PID (Proportion Integration Differentiation) intelligent pressure reducer controller

Abstract

The invention relates to the technical field of automatic control, and discloses a PID (Proportion Integration Differentiation) intelligent pressure reducer controller, which comprises the following steps of: judging whether an environment disturbance or a load sudden change event exists according to a system load change rate and sensor feedback stability, constructing a master control-controlled coupling characteristic matrix by adopting a dual-channel data fusion method, performing attribution splitting on each disturbance channel, and obtaining a PID (Proportion Integration Differentiation) intelligent pressure reducer. PID parameters are dynamically adjusted according to the principal component weight of a disturbance channel, a variable parameter PID adjustment matrix is constructed, filtering smoothing is carried out on sampled signals through an extended Kalman filtering algorithm, a feedback closed-loop correction vector based on sampling residual errors is constructed, a historical offset tracking and switching value boundary judgment mechanism is introduced, fine tuning suppression is carried out on control signals of a motor driving mechanism, and the control signals of the motor driving mechanism are output. Judging whether a mechanical limiting protection device needs to be activated or not; and if yes, automatically triggering voltage limiting protection logic and locking the current control quantity. The method has the advantage of improving efficient and stable operation.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of automatic control, in particular to a PID intelligent pressure reducer controller. BACKGROUND

[0002] The existing PID intelligent pressure reducer controller is widely used in industrial gas or liquid pipeline systems to realize precise regulation of output pressure. Its basic control principle is to collect feedback signals, compare them with the set given value, calculate them according to the deviation signal by the PID control algorithm, and then adjust the opening of the pressure reducing valve by the drive mechanism through the execution motor, so as to realize closed-loop control of pressure and maintain stable system output pressure. At present, the common PID controller is usually composed of a control mainboard (such as a CPU processing unit), a pressure sensor, an execution mechanism, and a communication and power module. For example, the inlet pressure is collected by the high-pressure side P1 pressure sensor, and the control output pressure is obtained by the low-pressure side P2 pressure transmitter. The control board compares the deviation between the set value and the P2 feedback value, adjusts the execution motor in real time according to the PID control logic, and completes the precise control of the valve by the motor drive mechanism. The system also integrates a high-pressure mechanical safety mechanism, a CAN external communication interface, a DC 24V power module, two-way on-off signal input and output ports, and realizes multi-signal linkage and state monitoring. However, in actual working conditions, especially when the environmental factors fluctuate greatly (such as frequent changes in temperature and voltage) or the system load changes sharply (such as frequent opening / closing of the gas end), the traditional PID control system often has difficulty in maintaining fast response and stable regulation. The specific performance is as follows: on the one hand, the sensor may have signal drift in high or low temperature environment, and the controller does not set a dynamic compensation mechanism, resulting in error accumulation; on the other hand, when the system load changes suddenly, the flow fluctuates sharply, the PID parameters are fixed, and the pressure control is delayed, the overshoot is obvious, and the system output is prone to transient instability. Therefore, it is necessary to design an intelligent pressure reducer controller that can improve efficient and stable operation. SUMMARY

[0003] In view of the deficiencies of the prior art, the present application provides a PID intelligent pressure reducer controller, which has the advantages of improving efficient and stable operation and solving the problems in the above background art.

[0004] To achieve the above-mentioned purpose of improving efficient and stable operation, the present application provides the following technical scheme: a PID intelligent pressure reducer controller, comprising: A disturbance pre-checking module: according to the system load change rate and the sensor feedback stability, it is judged whether there is an environmental disturbance or a load mutation event, if yes, it enters the variable decoupling module.

[0005] Variable decoupling module: a double-channel data fusion method is used to construct a master-slave coupling feature matrix, the influence weight of temperature, voltage and load transient flow on control accuracy is analyzed by using the covariant feature matrix, and each disturbance channel is attributed and split.

[0006] Parameter reconstruction module: the PID parameters are dynamically adjusted according to the principal component weight of the disturbance channel, the variable ratio correction factor is introduced into the proportional coefficient, integral time and differential time respectively, the variable parameter PID adjustment matrix is constructed, and it is judged whether the current parameters meet the pressure difference suppression threshold, if not, the feedback enhancement module is entered.

[0007] Feedback enhancement module: the sampling signal is filtered and smoothed by an extended Kalman filter algorithm, the state difference of P1 high-voltage side pressure sensor and P2 pressure transmitter is combined to construct a feedback closed-loop correction vector based on the sampling residual, and a control bias compensation value is output to the execution motor.

