership function to perform variable fuzzification identification on the standard pressure control dataset based on the input and output variables to obtain a pressure control fuzzy sample set; designing a fuzzy rule base, and performing fuzzy inference control training, testing, verification, and optimization on the pressure control fuzzy sample set based on the designed fuzzy rule base to construct the pressure controller.

[0047] Specifically, a pressure history control dataset refers to multi-dimensional time-series data collected and recorded during system debugging or historical operation, reflecting the pressure control process. It typically includes "pressure deviation," "deviation change rate," and the applied and verified effective "control outputs," such as valve position and current. A membership function describes the degree to which a precise numerical value belongs to a fuzzy linguistic concept; its function value is between [0, 1], also known as the membership degree. Common shapes include triangles, trapezoids, and Gaussian shapes. Fuzzy identification refers to the process of converting the precise input / output value of each record in a standard pressure control dataset into a set of linguistic labels and their membership degrees using membership functions. A fuzzy rule base refers to a set of conditional statements in the form of "IF-THEN." Each rule describes the mapping relationship from fuzzy input states to fuzzy output actions, for example: "IF deviation is 'positive large' AND deviation change rate is 'negative small', THEN control output is 'positive medium'."

[0048] Specifically, the operating data of the original PID controller under stable conditions is recorded to obtain a historical pressure control dataset. This dataset is preprocessed to remove outliers, smooth noise, and normalize to a unified dimension, thus obtaining a standard pressure control dataset. Input and output variables are defined, that is, reasonable universes of discourse and basic fuzzy subsets are set for deviation, rate of change, and control output, such as "negative large NB", "negative medium NM", "zero ZO", "positive medium PM", and "positive large PB". Membership functions are selected for these subsets. For example, a highly sensitive triangular membership function is selected for deviation and rate of change, and a smooth Gaussian membership function is selected for the output. These membership functions are used to fuzzify the variables in the standard pressure control dataset, transforming each historical data entry into a fuzzy language description, forming a pressure control fuzzy sample set.

[0049] A fuzzy rule base is designed. Specifically, the K-means clustering algorithm is used to perform cluster analysis on the fuzzy sample set, automatically extracting typical input-output mapping patterns from the data. An initial fuzzy rule is generated, represented by the center of each cluster. After redundancy merging and optimization, a preliminary fuzzy rule base is formed. Based on this rule base, another set of historical data or simulated operating conditions are used to train and test the controller for fuzzy inference control, evaluating its step response and other performance characteristics. Based on the evaluation results, the membership function parameters can be adjusted or the rule base can be pruned. After multiple iterative optimizations, a pressure controller with satisfactory performance and stable reliability is finally constructed.

[0050] This step enables the objectification and systematization of the fuzzy controller. It transforms fuzzy control from relying on expert subjective experience into an engineering technology driven by historical data and capable of automatic optimization through algorithms. The constructed pressure controller's rule base is directly derived from the controlled object, i.e., the actuator's own optimal historical operating data. Therefore, it can more realistically and completely characterize the system's dynamic characteristics and control strategies, overcoming the shortcomings of incomplete and inaccurate rules summarized manually. Through testing, verification, and optimization, it is ensured that the controller possesses good stability and dynamic performance before being put into actual operation.

[0051] Furthermore, a fuzzy rule base is designed, including: constructing a three-dimensional variable dataset for pressure control based on the fuzzy sample set for pressure control; performing K-means clustering on the three-dimensional variable dataset for pressure control to obtain multiple pressure control pattern data clusters; designing a fuzzy rule form; and performing rule extraction and redundancy merging optimization based on the cluster centers of the multiple pressure control pattern data clusters according to the fuzzy rule form to construct the fuzzy rule base.

