Load automatic tracking and intelligent adjustment system

Abstract

This invention relates to the field of industrial automatic control technology and discloses an automatic load tracking and intelligent adjustment system, comprising: a signal input module to acquire feedback signals and command signals of the controlled object; a state observation module to construct a total disturbance expansion state variable including model uncertainty and load disturbance; a dynamic boundary verification module to extract the rate of change of the variables and compare it with the dynamic response boundary of the actuator, identify and eliminate non-physical noise whose evolution rate exceeds the upper limit of the physical action frequency, and output the effective load disturbance component; and an error feedback compensation module to retrieve the transmission characteristic compensation table according to the component, and use the mechanical backlash compensation value and static friction parameters to perform advance compensation on the command signal. This invention blocks non-physical interference pulses through physical consistency verification, solves the constraint contradiction between control bandwidth and measurement noise, improves the operating stability of the controlled object, and effectively suppresses the mechanical resonance generated by the actuator.

Description

Technical Field

[0001] This invention belongs to the field of industrial automatic control technology, and in particular relates to an automatic load tracking and intelligent adjustment system. Background Technology

[0002] In current industrial process control, eliminating the deviation between the setpoint and the controlled variable is the foundation for ensuring production stability. Existing technologies mostly adopt proportional-integral-derivative (PID) control and active disturbance rejection (ADR) technology to handle load disturbances by constructing feedback closed loops. However, in nonlinear time-varying conditions such as temperature control in chemical polymerization reactors and torque control in precision servo stamping systems, the load exhibits high-frequency and large-amplitude random fluctuations. Conventional control strategies are limited by their error-driven characteristics, and the adjustment action lags behind the instantaneous node where the disturbance occurs. When the control bandwidth is increased to improve the tracking speed, the system is prone to capturing and amplifying stray signals in the measurement circuit, causing non-physical chatter in the actuator and increasing the risk of wear on transmission components.

[0003] Industrial practice shows that even if the control level captures high-frequency disturbances at the computational level, its output commands still encounter mechanical constraints when converted into physical displacements. Existing control models typically ignore the transmission backlash and initial static friction within the actuator. Such execution dissipation causes nonlinear absorption of control energy in the initial stage of adjustment, resulting in a structural misalignment between the calculated command and the actual physical response, thus worsening the system's phase margin. Existing technologies mainly suffer from the following shortcomings: 1. The adjustment mechanism possesses a hysteresis attribute driven by deviation, making it difficult to achieve active immunity to load disturbances; 2. There is a physical trade-off between bandwidth enhancement and noise suppression, making it difficult to balance response speed and operational stability; 3. Control commands lack anticipatory compensation for the mechanical losses of the actuator, leading to unexpected deviations in the initial stage of adjustment. Besides hardware limitations, software control methods also face similar challenges. However, this approach has its shortcomings. For example, Chinese invention patent CN106707756B discloses a method for coordinated control of supercritical thermal power units with integrated expansion observers. This method suppresses disturbances by superimposing multi-model predictions and expansion state observers. Under actual operating conditions, the observer gain setting is based on a mathematical model, neglecting the dynamic response boundary of the controlled object's actuator. When electromagnetic noise with an evolution rate far exceeding the mechanical response limit exists in the measurement circuit, it is identified as a disturbance to be compensated and triggers the adjustment action, causing non-physical high-frequency chattering of the actuator, deteriorating the system's phase margin and accelerating the wear of transmission components. The control model ignores the internal transmission gap and initial static friction of the actuator, and the execution dissipation phenomenon causes nonlinear absorption of control energy in the early stage of adjustment, resulting in a structural misalignment between the command calculation results and the actual physical response.

[0004] Therefore, how to achieve real-time observation of load disturbance through physical consistency verification, and how to construct a feedforward compensation mechanism in combination with the execution dissipation spectrum to eliminate control lag and execution chatter under high-frequency operating conditions, has become the technical problem to be solved by this invention. Summary of the Invention

[0005] This invention provides a load automatic tracking and intelligent adjustment system, the system comprising: The signal input module is used to acquire feedback signals and command signals from the controlled object; The state observation module is used to construct the total disturbance expansion state variables, which include uncertainties in the controlled object model and external load disturbances, using feedback signals. The dynamic boundary verification module is used to extract the rate of change of the total disturbance expansion state variable and compare the rate of change with the preset dynamic response boundary of the actuator of the controlled object. Within the dynamic response boundary, it identifies non-physical noise whose evolution rate exceeds the upper limit of the physical action frequency of the actuator and outputs the verified effective load disturbance component. The error feedback compensation module is used to retrieve the pre-stored transmission characteristic compensation table based on the effective load disturbance component, and extract the mechanical backlash compensation value and static friction parameters from the transmission characteristic compensation table. It then performs nonlinear advance compensation on the command signal to generate control commands acting on the actuator. The dynamic boundary verification module determines the rate of change by calculating the derivative of the total disturbance expansion state variable with respect to time, and establishes constraint logic between the rate of change and the maximum displacement response of the actuator per unit time. This constraint logic filters out interference signals that are higher than the dynamic response boundary.