[0008] Offset suppression module: the control signal of the motor driving mechanism is fine-tuned and suppressed by introducing a historical offset tracking and on-off quantity boundary judgment mechanism, it is judged whether the mechanical limit protection device needs to be activated, if yes, the pressure limiting protection logic is automatically triggered and the current control quantity is locked.

[0009] Preferably, the process of judging whether there is an environmental disturbance or load mutation event is: Collecting continuous pressure change data of P1 high-voltage side pressure sensor and P2 pressure transmitter; Calculating the pressure change rate per unit time, and comparing it with the motor execution bias and historical load baseline; If the change rate of any sensing channel exceeds the preset threshold, it is determined that there is an environmental disturbance or load mutation event, and the variable decoupling module is triggered for disturbance source identification and attribution.

[0010] Preferably, the process of constructing a master-slave coupling feature matrix by a double-channel data fusion method is: Extracting high-frequency sampling data of input pressure and output pressure in the disturbance period; Based on time synchronization mechanism and signal fitting function, the fluctuation characteristics of the two channels are time-aligned; Introducing a feature weight synergy factor to fuse and model the input and output channel signals, and generating a master-slave coupling feature matrix.

[0011] Preferably, the process of constructing a master-slave coupling feature matrix by a double-channel data fusion method is: Extracting high-frequency sampling data of input pressure and output pressure in the disturbance period; Based on time synchronization mechanism and signal fitting function, the fluctuation characteristics of the two channels are time-aligned; The characteristic weight synergy factor is introduced to fuse and model the input and output channel signals to generate a master-slave coupling characteristic matrix.

[0012] Preferably, the process of dynamically adjusting the PID parameter according to the principal component weight of the disturbance channel is as follows: The disturbance variable after the attribution split is extracted by the principal component analysis method to obtain the first several principal components; Based on the principal component variance contribution rate and the actual influence weight, the proportional coefficient, integral time and differential time distribution adjustment factor in the PID controller is allocated; The PID control parameters under the current system state are dynamically generated using a parameter mapping function and input into the PID adjustment unit.

[0013] Preferably, the process of constructing the variable parameter PID adjustment matrix is as follows: A three-dimensional parameter matrix structure containing a dynamic correction factor is designed to correspond to the adjustment factors in the proportional, integral and differential directions; According to the change trend of the current disturbance principal component, the corresponding value of the three-dimensional matrix is updated in real time; The variable parameter matrix is input into the control loop to detect whether the pressure difference suppression threshold is met. If the pressure difference has not converged to the tolerance range, the feedback enhancement module is triggered.

[0014] Preferably, the process of filtering and smoothing the sampling signal by the extended Kalman filter algorithm is as follows: An extended state transition model containing sensor state, control input and system output is constructed; The state of P1 and P2 pressure sampling signals is estimated using the prediction-correction mechanism, and noise and jitter components are filtered out. The filtered and smoothed signal is used as the system error evaluation reference.

[0015] Preferably, the process of constructing a feedback closed-loop correction vector based on the sampling residual is as follows: The residual vector between the expected output and the actual output is calculated using the filtered P1 and P2 pressure values as input; The control execution lag error is estimated by combining the motor response delay model; The residual and lag error are combined to construct a closed-loop feedback correction vector by linear weighting.

[0016] Preferably, the process of automatically triggering the voltage limiting protection logic and locking the current control quantity is as follows: The response difference between the system output low pressure signal and the motor control command is continuously monitored; If the control deviation amplitude continuously exceeds the set tolerance band, it is judged as an overshoot risk; The mechanical limit protection logic is automatically activated, the current control signal update is suspended, and the state of the execution motor is locked.