[0052] Specifically, the pressure control three-dimensional variable dataset is a structured numerical collection where each data point consists of three precise numerical dimensions. These three dimensions correspond to the two input variables (pressure deviation and deviation change rate) and one output variable (pressure control data) of the fuzzy controller. Essentially, it involves reorganizing the most representative precise values ​​of the standard pressure control dataset before fuzzification or the fuzzified samples—such as the center value of the domain corresponding to the linguistic value with the highest membership degree—into a three-dimensional point set for numerical clustering analysis. Rule extraction refers to starting from the cluster center of each pressure control pattern data cluster, a typical representative point, and transforming its three-dimensional precise coordinate values ​​through "reverse fuzzification"—that is, finding the fuzzy linguistic label with the highest membership degree for each precise value—into a fuzzy rule that conforms to the form of a fuzzy rule and is described in natural language.

[0053] Specifically, based on the fuzzy sample set of pressure control, a three-dimensional variable dataset for pressure control is constructed. From each fuzzy sample, the most representative precise values ​​of the three variables—pressure deviation, deviation change rate, and pressure control data—are extracted. For example, the universe center value corresponding to the fuzzy subset with the highest membership degree of the variable across all fuzzy sets is extracted, thus mapping each sample to a three-dimensional spatial point. All points constitute the dataset. K-means clustering is performed on the three-dimensional variable dataset of pressure control, automatically dividing the data points into K clusters to obtain multiple pressure control pattern data clusters and their respective cluster centers. Fuzzy rule forms are designed, such as the standard two-input single-output Mamdani form: "IF e is A and Δe is B, THEN u is C". Rule extraction is performed based on the cluster centers according to this form: for each cluster center, it is mapped to the fuzzy linguistic value closest to it, thus generating an initial rule. Redundancy merging and optimization are performed on all extracted initial rules, and duplicate and conflicting rules are deleted to construct a fuzzy rule base.

[0054] This step, by introducing K-means clustering, a data mining technique, automates, objectifies, and optimizes the design of the fuzzy rule base. It changes the traditional approach of relying on subjective, tedious, and potentially incomplete manual experience-based rule base summaries, instead automatically mining and summarizing the inherent control patterns and rules from the system's own historical best-performing data.

[0055] Furthermore, based on the designed fuzzy rule base, fuzzy inference control training, testing, verification, and optimization are performed on the pressure control fuzzy sample set to construct a pressure controller. This includes: training the pressure control fuzzy sample set based on the designed fuzzy rule base to obtain an initial pressure control model; testing, verifying, and evaluating the initial pressure control model to obtain initial model performance parameters, and selecting a model optimization strategy based on the initial model performance parameters; iteratively optimizing and updating the initial pressure control model based on the model optimization strategy to construct the pressure controller.

[0056] Specifically, the initial pressure control model is a prototype of a complete but not fully verified fuzzy logic system that combines a fuzzy rule base, defined input / output variables and their membership functions according to a specific fuzzy inference mechanism. This system can accept precise inputs and produce precise control outputs.

[0057] Specifically, fuzzy inference control training is performed on a fuzzy sample set for pressure control based on a designed fuzzy rule base, completed in a simulation environment. In this environment, the mathematical model of the actuator and the fuzzy inference mechanism built based on the rule base form a closed loop. The input sequence (deviation and rate of change) from the fuzzy sample set of pressure control is used as excitation, and simulation is run to observe the controller's output behavior, thus obtaining a runnable initial pressure control model. Next, this initial model is tested, verified, and evaluated: more stringent and comprehensive test scenarios are designed in the simulation platform, such as step changes in pressure setpoints of different amplitudes and the introduction of simulated load mutations. The model is run and response data is collected, and the initial model performance parameters are calculated, such as using MATLAB scripts to calculate overshoot and steady-state error. Based on these performance parameters, a model optimization strategy is selected: for example, if the response is too slow, the membership function of the deviation may be adjusted to increase control sensitivity; if there is continuous oscillation, the rule weights may need to be optimized or an integral term introduced. Finally, iterative optimization and updates are performed based on the selected model optimization strategy: for example, a genetic algorithm is used to search for a set of membership function parameters that optimize the overall performance index; or specific problematic rules are manually adjusted. After each round of optimization, testing and verification are performed again. This process is repeated until all performance parameters of the controller meet the design requirements.