[0006] Preferably, the dynamic boundary verification module performs the following process: extracting the first derivative value of the total disturbance expansion state variable as the rate of change in real time; determining the physical feasibility of the total disturbance expansion state variable by judging whether the first derivative value is within the dynamic response slope limit of the actuator; if the first derivative value exceeds the dynamic response slope limit, the dynamic boundary verification module locks the effective load disturbance component of the current period to the historical temporary value of the previous sampling period to block non-physical pulses from entering the control loop.

[0007] Preferably, the dynamic boundary check module follows the following gating rules when outputting the effective load disturbance component: Where v is the instantaneous derivative of the total disturbance expanded state variable with respect to time. The dynamic response boundary value is preset for the controlled object, and G(v) is the logic gated state factor. When G(v) is 1, the total disturbance expansion state variable is transmitted as the effective load disturbance component. When G(v) is 0, the value of the effective load disturbance component is kept constant.

[0008] Preferably, the transmission characteristic compensation table stores nonlinear compensation values ​​characterizing the actuator under different load ranges; the nonlinear compensation values ​​include mechanical backlash compensation values ​​for the commutation stage and friction torque gain for the low-speed stage; the error feedback compensation module locates the index address in the transmission characteristic compensation table according to the amplitude of the effective load disturbance component, and injects the extracted compensation value into the control command.

[0009] Preferably, the system further includes an instruction differentiation module for smoothing the trajectory of the instruction signal and extracting the differential feedforward signal; the error feedback compensation module performs vector superposition of the differential feedforward signal and the effective load disturbance component to offset the dynamic lag of the controlled object in the transmission chain.

[0010] Preferably, the signal input module includes an analog signal isolator and an analog-to-digital converter; the analog signal isolator is used to acquire the induced current or feedback voltage of the controlled object through electromagnetic coupling; the analog-to-digital converter is used to convert the induced current or feedback voltage into a digital signal, and the sampling frequency is not less than 10kHz.

[0011] Preferably, the state observation module runs in the real-time controller; the state observation module adopts a third-order nonlinear observer topology, and uses the deviation between the feedback signal and the internal state estimate to achieve convergence of the total disturbance extended state variable by adjusting the observer gain factor.

[0012] Preferably, the error feedback compensation module includes a stability monitoring unit for monitoring the pulsation frequency of the effective load disturbance component; if the pulsation frequency reaches more than 80% of the system closed-loop bandwidth, the stability monitoring unit automatically attenuates the injection intensity of the nonlinear compensation quantity to suppress the mechanical resonance generated by the actuator.

[0013] Preferably, the control command is converted into a control pulse with a frequency between 2kHz and 20kHz via a pulse width modulation interface to drive the power conversion device to adjust the input energy of the controlled object.

[0014] Preferably, the controlled parameters of the controlled object are selected from temperature data in the range of 0°C to 200°C in the reactor, or torque data in the range of 0 N·m to 500 N·m in the servo mechanism; the system limits the rate of change of the control command in real time through the logical association between the effective load disturbance component and the transmission characteristic compensation table, so that it does not exceed the upper limit of the dynamic response frequency of the actuator.

[0015] Compared with existing technologies, the automatic load tracking and intelligent adjustment system of the present invention has the following advantages: 1. In automatic load tracking, the system achieves high synchronization between control commands and physical actions in micro-time by deeply coupling the transient process planning module and the physical consistency gating observation module. Using the physical response envelope as a criterion, the system retrieves the corresponding compensation data stream at the moment it identifies the actual load fluctuation, combined with the preset execution dissipation spectrum, and feeds the compensation amount forward into the control command. This allows the control energy to offset the inherent transmission gap and initial static friction dissipation of the actuator before it is converted into actual mechanical action. This avoids the phase shift that inevitably occurs when traditional closed-loop regulation overcomes physical lag, enabling the system to maintain high dynamic tracking accuracy when dealing with large load changes.