[0017] Compared with the prior art, the PID intelligent pressure reducer controller has the following beneficial effects: Through the PID intelligent pressure reducer controller, a complete disturbance perception and parameter adaptive closed-loop system is constructed, which significantly improves the response speed and control accuracy of pressure regulation. In the disturbance pre-check link, the system can identify the external environmental disturbance and load mutation risk in advance to ensure the agility of the control link. The variable decoupling module realizes fine attribution of temperature, voltage and transient flow and other factors through double-channel data fusion and covariant feature analysis, thereby improving the accuracy of interference identification. The parameter reconstruction mechanism adaptively adjusts the PID parameters according to the disturbance principal component weight, so that the control strategy has stronger dynamic adaptability. In the feedback enhancement stage, the extended Kalman filter and the sampling residual vector are introduced to build the controller with strong noise resistance and error self-correction ability. The offset suppression link further improves the stability and safety of the system through historical deviation tracking and boundary protection mechanism. The controller has multi-dimensional disturbance perception, parameter self-reconstruction and high robustness feedback characteristics, and is suitable for high-precision differential pressure control requirements under complex working conditions. BRIEF DESCRIPTION OF DRAWINGS

[0018] Figure 1 The PID intelligent pressure reducer control flowchart of the present application. DETAILED DESCRIPTION

[0019] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0020] Embodiment 1: Please refer to Figure 1 The PID intelligent pressure reducer controller described in the embodiments of the present application comprises: The disturbance pre-check module: according to the system load change rate and the sensor feedback stability, it is judged whether there is an environmental disturbance or a load mutation event, if yes, it enters the variable decoupling module.

[0021] The process of judging whether there is an environmental disturbance or a load mutation event in the disturbance pre-check module is: Collect continuous pressure change data of P1 high-pressure side pressure sensor and P2 pressure transmitter; Through the P1 pressure sensor arranged on the high-pressure side of the system and the P2 pressure sensor arranged on the low-pressure feedback side, the pressure values of the liquid inlet end and the liquid outlet end are sampled in real time and synchronously. The sampling frequency is set according to the response period of the control system, so as to ensure that the transient fluctuation characteristics can be captured. The collected data include the absolute pressure value at each time and the time stamp, which are used for subsequent change rate and trend analysis.

[0022] The pressure change rate per unit time is calculated, and the execution deviation of the motor is compared with the historical load baseline; In a set time window range, the sliding window algorithm is used to analyze the continuous pressure sampling data of P1 high-pressure side and P2 low-pressure side frame by frame. According to the numerical change between the continuous sampling points, the system calculates the pressure fluctuation speed per unit time to evaluate the fast and slow trend of pressure change. Combined with the current execution state of the motor, the execution deviation value, i.e. the deviation between the system set pressure and the actual feedback pressure, is extracted and compared with the historical load baseline modeled in the system. The baseline is generated by long-term running data and has certain statistical stability, which is used to judge whether the current fluctuation is within the acceptable range of normal working condition.

[0023] If the change rate of any sensor channel exceeds the preset threshold, it is determined that there is an environmental disturbance or a load mutation event, and the variable decoupling module is triggered to identify and attribute the disturbance source; The multi-channel disturbance judgment threshold is set. If the change rate of any sensor channel exceeds the set disturbance identification threshold, or the change trend obviously deviates from the predicted value range of the load baseline model, it is determined that there is a sudden disturbance or a load mutation event in the system. Once it is determined that a disturbance occurs, the current pressure fluctuation trend of P1 and P2, the motor driving state, the load feedback and other information are input to the variable decoupling module, and the module is triggered to enter the working state to identify, split and attribute the disturbance source, providing a basis for subsequent PID parameter adjustment and feedback compensation.

[0024] The variable decoupling module: a double-channel data fusion method is used to construct a master-slave controlled coupling feature matrix, and the influence weight of temperature, voltage and load transient flow on control accuracy is analyzed by using the covariant feature matrix, and each disturbance channel is attributed and split.

[0025] The process of constructing a master-slave controlled coupling feature matrix in the variable decoupling module by using a double-channel data fusion method is as follows: High-frequency sampling data of input pressure and output pressure in the disturbance period are extracted; After detecting that there is a suspected disturbance signal in the system, the system immediately samples the input pressure and the output pressure at a high frequency, and the sampling frequency is usually set in the range of 100 Hz to 1 kHz to capture the short-term fluctuation characteristics. The sampling period covers the three stages before, during and after the disturbance, ensuring that the whole process of the disturbance is recorded completely.

[0026] Based on the time synchronization mechanism and signal fitting function, the fluctuation characteristics of two channels are time-aligned; In order to realize the alignment of input and output channel signals on the time axis, the system adopts a time synchronization mechanism, including timestamp standardization processing and interpolation resampling, to unify the time precision of different sensor data. On this basis, a signal fitting function is introduced to reconstruct the curve of the sampling data, thereby reducing the influence of sampling noise and extracting the main fluctuation characteristics of each channel, realizing the time consistency analysis of the input / output pressure change behavior.