[0058] This step, by introducing a systematic training, testing, verification, and iterative optimization process, improves the reliability, accuracy, and adaptability to complex operating conditions of the controller. It is a key quality assurance step to ensure that the entire adaptive control system can operate stably and efficiently in practical applications.

[0059] Furthermore, generating the mechanism pressure control signal includes: calculating a target pressure control signal based on the pressure control update parameters; obtaining the control actuator of the actuator; and converting the target pressure control signal into the mechanism pressure control signal of the control actuator.

[0060] Specifically, the target pressure control signal is an intermediate computational quantity, referring to a digital quantity calculated based on the meaning of the pressure control update parameters that can directly describe the desired physical execution action. The mechanism pressure control signal refers to a physical electrical signal with specific standards and formats that can be directly recognized and driven by the electrical interface of the controlled actuator. It serves as a "bridge" connecting the digital controller and the physical world, and commonly takes the forms of analog and digital signals.

[0061] Specifically, the target pressure control signal is calculated based on the pressure control update parameters. If the update parameters are Kp, Ki, and Kd of a PID controller, the PID function block in the controller will use these new parameters and the real-time deviation to calculate a new control output value U, which is the target pressure control signal. If the update parameter is already a control quantity U, it is directly used as the target signal. Next, the controller actuator is obtained, that is, the specific model and interface type of the final driven object are determined. The controller's output module is responsible for converting the target pressure control signal into a mechanism pressure control signal that the controller actuator can recognize. For example, if the controller actuator is an analog proportional valve, the controller's analog output module will linearly convert the digital quantity U (e.g., 0.0-1.0) into a 0-10V voltage signal and output it. If the controller actuator is a servo drive that supports bus communication, the controller's communication master module will encapsulate the digital quantity U into a specific EtherCAT message and send it to the drive. Finally, this mechanism pressure control signal is sent to the controller actuator to drive it to produce the corresponding physical action, thereby completing the final control of the actuator.

[0062] This step achieves high-fidelity, low-latency output of control commands. By clearly separating the two sub-processes of "target signal calculation" and "signal physical conversion," and by standardizing signal matching for different types of control actuators, it ensures that the controller can generate perfectly compatible drive commands regardless of the actuator being used.

[0063] Furthermore, driving the actuator to perform pressure adaptive regulation and compensation closed-loop control based on the mechanism pressure control signal includes: driving the actuator to perform pressure adaptive regulation and monitoring feedback based on the mechanism pressure control signal to obtain pressure control feedback parameters; and using a PID controller to perform deviation compensation closed-loop control on the mechanism pressure control signal based on the pressure control feedback parameters.

[0064] Specifically, pressure control feedback parameters refer to key real-time data obtained through monitoring feedback, specifically the instantaneous deviation between the actual pressure value and the target pressure value, and usually also include the rate of change of this deviation.

[0065] Specifically, the actuator is driven by a pressure control signal to perform adaptive pressure regulation. This signal is sent to a proportional valve amplifier or servo driver, driving the valve core to move or the motor to rotate, thereby physically changing the system pressure. Simultaneously, monitoring feedback is implemented. A high-precision sensor deployed at the actuator end detects the actual pressure in real time and sends the signal back to the controller via a data acquisition card. The controller compares the actual pressure with the target pressure to obtain pressure control feedback parameters, such as the instantaneous pressure deviation e. A PID controller is used to perform deviation compensation closed-loop control based on this feedback parameter. The deviation e is calculated using an independently configured and adjustable PID controller. For example, its integral (I) term continuously accumulates a small steady-state deviation to eliminate steady-state error, while the derivative (D) term responds to rapid changes in deviation by suppressing overshoot. The compensation output from the PID calculation, typically a small increment ΔU, is superimposed in real time on the main mechanism pressure control signal output by the fuzzy controller within the controller to generate a drive command. This new command is then output again to drive the actuator for more precise adjustments.