[0016] 2. The constructed momentum-based interference classification and shielding mechanism effectively resolves the inherent constraint between control bandwidth and measurement noise in industrial settings. The physical consistency gating observation module does not rely on conventional frequency domain filtering methods. Instead, it calculates the rate of change of the total disturbance expansion state variable and maps it to the system's maximum physical response envelope space. Since the evolution rate of non-physical electromagnetic interference spikes far exceeds the physical power boundary of the actuator, the system can deterministically identify and block false disturbances through this physical consistency verification. This ensures that the controller maintains high bandwidth to capture the real load without triggering invalid adjustment actions due to sensor noise, thereby improving the operational stability of the control loop in complex electromagnetic environments.

[0017] 3. By leveraging the logical linkage between the nonlinear error feedback compensation module and the execution dissipation graph library, the system avoids non-physical fatigue wear of the actuator in principle. It verifies the trajectory characteristics within the initial time window of the disturbance signal and calls the matching logical rack bounce parameters for real-time adjustment, constructing a physical-level logical firewall. This not only enables adjustment commands to accurately bypass the actuator's dead zone but also filters out all high-frequency command components exceeding the physical object's response capability, preventing invalid high-frequency chattering in regulating valves or servo motors, reducing physical losses in key transmission components, and extending the service life of the core hardware of the industrial control system. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the internal interference resistance and signal flow topology of the intelligent adjustment system of the present invention; Figure 2 This is the logic diagram of the multi-dimensional signal feedforward and error compensation for the transmission characteristics of this invention. Detailed Implementation

[0019] The technical solutions of the embodiments of this application will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of this application are within the scope of protection of this application.

[0020] It should be noted that all directional and positional terms used in this invention, such as: up, down, left, right, front, back, vertical, horizontal, inner, outer, top, bottom, transverse, longitudinal, center, etc., are only used to explain the relative positional relationship and connection between components in a specific state (as shown in the accompanying drawings). They are only for the convenience of describing this invention and do not require that this invention be constructed and operated in a specific orientation. Therefore, they should not be construed as limiting this invention. In addition, the descriptions of "first," "second," etc., in this invention are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated.

[0021] In the description of this invention, unless otherwise explicitly specified and limited, the terms installation, connection, and linking should be interpreted broadly. For example, they can refer to fixed connections, detachable connections, or integral connections; they can refer to mechanical connections; they can refer to direct connections or indirect connections through an intermediate medium; they can refer to the internal connection of two components. For those skilled in the art, the specific meaning of the above terms in this invention can be understood according to the specific circumstances.

[0022] In the description of this specification, references to the terms "an embodiment," "some embodiments," "illustrative embodiments," "examples," "specific examples," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example, and the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0023] An automatic load tracking and intelligent regulation system, the system comprising: The signal input module is used to acquire feedback signals and command signals from the controlled object; The state observation module is used to construct the total disturbance expansion state variables, which include uncertainties in the controlled object model and external load disturbances, using feedback signals. The dynamic boundary verification module is used to extract the rate of change of the total disturbance expansion state variable and compare the rate of change with the preset dynamic response boundary of the actuator of the controlled object. Within the dynamic response boundary, it identifies non-physical noise whose evolution rate exceeds the upper limit of the physical action frequency of the actuator and outputs the verified effective load disturbance component. The error feedback compensation module is used to retrieve the pre-stored transmission characteristic compensation table based on the effective load disturbance component, and extract the mechanical backlash compensation value and static friction parameters from the transmission characteristic compensation table. It then performs nonlinear advance compensation on the command signal to generate control commands acting on the actuator. The dynamic boundary verification module determines the rate of change by calculating the derivative of the total disturbance expansion state variable with respect to time, and establishes constraint logic between the rate of change and the maximum displacement response of the actuator per unit time. This constraint logic filters out interference signals that are higher than the dynamic response boundary.

[0024] Preferably, the dynamic boundary verification module performs the following process: extracting the first derivative value of the total disturbance expansion state variable as the rate of change in real time; determining the physical feasibility of the total disturbance expansion state variable by judging whether the first derivative value is within the dynamic response slope limit of the actuator; if the first derivative value exceeds the dynamic response slope limit, the dynamic boundary verification module locks the effective load disturbance component of the current period to the historical temporary value of the previous sampling period to block non-physical pulses from entering the control loop.

