An intelligent remote control parameter adaptive adjustment system of an automation integrated system

Abstract

This invention relates to the field of industrial automatic control and discloses an intelligent remote control parameter adaptive adjustment system for an integrated automation system. The system includes: a state parameter acquisition unit for extracting real-time state features; a state space mapping module for mapping real-time state features to a reduced-dimensional state space image model and determining the adaptation deviation; a dynamic compensation calculation module for calculating the state evolution displacement based on the adaptation deviation, calculating the inverse compensation phase angle, and generating a dynamic envelope for the adjustment parameters; and a control signal output module for issuing control signals with phase lead. This invention achieves phase feedforward compensation of parameters by mapping the mechanical hysteresis and time delay of the power end to a reduced-dimensional state space, eliminating timing misalignment between command issuance and response, suppressing parasitic oscillations, and improving the system's stability under non-ideal network conditions.

Description

Technical Field

[0001] This invention relates to an intelligent remote control parameter adaptive adjustment system for an automated integrated system, belonging to the field of industrial automatic control technology. Background Technology

[0002] Currently, industrial automation integrated systems generally adopt a distributed architecture for remote control. This architecture uses feedback control loops to set operating parameters and maintain the stable operating state of the controlled object. In the remote control mode, the automation integrated system is affected by the communication link delay and the inherent mechanical inertia of the actuator. During the process of sending the command to the remote actuator, the transmission delay and physical inertia are superimposed. When the command acts on the physical entity, the physical reference plane of the controlled object is displaced compared with the time of command generation, forming an asymmetric misalignment between the command sending phase and the physical response phase.

[0003] Increasing the sampling frequency or using proportional-integral-derivative (PID) adjustment algorithms for compensation cannot eliminate the nonlinear disturbances caused by the aforementioned asymmetric time delay. Multi-parameter coupling interference generates nonlinear amplification in the hysteresis link, inducing parasitic oscillations. To maintain system stability, existing technologies typically reduce the adjustment gain, thereby reducing system response efficiency.

[0004] The aforementioned problems partly stem from the limitations of the underlying hardware entities, such as the rigid constraints of mechanical transmission or roller shape, which are difficult to overcome. Conventional solutions at the software communication level, such as network architecture and control methods, also have shortcomings. For example, Chinese invention patent application CN119766813A discloses a data processing method and device, a non-volatile storage medium, which divides logical partitions and creates a cloud controller in a dedicated cloud network. It uses direct-connect switches to reduce routing layers and avoid communication delays at the physical topology level. The underlying logic of the reconstructed network path control strategy implicitly presets an idealized local low-hysteresis network environment. In cross-regional long-distance industrial interconnection scenarios, the inherent latency jitter of physical links and the hysteresis of remote entity mechanical dynamic response objectively exist and are dynamically intertwined. This existing technology does not penetrate the communication and mechanical spatiotemporal coupling barriers. The data model only focuses on the rapid issuance of instructions and protocol conversion, lacking a feedforward phase offset mechanism for mechanical resistance and parasitic disturbances. When applied to highly dynamic and non-ideal network conditions, the static communication adjustment instructions and the reference plane of the remote entity's physical response produce an asymmetric misalignment, which cannot neutralize the overshoot and resonance induced by spatiotemporal hysteresis in the system.

[0005] Therefore, how to establish a coupling adjustment mechanism between the dynamic characteristics of the actuator and the characteristics of the communication link, realize the spatiotemporal offset of control parameters, and eliminate coupling interference caused by timing misalignment has become the technical problem to be solved by this invention. Summary of the Invention