[0027] Introducing a feature weight synergy factor, the input and output channel signals are fused and modeled to generate a master-slave coupling feature matrix; After alignment, the system extracts key disturbance characteristic quantities of the input and output channels, including but not limited to: pressure change amplitude, change rate, number of inflection points, response lag time, signal fluctuation frequency, etc. Then, a feature weight synergy factor is introduced, which is allocated according to historical working condition data statistics or model training results, and is used to measure the influence weight of each feature on the control effect; through the normalization processing and weighted fusion of multiple disturbance characteristics of two channels, a master-slave coupling feature matrix is constructed.

[0028] The attribution splitting process of each disturbance channel in the variable decoupling module is: Based on the fused covariant feature matrix, the participation degree of each physical quantity in the disturbance process is identified; After the master-slave coupling feature matrix is constructed, the system extracts the disturbance characteristics corresponding to each physical variable to form a fused covariant feature matrix. This matrix is organized as follows: each column represents a disturbance variable; each row represents a sampling time or a feature window in the disturbance process; each cell records the normalized disturbance intensity or change rate of the variable at the current time; to identify the "participation degree" of each physical quantity in the disturbance, the system calculates the disturbance activation frequency of the variable and its signal occupancy rate in the system response stage, to determine whether the variable actively participates in the disturbance response.

[0029] Calculate the correlation coefficient of the disturbance response intensity of each variable and the system deviation to form a disturbance influence ranking; The system analyzes the correlation between the disturbance intensity of each variable and the actual deviation in the control system; the Pearson correlation coefficient or the Spearman rank correlation coefficient is used to evaluate the linear or nonlinear coupling relationship between the numerical change of each disturbance variable and the fluctuation of the system deviation; the energy, gradient change rate, and time synchronization of the system response peak in the variable disturbance interval are quantitatively scored to form a "disturbance force" measurement index for each disturbance channel; finally, the system generates a disturbance impact list using a weighted ranking method based on the correlation coefficient between the variable and the system deviation and the disturbance response intensity, and arranges all covariates according to their comprehensive influence on the control error from large to small.

[0030] According to the ranking results, the main disturbance variables affecting the control accuracy are attributed to label division; According to the above ranking results, the system sets the disturbance attribution label division standard, which usually includes the following categories: dominant disturbance channel: the top 1-2 variables, usually the disturbance source with the greatest impact on control accuracy and the most prominent change intensity, which is marked as a first-level attribution target; secondary disturbance channel: middle-ranked variables that have a moderate impact on the system, marked as a secondary disturbance channel, and may participate in PID parameter adjustment; invalid disturbance or background fluctuation channel: channels with low correlation or small disturbance intensity, marked as background disturbance, only used for system monitoring and recording, and does not trigger control compensation behavior. In addition, the attribution label supports dynamic updating, and the system can reevaluate and adjust the labels of the disturbance variables based on new data during continuous operation to ensure the real-time and robustness of the disturbance tracing mechanism.

[0031] Parameter reconstruction module: dynamically adjust PID parameters according to the principal component weight of the disturbance channel, introduce variable ratio correction factors to the proportional coefficient, integral time, and differential time, respectively, construct a variable parameter PID adjustment matrix, and determine whether the current parameters meet the pressure difference suppression threshold. If not, enter the feedback enhancement module.

[0032] The process of dynamically adjusting PID parameters in the parameter reconstruction module according to the principal component weight of the disturbance channel is: Extract the first several principal components of the disturbance variables after attribution splitting by principal component analysis method; After completing the attribution splitting and impact ranking of each disturbance variable, the system takes the variables with significant disturbance effect as input, constructs a disturbance factor matrix, and performs principal component analysis. The specific steps are as follows: The disturbance variables are standardized to eliminate the influence of dimension; the covariance matrix is calculated to extract the coupling degree between variables; the eigenvalues and eigenvectors of the covariance matrix are solved to obtain the linear combination of each principal component; the variance contribution rate of each principal component is calculated according to the size of the eigenvalue, and the first several principal components with a cumulative contribution rate reaching a set threshold are selected as the disturbance principal axes, which represent the main change direction and energy concentration of the disturbance variables in the disturbance period, and have the functions of dimension reduction, redundancy removal and structure extraction.