[0025] Preferably, the dynamic boundary check module follows the following gating rules when outputting the effective load disturbance component: Where v is the instantaneous derivative of the total disturbance expanded state variable with respect to time. The dynamic response boundary value is preset for the controlled object, and G(v) is the logic gated state factor. When G(v) is 1, the total disturbance expansion state variable is transmitted as the effective load disturbance component. When G(v) is 0, the value of the effective load disturbance component is kept constant.

[0026] Preferably, the transmission characteristic compensation table stores nonlinear compensation values ​​characterizing the actuator under different load ranges; the nonlinear compensation values ​​include mechanical backlash compensation values ​​for the commutation stage and friction torque gain for the low-speed stage; the error feedback compensation module locates the index address in the transmission characteristic compensation table according to the amplitude of the effective load disturbance component, and injects the extracted compensation value into the control command.

[0027] Preferably, the system further includes an instruction differentiation module for smoothing the trajectory of the instruction signal and extracting the differential feedforward signal; the error feedback compensation module performs vector superposition of the differential feedforward signal and the effective load disturbance component to offset the dynamic lag of the controlled object in the transmission chain.

[0028] Preferably, the signal input module includes an analog signal isolator and an analog-to-digital converter; the analog signal isolator is used to acquire the induced current or feedback voltage of the controlled object through electromagnetic coupling; the analog-to-digital converter is used to convert the induced current or feedback voltage into a digital signal, and the sampling frequency is not less than 10kHz.

[0029] Preferably, the state observation module runs in the real-time controller; the state observation module adopts a third-order nonlinear observer topology, and uses the deviation between the feedback signal and the internal state estimate to achieve convergence of the total disturbance extended state variable by adjusting the observer gain factor.

[0030] Preferably, the error feedback compensation module includes a stability monitoring unit for monitoring the pulsation frequency of the effective load disturbance component; if the pulsation frequency reaches more than 80% of the system closed-loop bandwidth, the stability monitoring unit automatically attenuates the injection intensity of the nonlinear compensation quantity to suppress the mechanical resonance generated by the actuator.

[0031] Preferably, the control command is converted into a control pulse with a frequency between 2kHz and 20kHz via a pulse width modulation interface to drive the power conversion device to adjust the input energy of the controlled object.

[0032] Preferably, the controlled parameters of the controlled object are selected from those inside the reactor. to Temperature data within the range, or torque data within the range of 0 N·m to 500 N·m in the servo mechanism; the system limits the rate of change of control commands in real time through the logical association between the effective load disturbance component and the transmission characteristic compensation table, so that it does not exceed the upper limit of the dynamic response frequency of the actuator.

[0033] Example 1: In a nonlinear time-varying industrial process scenario where torque data in a servo mechanism ranging from 0 N·m to 500 N·m is controlled, the controlled object faces high-frequency and random large-amplitude external load disturbances that occur with the process progress. The deviation-driven adjustment mechanism exhibits phase lag under such conditions. Furthermore, increasing the control closed-loop bandwidth to improve tracking speed synchronously amplifies stray signals induced in the measurement circuit, causing non-physical high-frequency chatter and mechanical wear in the underlying actuator. Simultaneously, the inherent transmission gap and initial static friction dissipation within the actuator cause nonlinear absorption of control energy in the initial adjustment phase, resulting in a structural phase misalignment between the calculated command and the actual physical displacement. The automatic load tracking and intelligent adjustment system observes the disturbance and performs dynamic compensation at the control end. The signal input module of this system... The module includes an analog signal isolator and an analog-to-digital converter. The analog signal isolator acquires the feedback signal and command signal of the controlled object through electromagnetic coupling. The analog-to-digital converter with a sampling frequency of not less than 10kHz converts the feedback signal into a digital signal. The state observation module running in the real-time controller uses the feedback signal to construct a total disturbance expansion state variable that includes the uncertainty of the controlled object model and external load disturbances. The dynamic boundary verification module calculates the derivative of the total disturbance expansion state variable with respect to time to determine the rate of change and establishes the constraint logic between the rate of change and the maximum displacement response of the actuator per unit time. The rate of change is compared with the preset dynamic response boundary of the actuator of the controlled object. Within the dynamic response boundary, non-physical noise whose evolution rate exceeds the upper limit of the physical action frequency of the actuator is identified.