[0006] To address the problems in the background art, the technical solution of the present invention is as follows: An intelligent remote control parameter adaptive adjustment system for an automated integrated system, comprising: The status parameter acquisition unit is used to acquire the real-time operating parameters of the controlled object and extract the real-time status features that characterize the operating conditions of the controlled object. The state space mapping module, whose input end is connected to the state parameter acquisition unit, is used to map real-time state features to a preset dimensionality-reduced state space image model, and calculate the topological displacement of the node migration trajectory relative to the preset steady-state trajectory based on the node migration trajectory of the real-time state features in the dimensionality-reduced state space image model, so as to determine the degree of adaptation deviation of the remote control parameters relative to the operating conditions. The dynamic compensation solution module, which is connected to the state space mapping module, is used to generate a control increment sequence based on the adaptation deviation when the adaptation deviation exceeds the preset limit and input it into the dimensionality-reduced state space mapping model. By calculating the state evolution displacement of the control increment sequence under the constraints of transmission delay and physical inertia of the actuator, the corresponding inverse compensation phase angle is calculated, and a dynamic envelope of adjustment parameters with time gradient is generated. The control signal output module is used to convert the dynamic envelope of the adjustment parameters into a control signal with a phase lead and send it to the actuator. The phase lead is used to offset the lag of the actuator and parameter coupling interference.

[0007] Preferably, the real-time status characteristics include: the output dead zone data of the actuator, the dynamic response hysteresis, and the time delay jitter sequence characterizing the transmission fluctuation of the remote communication link.

[0008] Preferably, the state space mapping module establishes a dimensionality-reduced state space image model in the following manner: obtaining the controlled physical quantity feature vector of the actuator and mapping the controlled physical quantity feature vector to the reference state node in the dimensionality-reduced state space image model; performing real-time offset modulation on the reference state node according to the real-time state characteristics to generate a topology evolution vector characterizing the system state drift trend.

[0009] Preferably, the dynamic compensation solution module follows the following judgment rules when calculating the reverse compensation phase angle: Where Δθ is the reverse compensation phase angle, To control the real-time round-trip transmission delay of signals in a remote communication link, Let be the inherent response hysteresis constant of the i-th actuator. Let represent the coupling association weight of the i-th actuator in the multi-parameter coupled system, and n be the total number of controlled actuators.

[0010] Preferably, the dynamic envelope of the adjustment parameter consists of a set of instruction sequences with time-related relationships, wherein the phase of each control point in the instruction sequence is within the leading range determined by the inverse compensation phase angle, and the amplitude change rate of the instruction sequence is limited by the rated dynamic response limit of the actuator.

[0011] Preferably, the dynamic compensation solution module is also used to identify multi-parameter coupling relationships in the automated integrated system. By orthogonalizing the evolution trajectories of different controlled variables in the dimensionality-reduced state space mapping model, the cross interference in the adjustment process of each controlled variable is eliminated.

[0012] Preferably, before issuing the control signal, the control signal output module is also used to acquire the real-time feedback signal returned by the state parameter acquisition unit, and to correct the feedback gain of the dynamic envelope of the adjustment parameter according to the deviation between the real-time feedback signal and the preset control target.

[0013] Preferably, the control signal output module includes a timing alignment unit, which adjusts the timing of the phase lead applied to the actuator according to the time marker in the dynamic envelope of the adjustment parameters, so that the phase lead cancels out the mechanical inertia of the actuator in situ.

[0014] Preferably, the system also includes a health evaluation module, which is used to obtain the performance degradation parameters of the actuator and the historical packet loss characteristics of the remote communication link, and update the calculation step size in the dimensionality reduction state space mapping model based on the performance degradation parameters and the historical packet loss characteristics.

[0015] Preferably, the system also includes a redundant safety unit, which is used to block the output of the dynamic compensation calculation module when the packet loss rate of the remote communication link exceeds the safety threshold, and lock the operating state of the actuator within a preset safety envelope range according to the historical steady-state parameters stored in the dimensionality-reduced state space image model.

[0016] Compared with the prior art, the beneficial effects of the present invention are: 1. In the adaptive adjustment of intelligent remote control parameters, by mapping the inherent mechanical hysteresis of each physical execution component to a preset computational auxiliary topology model, and combining the real-time round-trip delay of the communication link to calculate the topology bias vector, the spatiotemporal evolution trajectory of the control command can be predicted. This mapping mechanism based on topological displacement transforms the generation process of control parameters from traditional passive error correction to active phase feedforward compensation, effectively eliminating the asymmetric timing misalignment between remote command issuance and physical response, and avoiding parameter adjustment failure caused by control reference plane drift.