[0066] This step enhances the system's robustness and control accuracy in the face of internal parameter perturbations and external disturbances, and is the ultimate key to the entire method's ability to achieve high-performance adaptive and closed-loop control.

[0067] In summary, the pressure adaptive control method for actuators provided in this application has the following technical effects: 1. By deploying high-precision sensors to collect pressure data in real time, and combining fuzzy composite calculation and dynamic parameter adjustment to generate control signals, the actuator is driven to form a compensation closed loop. This claim constructs a complete "perception-decision-execution-compensation" adaptive control architecture, realizing high-precision and robust intelligent control of nonlinear and time-varying pressure systems.

[0068] 2. By monitoring the delay duration during gain switching and calculating a compensation coefficient, the signal is over-compensated. This claim effectively eliminates signal jumps or interruptions during gain switching, ensuring the instantaneous continuity and authenticity of the feedback signal and avoiding control malfunctions.

[0069] 3. By collecting historical data, defining variables, selecting membership functions to fuzzify the sample set, and designing a rule base based on this for training and verification, this claim provides a systematic, data-driven method for constructing a fuzzy controller. This overcomes the limitations of relying on subjective experience and lays the foundation for intelligent decision-making.

[0070] Example 2, based on the same inventive concept as the pressure adaptive control method for the actuator in the foregoing examples, such as... Figure 2 As shown in the figure, this application embodiment provides a pressure adaptive control system for an actuator, the system comprising: The pressure data acquisition module 11 is used to deploy a high-precision pressure sensor at the output end of the actuator, collect system pressure data in real time through the high-precision pressure sensor, and perform signal conditioning on the system pressure data to obtain standard system pressure data; the deviation data acquisition module 12 is used to set a target pressure value according to process requirements, and calculate the pressure deviation value and deviation change rate between the standard system pressure data and the target pressure value; the update parameter acquisition module 13 is used to construct a pressure controller, and perform fuzzy composite calculation and dynamic parameter adjustment on the pressure deviation value and deviation change rate based on the pressure controller to generate pressure control update parameters; the control signal generation module 14 is used to generate a mechanism pressure control signal according to the pressure control update parameters, and drive the actuator to perform pressure adaptive regulation and compensation closed-loop control based on the mechanism pressure control signal.

[0071] Furthermore, the pressure data acquisition module 11 is also used to perform the following steps: deploying a signal conditioner, connecting the input end of the signal conditioner to the high-precision pressure sensor, and connecting the output end to the data preprocessing module; dynamically adjusting the gain of the system pressure data based on the signal conditioner to obtain usable system pressure data; transmitting the usable system pressure data to the data preprocessing module, wherein the data preprocessing module includes a data filtering unit and a normalization processing unit; and using the data preprocessing module to perform signal conditioning on the usable system pressure data to obtain standard system pressure data.

[0072] Furthermore, the pressure data acquisition module 11 is also used to perform the following steps: acquiring a pressure signal distribution threshold according to the pressure application requirements of the actuator; dividing the pressure signal distribution threshold into intervals and analyzing the gain strategy to construct a pressure signal partition gain strategy; the signal conditioner performs dynamic gain analysis on the system pressure data based on the pressure signal partition gain strategy to obtain a signal dynamic gain coefficient; and performs gain amplification adjustment on the system pressure data based on the signal dynamic gain coefficient to obtain usable system pressure data.

[0073] Furthermore, the pressure data acquisition module 11 is also used to perform the following steps: when the signal conditioner is in the gain switching transition phase, monitor the switching time of the signal conditioner to obtain the signal switching delay duration; calculate the compensation coefficient of the signal switching delay duration according to the amplitude continuity compensation target to determine the signal switching compensation coefficient; perform gain compensation on the system pressure data based on the signal switching compensation coefficient, and perform gain amplification adjustment on the system pressure data after the compensation transition based on the signal dynamic gain coefficient.