[0033] Based on the principal component variance contribution rate and the actual influence weight, the proportional coefficient, integral time and differential time distribution adjustment factor in the PID controller is adjusted; After extracting the principal components, the system fuses the variance contribution rate corresponding to each principal component with the attribution label influence weight of the original variable to construct a disturbance adjustment weight model of the PID; the specific method includes: The comprehensive disturbance influence coefficient of each principal component is calculated, denoted as α_i, defined as: α_i=variance contribution rate × average disturbance weight According to the potential influence degree of each principal component on the system stability, a PID mapping weight matrix is established to determine: which type of disturbance should the proportional coefficient Kp respond to; how to adjust the time accumulation effect of the integral time Ti to slow disturbances; whether the differential time Td needs to strengthen the prediction ability to short-term high-speed fluctuations; the system constructs an adjustment factor vector β=[β_Kp,β_Ti,β_Td], whose value is determined by the principal component contribution rate and the system sensitivity mapping function.

[0034] The PID control parameters in the current system state are dynamically generated using the parameter mapping function and input to the PID adjustment unit; Taking the above adjustment factor as input, the system dynamically calculates and outputs the PID parameters in the current state using the preset PID parameter mapping function. The mapping function supports nonlinear adjustment, for example: Kp=Kp_base×(1+β_Kp×f1(t)) Ti=Ti_base÷(1+β_Ti×f2(t)) Td=Td_base×(1+β_Td×f3(t)) Where Kp_base, Ti_base, and Td_base are static design values; f1(t), f2(t), and f3(t) are disturbance time evolution functions or sampling window statistical characteristics. After completing parameter generation, the dynamically adjusted Kp, Ti, and Td values are input into the PID adjustment unit in real time to realize adaptive control according to the disturbance characteristics.

[0035] The process of constructing the variable-parameter PID adjustment matrix in the parameter reconstruction module is: The three-dimensional parameter matrix structure is designed to contain dynamic correction factors, which correspond to the adjustment factors of the proportional, integral, and differential directions. The system first designs a three-dimensional parameter matrix based on the traditional PID control structure, with each dimension corresponding to the proportional coefficient, integral time, and differential time. To achieve adaptive changes in controller parameters with system disturbances, the matrix introduces correction factors that dynamically adjust over time, enabling it to respond to different control stages. The correction factor in each direction is determined by the variance contribution rate of the disturbance variable in the principal component analysis process, the disturbance response intensity, and the feedback deviation information, ensuring that the distribution of each adjustment term weight is closely related to the current control state. The adjustment matrix constructed in this way is no longer a fixed parameter combination but has the ability to evolve over time, allowing flexible adjustment of the proportion, integral, and differential forces in the three links according to control requirements.

[0036] According to the current disturbance principal component change trend, the corresponding value of the three-dimensional matrix is updated in real time. The system continuously monitors the main influencing factors in the disturbance channel and extracts their current change trend, such as growth rate, fluctuation frequency, and disturbance direction, through a sliding time window. Through principal component analysis, the dominant disturbance principal axis in the current control system is identified, and its weight value in the disturbance space is extracted. These principal component dynamic parameters are mapped as the basis for updating the adjustment factor. During the update process: the proportional adjustment factor automatically strengthens or weakens according to the sudden rise or fall of disturbance intensity to adjust the response speed; the integral adjustment factor depends on the feedback deviation accumulation and disturbance stability to determine the time span of the integral term; the differential adjustment factor controls the predictive adjustment of future trends of the system by detecting the disturbance slope and pressure gradient change. Each update is based on real-time data to ensure that the adjustment matrix always reflects the actual operating state of the current system.

[0037] The variable parameter matrix is input into the control loop to detect whether the pressure difference suppression threshold is met. If the pressure difference has not converged to the tolerance range, the feedback enhancement module is triggered. The updated variable parameter PID matrix directly acts on the controller output to issue adjustment instructions to the actuator. The control loop continuously detects the pressure difference between P1 and P2 pressure sensors and compares it with the set tolerance range. The judgment logic includes: if the pressure difference is stable within the set threshold, it is considered that the current parameter adjustment is effective, and the control process remains in the existing state; if the pressure difference continues to fluctuate or fails to converge within the preset tolerance range, it indicates that the current parameter adjustment has not completely offset the disturbance effect; at this time, the system will automatically activate the feedback enhancement module to introduce an auxiliary control strategy based on the state residual, further improving the control accuracy and response ability, and ensuring the realization of the system pressure difference control target.