[0017] 2. By utilizing the dynamically generated envelope of the adjustment parameters with time gradient, a phase lead component is preset at the control end, so that it offsets mechanical resistance and parasitic interference in situ when it reaches the physical execution end. This mechanism cuts off the nonlinear amplification path of multi-parameter coupling interference in the hysteresis link from the physical level, suppresses parasitic oscillations and system overshoot that are very easy to occur during remote control, and improves the operational stability of the controlled object under non-ideal network conditions and complex mechanical inertia conditions.

[0018] 3. The system's operational status is collected in all dimensions, its status features are extracted, its adaptability is determined, and its multi-parameter coupled calculations are mutually supportive. This constructs an adaptive iterative adjustment system based on a virtual computation kernel. By using computation-aided spatiotemporal evolution deduction within the topology model, it provides a mapping basis with physical and dynamic characteristics for multi-parameter adjustment, enhancing the deep matching between remote control commands and the operational status of underlying equipment. This solves the technical bottleneck of traditional static adjustment methods being unable to adapt to dynamic environments. Through the closed-loop linkage of adaptive adjustment of control parameters, execution of commands, and verification of operational status feedback, the system uses inverse phase calculation to dynamically correct the adjustment parameters. This makes the physical characteristics of the sent signals and the dynamic characteristics of the actuators highly cohesive. This closed-loop adjustment logic significantly reduces the lag in remote control parameter adjustment, ensuring that the automated integrated system maintains high-precision parameter adaptation capabilities and system response performance even in environments with network fluctuations and physical execution resistance. Attached Figure Description

[0019] Figure 1 This is a flowchart illustrating the overall architecture and parameter adjustment process of the automated control system of the present invention. Figure 2 This is a diagram illustrating the data evolution and topological displacement analysis of the state-space mapping model of this invention.

[0020] The objectives, features, and advantages of this invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0021] 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.

[0022] An intelligent remote control parameter adaptive adjustment system for an automated integrated system includes: The status parameter acquisition unit is used to acquire the real-time operating parameters of the controlled object and extract the real-time status features that characterize the operating conditions of the controlled object. The state space mapping module, whose input end is connected to the state parameter acquisition unit, is used to map real-time state features to a preset dimensionality-reduced state space image model, and calculate the topological displacement of the node migration trajectory relative to the preset steady-state trajectory based on the node migration trajectory of the real-time state features in the dimensionality-reduced state space image model, so as to determine the degree of adaptation deviation of the remote control parameters relative to the operating conditions. The dynamic compensation solution module, which is connected to the state space mapping module, is used to generate a control increment sequence based on the adaptation deviation when the adaptation deviation exceeds the preset limit and input it into the dimensionality-reduced state space mapping model. By calculating the state evolution displacement of the control increment sequence under the constraints of transmission delay and physical inertia of the actuator, the corresponding inverse compensation phase angle is calculated, and a dynamic envelope of adjustment parameters with time gradient is generated. The control signal output module is used to convert the dynamic envelope of the adjustment parameters into a control signal with a phase lead and send it to the actuator. The phase lead is used to offset the lag of the actuator and parameter coupling interference.

[0023] Preferably, the real-time status characteristics include: the output dead zone data of the actuator, the dynamic response hysteresis, and the time delay jitter sequence characterizing the transmission fluctuation of the remote communication link.

[0024] Preferably, the state space mapping module establishes a dimensionality-reduced state space image model in the following manner: obtaining the controlled physical quantity feature vector of the actuator and mapping the controlled physical quantity feature vector to the reference state node in the dimensionality-reduced state space image model; performing real-time offset modulation on the reference state node according to the real-time state characteristics to generate a topology evolution vector characterizing the system state drift trend.