[0034] The dynamic boundary verification module performs dynamic boundary verification by extracting the first derivative of the total disturbance expansion state variable as the rate of change in real time. It determines the physical feasibility of the total disturbance expansion state variable by judging whether the first derivative is within the limit of the dynamic response slope of the actuator. When outputting the effective load disturbance component, this module follows a gating rule based on the logic-gated state factor G(v), where v is the instantaneous derivative of the total disturbance expansion state variable with respect to time. The preset dynamic response boundary value for the controlled object, when v is less than or equal to When G(v) takes a value of 1 and transmits the total disturbance expansion state variable as the effective load disturbance component, when v is greater than 1... When the first derivative value exceeds the dynamic response slope limit, G(v) takes the value of 0 and keeps the effective load disturbance component constant. The effective load disturbance component of the current period is locked to the historical temporary value of the previous sampling period. In this way, non-physical pulses are blocked from entering the control loop, and the verified effective load disturbance component is output. The effective load disturbance component output by the dynamic boundary verification module provides a feedforward physical quantity input for the error feedback compensation module. The error feedback compensation module locates the index address in the pre-stored transmission characteristic compensation table according to the amplitude of the effective load disturbance component. This transmission characteristic compensation table stores the characteristics of the transmission characteristic compensation module. The nonlinear compensation amount of the actuator under different load ranges is calculated. The error feedback compensation module extracts the mechanical backlash compensation value for the commutation stage and the static friction parameter for the low speed stage from the transmission characteristic compensation table. The extracted nonlinear compensation amount is injected into the control command to perform nonlinear advance compensation on the command signal to offset the inherent execution dissipation of the actuator. At the same time, the command differentiation module included in the system performs trajectory smoothing processing on the command signal and extracts the differential feedforward signal. The error feedback compensation module performs vector superposition of the differential feedforward signal and the effective load disturbance component to offset the dynamic lag of the controlled object in the transmission chain.

[0035] The stability monitoring unit inside the error feedback compensation module continuously monitors the pulsation frequency of the effective load disturbance component. When the pulsation frequency reaches more than 80% of the system closed-loop bandwidth, the stability monitoring unit automatically attenuates the injection intensity of the nonlinear compensation quantity. The control command generated and applied to the actuator is converted into a control pulse with a frequency between 2kHz and 20kHz through the pulse width modulation interface to drive the power conversion device to adjust the input energy of the controlled object. The load automatic tracking and intelligent adjustment system limits the rate of change of the control command in real time through the logical association between the effective load disturbance component and the transmission characteristic compensation table, so that it does not exceed the upper limit of the dynamic response frequency of the actuator. The actuator avoids the generation of high-frequency chatter while crossing the static friction dead zone. Under complex electromagnetic interference conditions, the control pulse and the input energy adjustment of the controlled object achieve deterministic synchronization of physical actions.

[0036] Example 2: In the scenario of measuring control parameters of a stirring servo mechanism in a nonlinear time-varying industrial polymerization reactor, the servo test platform is equipped with a permanent magnet synchronous motor with a rated torque of 500 N·m and a dynamic torque sensor with a resolution of 0.01 N·m. The signal generator injects Gaussian white noise with a signal-to-noise ratio of 20 dB and power frequency interference harmonics with a frequency of 50 Hz into the input of the servo test platform to simulate background noise in the industrial electromagnetic environment. The main technical consideration for setting the sampling frequency is to balance the real-time capture of high-frequency disturbance signals with the data processing load of the real-time controller. When the bandwidth of the monitored feedback signal spectrum contains transient step changes, in order to avoid signal aliasing, the sampling frequency is required to be greater than five times the highest effective frequency component of the feedback signal. The system determines the sampling frequency of the analog-to-digital converter to be 10 kHz based on this judgment rule.

[0037] To determine the boundary blocking mechanism of the total disturbance expansion state variable, the test procedure set three gradients of transient torque disturbance change rate inputs: low, medium, and high, corresponding to step disturbances of 50 N·m / ms, 200 N·m / ms, and 400 N·m / ms, respectively. The state observation module inside the real-time controller received a feedback signal superimposed with 20 dB Gaussian white noise and constructed the total disturbance expansion state variable. The dynamic boundary verification module calculated the instantaneous derivative v of the total disturbance expansion state variable with respect to time in real time, where v represents the instantaneous evolution rate of the disturbance physical quantity. Test data showed that under the high disturbance input of 400 N·m / ms, the original first-order derivative value, including the high-frequency noise superposition, instantly rose to 452.3 N·m / ms, which exceeded the preset dynamic response boundary value of the actuator connected to the servo test platform. The subscript max represents the physical extreme value of the response boundary; according to the gating rule of the logic gating state factor G(v), when v is greater than When G(v) is triggered to be 0, the system locks the effective load disturbance component of the current period to the historical temporary value of the previous sampling period, filters out the aforementioned spike pulse of 452.3 N·m / ms, and outputs a smooth amplitude constrained to within... Effective load disturbance components within the range.