[0038] Feedback enhancement module: through the extended Kalman filtering algorithm for filtering and smoothing of sampling signal, combined with the state difference of P1 high voltage side pressure sensor and P2 pressure transmitter, construct the feedback closed loop correction vector based on the sampling residual, and output the control deviation compensation value to the execution motor.

[0039] The filtering and smoothing process of the sampling signal in the feedback enhancement module by the extended Kalman filtering algorithm is: An extended state transition model is constructed, which includes sensor state, control input and system output. In a typical PID control system, the data of P1 high voltage side and P2 low voltage side pressure sensor has nonlinear characteristics and noise interference. In order to more accurately describe the actual physical system, the system uses extended Kalman filtering method; first, define the system state vector, usually including: pressure state; control input system output; environmental variables; these variables constitute an extended state vector, which is used to describe the true running state of the system at the current time. Based on this, a nonlinear state transition function is established, as well as a nonlinear observation function related to the sensor observation value.

[0040] The state of P1 and P2 pressure sampling signal is estimated by using the prediction-correction mechanism, and the noise and jitter components are filtered out, and the filtered and smoothed signal is used as the system error evaluation reference; The extended Kalman filtering algorithm is based on the "prediction-correction" two-stage mechanism: in the prediction stage, the system uses the nonlinear state transition function to predict the system state at the current time according to the state estimation value at the last time and the control input. This step provides a priori estimate, that is, "how the system will evolve without new observation data" At the same time, the predicted covariance matrix is generated according to the system noise model, which reflects the uncertainty of the current estimate. In the correction stage, the current observation value of P1 and P2 sensor is collected and substituted into the observation function, which is compared with the predicted result to calculate the observation residual; through the linearization Jacobian matrix, the system makes a local linear approximation to the nonlinear part, and then uses Kalman gain to dynamically adjust the priori estimate, so as to generate a more accurate posteriori estimate; the final output state estimation value is the filtering and smoothing result of P1 and P2 pressure channel, which has higher precision and stability. The system uses this smoothed value as the reference input of the subsequent control module, which ensures that the control deviation calculation is more true and accurate, the system response does not appear misjudgment under the background of high frequency disturbance, and the stability and robustness of the overall control system are improved. This step not only reduces the false triggering or false adjustment, but also significantly improves the response quality of the feedback channel, providing more reliable adjustment input for the actuator.

[0041] The feedback enhancement module constructs the feedback closed loop correction vector based on the sampling residual process: With the filtered P1 and P2 pressure values as inputs, calculate the residual vector between the expected output and the actual output. In the extended Kalman filter stage, the P1 and P2 pressure signals have been filtered and smoothed, resulting in more accurate and precise pressure data. Next, we calculate the output residual step based on the system's expected output value and the actual observed value: expected output: the ideal pressure value calculated based on the system's set target or through a predictive model; actual output: the real-time, filtered P1 and P2 pressure values obtained through the sensor; calculate residual vector: residual vector = expected output - actual output residual vector = expected output - actual output; this step produces a residual vector that reflects the performance deviation of the control system at the current time, reflecting the difference between the target pressure and the actual pressure.

[0042] Estimate the control execution lag error in combination with the motor response delay model; In a dynamic control system, the response of the motor and the actuator usually has a certain time delay. In order to improve the accuracy of the system, we need to estimate the error caused by this delay in the control result. First, the motor response delay model describes that after the motor receives the control command, the actual pressure value does not change immediately, but there is a certain time lag. This lag may be due to the inertia of the motor, the physical characteristics of the system, and the reaction speed of the actuator; to estimate this lag error, we analyze historical data to identify the impact of this delay. Specifically, we compare the difference between past control commands and actual feedback signals to estimate the error caused by the lag. By combining historical data with the delay characteristics of the system response, we can more accurately assess the impact of the current lag on the control effect. This helps us consider these delays in the control algorithm, thereby improving the response speed and stability of the system.