[0025] Preferably, the dynamic compensation solution module follows the following judgment rules when calculating the reverse compensation phase angle: Where Δθ is the reverse compensation phase angle, To control the real-time round-trip transmission delay of signals in a remote communication link, Let be the inherent response hysteresis constant of the i-th actuator. Let represent the coupling association weight of the i-th actuator in the multi-parameter coupled system, and n be the total number of controlled actuators.

[0026] Preferably, the dynamic envelope of the adjustment parameter consists of a set of instruction sequences with time-related relationships, wherein the phase of each control point in the instruction sequence is within the leading range determined by the inverse compensation phase angle, and the amplitude change rate of the instruction sequence is limited by the rated dynamic response limit of the actuator.

[0027] Preferably, the dynamic compensation solution module is also used to identify multi-parameter coupling relationships in the automated integrated system. By orthogonalizing the evolution trajectories of different controlled variables in the dimensionality-reduced state space mapping model, the cross interference in the adjustment process of each controlled variable is eliminated.

[0028] Preferably, before issuing the control signal, the control signal output module is also used to acquire the real-time feedback signal returned by the state parameter acquisition unit, and to correct the feedback gain of the dynamic envelope of the adjustment parameter according to the deviation between the real-time feedback signal and the preset control target.

[0029] Preferably, the control signal output module includes a timing alignment unit, which adjusts the timing of the phase lead applied to the actuator according to the time marker in the dynamic envelope of the adjustment parameters, so that the phase lead cancels out the mechanical inertia of the actuator in situ.

[0030] Preferably, the system also includes a health evaluation module, which is used to obtain the performance degradation parameters of the actuator and the historical packet loss characteristics of the remote communication link, and update the calculation step size in the dimensionality reduction state space mapping model based on the performance degradation parameters and the historical packet loss characteristics.

[0031] Preferably, the system also includes a redundant safety unit, which is used to block the output of the dynamic compensation calculation module when the packet loss rate of the remote communication link exceeds the safety threshold, and lock the operating state of the actuator within a preset safety envelope range according to the historical steady-state parameters stored in the dimensionality-reduced state space image model.

[0032] Example 1: In a distributed multi-axis collaborative manufacturing production line scenario deployed across regions, the controlled object includes multiple synchronous servo drive axes and their corresponding actuators. The state parameter acquisition unit continuously acquires the real-time operating parameters of the controlled object at a sampling period of 10ms, extracting real-time state features characterizing the operating condition of the controlled object. These real-time state features include the output dead zone data of the actuator, the dynamic response hysteresis, and the time delay jitter sequence characterizing the transmission fluctuations of the remote communication link. Due to the physical distance between the main control center and the edge execution nodes, the time delay generated by the network transmission of the issued commands is superimposed on the mechanical climbing inertia of the servo motor, causing the physical reference plane of the controlled object to drift when the control parameters reach the physical entity. This induces parasitic oscillations at the multi-axis linkage coupling point and reduces automation. To ensure the stability of the integrated system operation, the state space mapping module maps the aforementioned real-time state characteristics to a preset dimensionality-reduced state space image model. By acquiring the controlled physical quantity feature vector of the actuator and mapping it to a reference state node in the dimensionality-reduced state space image model, real-time offset modulation is applied to the reference state node based on the real-time state characteristics to generate a topology evolution vector characterizing the system state drift trend. Based on the node migration trajectory of the real-time state characteristics in the dimensionality-reduced state space image model, the topology displacement of the node migration trajectory relative to the preset steady-state trajectory is calculated to determine the degree of adaptation deviation of the remote control parameters relative to the operating conditions. This topology displacement quantitatively characterizes the geometric displacement characteristics of the multivariable coupling time delay in the dimensionality-reduced topology space, constituting the physical and dynamic input basis for parameter phase compensation.