[0074] Furthermore, the update parameter acquisition module 13 is also used to perform the following steps: collecting the historical pressure control dataset of the actuator, preprocessing the historical pressure control dataset to obtain a standard pressure control dataset; defining input variables and output variables, wherein the input variables include pressure deviation data and deviation change rate data, and the output variables include pressure control data; selecting a membership function, using the membership function to perform variable fuzzification identification on the standard pressure control dataset based on the input variables and output variables to obtain a pressure control fuzzy sample set; designing a fuzzy rule base, and performing fuzzy inference control training, testing, verification, and optimization on the pressure control fuzzy sample set based on the designed fuzzy rule base to construct a pressure controller.

[0075] Furthermore, the update parameter acquisition module 13 is also used to perform the following steps: construct a pressure control three-dimensional variable dataset based on the pressure control fuzzy sample set; perform K-means clustering on the pressure control three-dimensional variable dataset to obtain multiple pressure control pattern data clusters; design a fuzzy rule form, and perform rule extraction and redundancy merging optimization based on the cluster centers of the multiple pressure control pattern data clusters according to the fuzzy rule form to construct a fuzzy rule library.

[0076] Furthermore, the update parameter acquisition module 13 is also used to perform the following steps: perform fuzzy inference control training on the pressure control fuzzy sample set based on the design fuzzy rule base to obtain an initial pressure control model; test, verify and evaluate the initial pressure control model to obtain initial model performance parameters, and select a model optimization strategy based on the initial model performance parameters; iteratively optimize and update the initial pressure control model based on the model optimization strategy to construct a pressure controller.

[0077] Furthermore, the control signal generation module 14 is also used to perform the following steps: calculate the target pressure control signal according to the pressure control update parameters; obtain the control actuator of the actuator, and convert the target pressure control signal into the mechanism pressure control signal of the control actuator.

[0078] Furthermore, the control signal generation module 14 is also used to perform the following steps: drive the actuator to perform pressure adaptive regulation and monitoring feedback based on the mechanism pressure control signal to obtain pressure control feedback parameters; and use a PID controller to perform deviation compensation closed-loop control on the mechanism pressure control signal based on the pressure control feedback parameters.

[0079] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A pressure adaptive control method for actuators, characterized in that, The method includes: A high-precision pressure sensor is deployed at the output end of the actuator. The system pressure data is collected in real time through the high-precision pressure sensor, and the system pressure data is conditioned to obtain standard system pressure data. Set a target pressure value according to process requirements, and calculate the pressure deviation and deviation rate between the standard system pressure data and the target pressure value; A pressure controller is constructed, and based on the pressure controller, fuzzy composite calculation and dynamic parameter adjustment are performed on the pressure deviation value and the rate of change of deviation to generate pressure control update parameters. Based on the pressure control update parameters, a mechanism pressure control signal is generated, and the actuator is driven to perform pressure adaptive regulation and compensation closed-loop control based on the mechanism pressure control signal.

2. The pressure adaptive regulation method for actuators as claimed in claim 1, wherein, Obtain standard system pressure data, including: A signal conditioner is installed, with its input end connected to the high-precision pressure sensor and its output end connected to the data preprocessing module. Based on the signal conditioner, the system pressure data is dynamically gained and adjusted to obtain usable system pressure data; The available system pressure data is communicated to the data preprocessing module, wherein the data preprocessing module includes a data filtering unit and a normalization processing unit; The available system pressure data is processed by the data preprocessing module to obtain standard system pressure data.

3. The pressure adaptive control method for an actuator as described in claim 2, characterized in that, Obtain available system stress data, including: Based on the pressure application requirements of the actuator, obtain the pressure signal distribution threshold; The pressure signal distribution threshold is divided into intervals and the gain strategy is analyzed to construct a pressure signal partition gain strategy; The signal conditioner performs dynamic gain analysis on the system pressure data based on the pressure signal partitioning gain strategy to obtain the signal dynamic gain coefficient. Based on the signal dynamic gain coefficient, the system pressure data is amplified and adjusted to obtain usable system pressure data.