[0038] The experiment included a partially missing control group with a state observation module but without a dynamic boundary verification module, and an out-of-range control group with dynamic response boundary values ​​deviating from physical extremes. In the partially missing control group, when faced with a 400 N·m / ms high-level disturbance and high-frequency noise superimposed on it, the unverified total disturbance expansion state variable directly drove the error feedback compensation module, causing high-frequency oscillations in the extracted nonlinear compensation. The high-frequency pulses output from the drive-end pulse width modulation interface triggered mechanical resonance in the actuator, with an amplitude increase to 2.53 mm. In the out-of-range control group, when… When the setpoint exceeds 150% of the upper limit of the physical action frequency of the actuator, the system tracking error surges from 0.48% in steady state to 8.72% and phase lag occurs. When the setpoint is lower than 50% of the actuator's rated response capability, the system suppresses effective intermediate frequency load changes, causing the tracking trajectory to deviate from the actual load state of the controlled object; the test group matched the physical limits. Under the given conditions, a verified effective load disturbance component positioning transmission characteristic compensation table is used to inject static friction parameters for the low-speed stage and mechanical backlash compensation values ​​for the commutation stage into the control command. Test results show that under high disturbance conditions, the test group suppresses the mechanical amplitude to 0.11 mm and maintains the full-band tracking error at 0.45%. The test data indicates that the total disturbance expansion state variable extraction step in the controlled object feedback signal and the gating mechanism based on the dynamic response boundary have a synergistic effect. The physical constraint condition based on the rate of change blocks the amplification of high-frequency pulses in the closed-loop control link. Nonlinear advance compensation offsets the static friction dead zone dissipation and avoids high-frequency chatter of the servo actuator. Under electromagnetic interference and high-frequency load change conditions, the system outputs smooth control pulses to match the actual mechanical displacement.

[0039] Example 3: In the initial deployment phase of the industrial automatic control system, the frictional dissipation and mechanical backlash in the physical transmission chain of the servo mechanism manifest as nonlinear hysteresis factors. If the feedforward command is generated based on the factory nominal value, this hysteresis factor causes phase deviation at the physical actuator. The load automatic tracking and intelligent adjustment system starts the offline calibration program to extract the physical parameters of the servo mechanism. The system drives the servo actuator to perform forward and reverse operation at a constant low speed of 1 rad / s without external load. The dynamic torque sensor located on the transmission shaft side collects the steady-state electromagnetic torque data in real time during this bidirectional motion process. The system calculates the absolute value of the difference between the forward steady-state electromagnetic torque and the reverse steady-state electromagnetic torque, and establishes half of this absolute value as the static friction parameter. At the same time, at the transient moment when the servo actuator receives the reverse voltage command, the position encoder records the angular hysteresis from the level flip to the actual mechanical shaft generating the first effective angular displacement. The system maps this angular hysteresis to the mechanical backlash compensation value. The controller establishes a mapping relationship between the measured data and the corresponding load range and writes it into the non-volatile memory to generate a transmission characteristic compensation table for online recall.

[0040] After offline extraction of physical parameters, the system enters a real-time closed-loop control cycle. The signal input module acquires the current feedback signal from the controlled object, and the state observation module extracts the external load state based on the discrete difference equation. The update logic follows mathematical expressions. ,in The total disturbance expansion state variable for the current period, The total disturbance expansion state variable of the previous period is given by h, where h is the sampling period. The tracking error between the current cycle feedback signal and the position estimate is β, which is the observer gain coefficient. The state observation module analyzes the transient resistance through this time integration calculation path. The error feedback compensation module uses the nonlinear compensation amount of the corresponding load range in the transmission characteristic compensation table to superimpose the command signal. The mechanical dissipation in the transmission chain is offset through the physical alignment mechanism, so that the actual drive trajectory and the adjustment curve remain synchronized in the time domain.