[0043] Combine the residual and lag error to construct a closed-loop feedback correction vector through linear weighting; In order to consider the output residual and the error of the execution lag, the system uses linear weighting to combine these two errors into a correction vector. In this way, the system can dynamically adjust the control strategy according to the current residual and estimated lag error; steps: weighting coefficients: set appropriate weighting coefficients α and β according to the real-time performance requirements and control strategy of the system, respectively for the weight adjustment of the residual and lag error. The weighting coefficients are usually set according to experimental data or through adaptive adjustment; correction vector calculation: combine the residual vector and the lag error according to the weighting coefficients to get the final correction vector: feedback correction vector = residual vector + lag error; feedback correction vector = α⋅residual vector + β⋅lag error correction vector effect: this correction vector will be used in the feedback loop to correct the control command, providing more accurate error compensation to ensure fast response and efficient convergence of the system.

[0044] Offset suppression module: Introduce historical offset tracking and switch boundary judgment mechanism, fine-tune and suppress the control signal of motor driving mechanism, judge whether to activate mechanical limit protection device, if yes, automatically trigger voltage limiting protection logic and lock the current control quantity.

[0045] The process of automatically triggering voltage limiting protection logic and locking the current control quantity in the offset suppression module is: Continuous monitoring of the response difference between the low voltage pressure signal at the system output end and the motor control instruction; Data acquisition: The system collects real-time pressure data feedback from the low voltage side pressure sensor and motor control instruction; Deviation calculation: Calculate the response difference between the low voltage pressure and the motor control instruction, i.e. the offset, regularly, to evaluate whether the control signal is accurately transmitted to the system execution end; Response difference monitoring: Compare the calculated deviation with the preset normal working range to judge whether there is abnormal offset.

[0046] If the control offset amplitude continuously exceeds the set tolerance band, it is judged as over-adjustment risk; Over-adjustment detection: If the deviation value continuously exceeds the set tolerance band for several times, the system considers that there is a risk of over-adjustment. This indicates that the control response of the system has deviated from the ideal value, which may lead to instability or damage to the system; Persistence judgment: Not only the single deviation exceeds the tolerance band, but also the deviation amplitude continuously exceeds the standard within a certain time window, indicating that the risk is continuous rather than occasional.

[0047] Automatically activate mechanical limit protection logic, suspend current control signal update and lock the state of the execution motor; Voltage limiting protection logic triggering: Once the system determines that there is an over-adjustment risk, the voltage limiting protection logic is automatically triggered. At this time, the system will interrupt the current control signal update to avoid further over-adjustment; Control signal freezing: The system suspends the update of the motor control signal, so that the execution motor stops receiving new instructions, locks the current control state, and prevents the motor state from continuing to change; Motor state locking: By locking the motor state, further misoperation or mechanical damage caused by over-adjustment is avoided. At the same time, the system enters protection mode and prepares to correct or restore normal control according to subsequent conditions.

[0048] It is to be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting; it is not intended to exclude myriad other embodiments of the present application that other inventors can develop based on the same general inventive concepts embodied by the described embodiments. That is, although the present application is described in terms of particular embodiments and implementations, it is to be understood that the terminology used is for the purpose of descriptive clarity and that it should be taken in its broadest possible sense. This is especially so where "comprising", "including", "containing", "characterized by" or the like are used as subordinating conjunctions. In this context, "comprising" is intended to mean that the embodiments can include a combination of one or more elements and / or features, including those not expressly described, without necessarily relying on any of the other features or elements not expressly described. In addition, it is to be understood that the described embodiments include any alterative or equivalent implementations or substitutions for one or more of the elements or features described. Any such alternatives or equivalents are intended to be encompassed by the description of the described embodiments.

[0049] While the embodiments of the application have been shown and described, it is to be understood that the application is not limited to the details of the embodiments described, since the scope of the expenditure of the application is defined with respect to the appended claims.

Claims

1. PID intelligent pressure reducer controller, characterized in that, include: Disturbance pre-check module: Based on the system load change rate and sensor feedback stability, it determines whether there is an environmental disturbance or load mutation event. If so, it enters the variable decoupling module; Variable decoupling module: A dual-channel data fusion method is used to construct a master-controlled coupling characteristic matrix. The covariate characteristic matrix is ​​used to analyze the weights of the impact of temperature, voltage, and load transient flow on control accuracy, and each disturbance channel is attributed and split. Parameter reconstruction module: Dynamically adjusts PID parameters according to the main component weights of the disturbance channel, introduces ratio correction factors to the proportional coefficient, integral time and differential time respectively, constructs a variable parameter PID adjustment matrix, and determines whether the current parameters meet the pressure difference suppression threshold. If not, it enters the feedback enhancement module; Feedback enhancement module: This module uses an extended Kalman filter algorithm to filter and smooth the sampled signal. It then combines the state differences between the P1 high-pressure side pressure sensor and the P2 pressure transmitter to construct a feedback closed-loop correction vector based on the sampling residual and outputs the control deviation compensation value to the actuator motor. Offset suppression module: Introduces historical offset tracking and switch boundary judgment mechanisms to fine-tune and suppress the control signal of the motor drive mechanism, and determines whether the mechanical limit protection device needs to be activated. If so, it automatically triggers the voltage limit protection logic and locks the current control quantity.