4. The pressure adaptive regulation method for actuators as claimed in claim 3, wherein, Before adjusting the gain of the system pressure data based on the signal dynamic gain coefficient, the following steps are included: When the signal conditioner is in the gain switching transition phase, the switching time of the signal conditioner is monitored to obtain the signal switching delay duration; The signal switching delay duration is calculated according to the amplitude continuity compensation target to determine the signal switching compensation coefficient; Gain compensation is performed on the system pressure data based on the signal switching compensation coefficient, and gain amplification and adjustment are performed on the system pressure data after compensation transition based on the signal dynamic gain coefficient.

5. The method for pressure adaptive regulation of actuators as recited in claim 1, wherein, Constructing a pressure controller includes: Collect the historical pressure control dataset of the actuator, preprocess the historical pressure control dataset to obtain the standard pressure control dataset; Define input variables and output variables, wherein the input variables include pressure deviation data and deviation change rate data, and the output variables include pressure control data; Select a membership function, and use the membership function to perform variable fuzzification identification on the standard pressure control dataset based on the input and output variables to obtain a pressure control fuzzy sample set; Design a fuzzy rule base, and based on the designed fuzzy rule base, perform fuzzy inference control training, testing, verification and optimization on the pressure control fuzzy sample set to construct a pressure controller.

6. The pressure adaptive regulation method for actuators as claimed in claim 5, wherein, Design a fuzzy rule base, including: Based on the aforementioned pressure control fuzzy sample set, construct a pressure control three-dimensional variable dataset; K-means clustering was performed on the pressure control three-dimensional variable dataset to obtain multiple pressure control pattern data clusters; Design a fuzzy rule format, and perform rule extraction and redundancy merging optimization based on the cluster centers of the multiple pressure control mode data clusters according to the fuzzy rule format to construct a fuzzy rule library.

7. The pressure adaptive regulation method for actuators as claimed in claim 5, wherein, Based on the designed fuzzy rule base, fuzzy inference control training, testing, verification, and optimization are performed on the pressure control fuzzy sample set to construct a pressure controller, including: Based on the designed fuzzy rule base, fuzzy inference control training is performed on the pressure control fuzzy sample set to obtain an initial pressure control model; The initial pressure control model is tested, verified, and evaluated to obtain initial model performance parameters, and a model optimization strategy is selected based on the initial model performance parameters. The initial pressure control model is iteratively optimized and updated based on the model optimization strategy to construct a pressure controller.

8. The method for pressure adaptive regulation of actuators of claim 1, wherein, Generates pressure control signals for the mechanism, including: Calculate the target pressure control signal based on the pressure control update parameters; Obtain the control actuator of the actuator and convert the target pressure control signal into the mechanism pressure control signal of the control actuator.

9. The method for pressure adaptive regulation of actuators of claim 1, wherein, Based on the pressure control signal of the mechanism, the actuator is driven to perform pressure adaptive regulation and compensation closed-loop control, including: Based on the pressure control signal of the mechanism, the actuator is driven to perform pressure adaptive regulation and monitoring feedback to obtain pressure control feedback parameters. A PID controller is used to perform deviation compensation closed-loop control on the pressure control signal of the mechanism based on the pressure control feedback parameters.

10. A pressure adaptive regulation system for actuators, characterized by, The system is used to perform the pressure adaptive control method for an actuator according to any one of claims 1 to 9, the system comprising: The pressure data acquisition module is used to deploy a high-precision pressure sensor at the output end of the actuator, and to collect system pressure data in real time through the high-precision pressure sensor, and to perform signal conditioning on the system pressure data to obtain standard system pressure data. The deviation data acquisition module is used to set a target pressure value according to process requirements and calculate the pressure deviation value and deviation change rate between the standard system pressure data and the target pressure value. The parameter acquisition module is used to construct a pressure controller. Based on the pressure controller, the pressure deviation value and the rate of change of deviation are subjected to fuzzy composite calculation and dynamic parameter adjustment to generate pressure control update parameters. The control signal generation module is used to generate mechanism pressure control signals, and drive the actuator to perform pressure adaptive regulation and compensation closed-loop control based on the mechanism pressure control signals.

Images