[0041] Example 4: In the initial stage of operation of an industrial automatic control system, the physical transmission components between the actuator and the controlled object face differences in assembly tolerances and initial lubrication conditions. If the error feedback compensation module included in the system lacks a benchmark setting for the initial physical environment, the extracted nonlinear compensation quantity cannot accurately map the hindrance characteristics of the current environment. Before receiving control commands, the load automatic tracking and intelligent adjustment system starts the dynamic friction profile calibration procedure. The system controller drives the servo actuator to output a low-frequency sinusoidal detection signal with a frequency of 0.1Hz and a peak torque of 10% of the rated torque. The analog-to-digital converter synchronously acquires the rotational speed feedback sequence and displacement feedback sequence of the controlled object, calculates the energy integral of the detection signal in a complete cycle, and extracts the transient torque offset when the rotational speed crosses zero.

[0042] The system calculates the difference between the energy integral and the energy consumption of the theoretical frictionless model to generate an initial global viscosity coefficient characterizing the system's viscous resistance. Simultaneously, it maps the transient torque offset to an initial static friction dead zone value for the current assembly tolerance. The system writes the initial global viscosity coefficient and the initial static friction dead zone value as reference data into a non-volatile memory and uses this reference data to scale and correct the static friction parameters in the transmission characteristic compensation table. Based on the acquired reference data, a customized compensation baseline is established, enabling the system to output a feedforward control quantity that matches the actual mechanical clearance when dealing with high-frequency load changes.

[0043] Example 5: When a novel servo mechanism with unknown physical response characteristics is connected to an industrial automatic control system device, if the dynamic response boundary value of the controlled object cannot be quantified, and the dynamic boundary verification module lacks a benchmark for intercepting non-physical noise, the system initiates an offline dynamic limit extraction procedure before closed-loop operation. The controller drives the power conversion device to output a sinusoidal sweep excitation signal with constant amplitude and a frequency increasing from 10Hz to 500Hz in a logarithmic stepwise manner to the unloaded servo actuator. The position encoder captures the angular displacement sequence of the mechanical transmission shaft in real time and imports it into the fast Fourier transform unit. This unit calculates the continuous amplitude-frequency response trajectory. The system extracts the physical cutoff frequency corresponding to the point where the displacement amplitude decays to 70.7% of the amplitude in the low-frequency flat region of the trajectory. By calculating the differential product of this cutoff frequency and the nominal amplitude at this time, the maximum allowable evolution rate of the mechanical transmission chain is derived and written into the register as a hard threshold to set as the preset dynamic response boundary value of the controlled object. The specific numerical conversion logic for generating measured numerical inputs for the dynamic boundary verification module to execute the truncation logic is as follows: The system multiplies the extracted 125Hz physical cutoff frequency by the angular velocity constant of 6.28, then multiplies the resulting value by the maximum response amplitude of the actuator under rated load, 10.0mm, and finally divides it by the 1000 millisecond conversion coefficient to obtain the maximum physical displacement change that the controlled object can produce within 1ms, which is 7.85mm per millisecond. This value is written into the comparator register as a hard comparison threshold. If the instantaneous rate of change exceeds this limit, it is determined to be an electromagnetic interference pulse.

[0044] To address the control command output requirements when the system approaches high-frequency boundaries, the stability monitoring unit within the error feedback compensation module is equipped with a continuous attenuation algorithm path. When the pulsation frequency of the effective load disturbance component exceeds 80% of the system closed-loop bandwidth, this unit calculates the exponential compensation weight based on the discrete mapping relationship. Its mathematical model is as follows: Where α is the attenuation coefficient of the nonlinear compensation, and f is the extracted real-time pulsation frequency. The closed-loop bandwidth is calibrated for the system, and k is the slope constant set in the controller. The error feedback compensation module multiplies the extracted static friction parameters and mechanical hysteresis compensation values ​​by the attenuation coefficient α and then superimposes them onto the control command. The system uses exponential derating constraints to reduce redundant compensation energy at the frequency domain boundary, so that the actuator can filter out high-frequency resonant waves while maintaining a low-frequency nonlinear advance compensation trajectory.

[0045] The embodiments of this application have been described above with reference to the accompanying drawings. Unless otherwise specified, the embodiments and features in the embodiments of this application can be combined with each other. This application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit of this application and the scope of protection of this invention, and all of these forms are within the protection scope of this application.