2. The PID intelligent pressure reducer controller according to claim 1, characterized in that: The process of determining whether there is an environmental disturbance or load mutation event is as follows: Collect continuous pressure change data from the P1 high-pressure side pressure sensor and the P2 pressure transmitter; Calculate the pressure change rate per unit time and compare it with the historical load baseline in combination with the motor execution deviation; If the rate of change of any sensing channel exceeds the preset threshold, it is determined to be an environmental disturbance or a sudden load change event, triggering the variable decoupling module to identify and attribute the disturbance source.

3. The PID intelligent pressure reducer controller according to claim 2, characterized in that: The process of constructing the master-controlled coupling characteristic matrix using the dual-channel data fusion method is as follows: Extract high-frequency sampling data of input pressure and output pressure during the disturbance period; Based on the time synchronization mechanism and signal fitting function, the fluctuation characteristics of the two channels are time-series aligned; The feature weight synergy factor is introduced to perform fusion modeling on the input and output channel signals and generate the master-controlled coupling feature matrix.

4. The PID intelligent pressure reducer controller according to claim 3, characterized in that: The process of attributing and splitting each disturbance channel is as follows: Based on the fused covariate feature matrix, the participation of each physical quantity in the disturbance process is identified; Calculate the correlation coefficient between the disturbance response intensity of each variable and the system deviation to form the disturbance impact ranking; Based on the sorting results, the main disturbance variables that affect the control accuracy are attributed and divided.

5. The PID intelligent pressure reducer controller according to claim 4, characterized in that: The process of dynamically adjusting the PID parameters according to the principal component weights of the disturbance channel is: The disturbance variables after attribution splitting are used to extract the first several principal components through principal component analysis method; Based on the principal component variance contribution rate and actual influence weight, adjustment factors are allocated to the proportional coefficient, integral time and differential time in the PID controller; The parameter mapping function is used to dynamically generate the PID control parameters under the current system state and input them into the PID adjustment unit.

6. The PID intelligent pressure reducer controller according to claim 5, characterized in that: The process of constructing the variable parameter PID adjustment matrix is: Design a three-dimensional parameter matrix structure containing dynamic correction factors, corresponding to the adjustment factors in the three directions of proportion, integration, and differentiation; According to the changing trend of the current disturbance principal component, the corresponding value of the three-dimensional matrix is ​​updated in real time; The variable parameter matrix is ​​input into the control loop to detect whether the pressure difference suppression threshold is met. If the pressure difference does not converge to the tolerance range, the feedback enhancement module is triggered.

7. The PID intelligent pressure reducer controller according to claim 6, characterized in that: The process of filtering and smoothing the sampled signal using the extended Kalman filtering algorithm is as follows: Construct an extended state transition model that includes sensor states, control inputs, and system outputs; The prediction-correction mechanism is used to estimate the state of the pressure sampling signals of P1 and P2, and the noise and jitter components are filtered out. The filtered smoothed signal is used as the benchmark for system error evaluation.

8. The PID intelligent pressure reducer controller according to claim 7, characterized in that: The process of constructing a feedback closed-loop correction vector based on sampling residuals is: Taking the filtered P1 and P2 pressure values ​​as input, calculate the residual vector between the expected output and the actual output; Combined with the motor response delay model, the control execution hysteresis error is estimated; The residual error is combined with the hysteresis error to construct a closed-loop feedback correction vector through a linear weighted sum method.

9. The PID intelligent pressure reducer controller according to claim 8, characterized in that: The process of automatically triggering the voltage limiting protection logic and locking the current control quantity is as follows: Continuously monitor the response difference between the low-voltage pressure signal at the system output and the motor control command; If the control deviation amplitude is detected to continuously exceed the set tolerance band, it is judged as an overshoot risk; Automatically activate the mechanical limit protection logic, suspend the current control signal update and lock the state of the executing motor.

Images