Claims

1. A load automatic tracking and intelligent adjustment system, characterized in that, The system includes: The signal input module is used to acquire feedback signals and command signals from the controlled object; The state observation module is used to construct the total disturbance expansion state variables, which include uncertainties in the controlled object model and external load disturbances, using feedback signals. The dynamic boundary verification module is used to extract the rate of change of the total disturbance expansion state variable and compare the rate of change with the preset dynamic response boundary of the actuator of the controlled object. Within the dynamic response boundary, it identifies non-physical noise whose evolution rate exceeds the upper limit of the physical action frequency of the actuator and outputs the verified effective load disturbance component. The error feedback compensation module is used to retrieve the pre-stored transmission characteristic compensation table based on the effective load disturbance component, and extract the mechanical backlash compensation value and static friction parameters from the transmission characteristic compensation table. It then performs nonlinear advance compensation on the command signal to generate control commands acting on the actuator. The dynamic boundary verification module determines the rate of change by calculating the derivative of the total disturbance expansion state variable with respect to time, and establishes constraint logic between the rate of change and the maximum displacement response of the actuator per unit time. This constraint logic filters out interference signals that are higher than the dynamic response boundary.

2. The automatic load tracking and intelligent adjustment system according to claim 1, characterized in that, The dynamic boundary verification module performs the following process: real-time extraction of the first derivative value of the total disturbance expansion state variable as the rate of change; the dynamic boundary verification module determines the physical feasibility of the total disturbance expansion state variable by judging whether the first derivative value is within the dynamic response slope limit of the actuator; if the first derivative value exceeds the dynamic response slope limit, the dynamic boundary verification module locks the effective load disturbance component of the current period to the historical temporary value of the previous sampling period to block non-physical pulses from entering the control loop.

3. The automatic load tracking and intelligent adjustment system according to claim 1, characterized in that, The dynamic boundary check module follows the following gating rules when outputting effective load disturbance components: Where v is the instantaneous derivative of the total disturbance expanded state variable with respect to time. The dynamic response boundary value is preset for the controlled object, and G(v) is the logic gated state factor. When G(v) is 1, the total disturbance expansion state variable is transmitted as the effective load disturbance component. When G(v) is 0, the value of the effective load disturbance component is kept constant.

4. The automatic load tracking and intelligent adjustment system according to claim 1, characterized in that, The transmission characteristic compensation table stores nonlinear compensation values ​​that characterize the actuator under different load ranges; the nonlinear compensation values ​​include mechanical backlash compensation values ​​for the commutation phase and friction torque gain for the low-speed phase. The error feedback compensation module locates the index address in the transmission characteristic compensation table based on the amplitude of the effective load disturbance component, and injects the extracted compensation amount into the control command.

5. The automatic load tracking and intelligent adjustment system according to claim 1, characterized in that, The system also includes an instruction differentiation module, which performs trajectory smoothing on the instruction signal and extracts the differential feedforward signal; the error feedback compensation module performs vector superposition of the differential feedforward signal and the effective load disturbance component to offset the dynamic lag of the controlled object in the transmission chain.

6. The automatic load tracking and intelligent adjustment system according to claim 1, characterized in that, The signal input module includes an analog signal isolator and an analog-to-digital converter; the analog signal isolator is used to obtain the induced current or feedback voltage of the controlled object through electromagnetic coupling; the analog-to-digital converter is used to convert the induced current or feedback voltage into a digital signal, and the sampling frequency is not less than 10kHz.

7. The automatic load tracking and intelligent adjustment system according to claim 1, characterized in that, The state observation module runs in the real-time controller. The state observation module adopts a third-order nonlinear observer topology and uses the deviation between the feedback signal and the internal state estimate to achieve convergence of the total disturbance extended state variable by adjusting the observer gain factor.

8. The automatic load tracking and intelligent adjustment system according to claim 1, characterized in that, The error feedback compensation module includes a stability monitoring unit, which monitors the pulsation frequency of the effective load disturbance component. If the pulsation frequency reaches more than 80% of the system closed-loop bandwidth, the stability monitoring unit automatically attenuates the injection intensity of the nonlinear compensation.

9. The automatic load tracking and intelligent adjustment system according to claim 1, characterized in that, The control commands are converted into control pulses with a frequency between 2kHz and 20kHz via a pulse width modulation interface to drive the power conversion device to adjust the input energy of the controlled object.

10. The automatic load tracking and intelligent adjustment system according to claim 1, characterized in that, The controlled parameters of the controlled object are selected from temperature data in the reactor range of 0℃ to 200℃, or torque data in the servo mechanism range of 0N·m to 500N·m. The system limits the rate of change of the control command in real time through the logical association between the effective load disturbance component and the transmission characteristic compensation table, so that it does not exceed the upper limit of the dynamic response frequency of the actuator.

Images