[0033] When the adaptation deviation is determined to exceed a preset limit of 5% based on the sensor feedback data, the dynamic compensation calculation module generates a control increment sequence based on the adaptation deviation and inputs it into the dimensionality-reduced state space mapping model. It then calculates the state evolution displacement of the control increment sequence under the constraints of transmission delay and the physical inertia of the actuator, following the formula... Calculate the corresponding reverse compensation phase angle, where Δθ is the reverse compensation phase angle. To control the real-time round-trip transmission delay of signals in a remote communication link, Let be the inherent response hysteresis constant of the i-th actuator. Let be the coupling association weight of the i-th actuator in the multi-parameter coupled system, and n be the total number of controlled actuators. This generates a dynamic envelope of adjustment parameters with a time gradient. When generating this envelope, the control terminal does not issue a single global phase offset, but instead establishes a 256-byte instruction ring buffer for each of the three actuators involved in the linkage. The system compensates for the phase angle inversely based on the calculated 0.42 radians, combined with the specific inherent response hysteresis constants of axes 1, 2, and 3 (e.g., 18.5 ms for axis 1, 19.2 ms for axis 2). s) Differentiated timing shifts are performed within the buffer to create millisecond-level misalignment in the micro-timing of each axis command, ensuring precise synchronization when the compensation signal reaches each heterogeneous physical terminal. The dynamic envelope of the adjustment parameter consists of a set of timing-related command sequences. The phase of each control point in the command sequence is within the leading range determined by the inverse compensation phase angle, and the amplitude change rate of the command sequence is limited by the rated dynamic response limit of the actuator. This amplitude change rate limiting logic works in conjunction with the phase leading component to achieve the dual goals of response hysteresis compensation and control accuracy maintenance.

[0034] The calculated inverse compensation phase angle is converted into an executable action of the physical servo mechanism. The control signal output module has a built-in position command interpolation unit. Based on the evolution law of phase and time correlation in rigid body kinematics, the position command interpolation unit calculates the time lead feedforward by sampling the real-time position and given the operating angular frequency. The calculation formula is: , The angular frequency of the real-time motion command for the servo drive axis is given, and Δθ is the inverse compensation phase angle. To calculate the time advance feedforward amount, after obtaining the time advance feedforward amount, the position command interpolation unit will shift the absolute value of the time advance feedforward amount forward on the time axis in the next synchronization cycle of the control bus, and generate an additional feedforward speed bias signal that is directly superimposed on the input of the driver speed loop. The mathematical phase angle in the topology calculation space is objectively transformed into the timing drift cancellation action on the underlying electrical control bus.

[0035] The control signal output module converts the dynamic envelope of the adjustment parameters into a control signal with a phase lead and sends it to the actuator. The timing alignment unit inside the control signal output module adjusts the timing of the phase lead applied to the actuator based on the time marker in the dynamic envelope of the adjustment parameters, so that the phase lead cancels out the mechanical inertia of the actuator in situ. At the same time, the dynamic compensation solution module identifies the multi-parameter coupling relationship in the automated integrated system. By applying orthogonalization operations to the evolution trajectories of different controlled variables in the reduced state space image model, it eliminates the cross interference in the adjustment process of each controlled variable, and maintains the zero-phase deviation adjustment state of the automated integrated system under network conditions with time delay jitter and actuator physical inertia.

[0036] Example 2: In the remote control of multi-axis collaborative manufacturing in industrial applications, a hardware-in-the-loop semi-physical simulation test environment is built between the master control center and edge servo nodes. This environment includes a synchronous servo drive axis array with a sampling period of 10ms. Gaussian white noise with a signal-to-noise ratio of 20dB and a time delay jitter sequence with a random amplitude fluctuating between 45.3ms and 142.7ms are actively superimposed on the master-slave communication link. At the same time, 50Hz power frequency resonance interference is introduced at the mechanical output end to restore the fluctuating network and mechanical states in the objective engineering site. The 10ms sampling period is set to balance the real-time capture of the state and the computational load of the dimensionality reduction state space mapping model. When extracting the operating conditions of the controlled object, if the out-of-band white noise energy of the communication link jumps, the sampling period tends to the lower limit of the value to prevent low-frequency signal aliasing distortion under the limitation of the Nyquist sampling theorem.

[0037] In the test environment startup state, the state parameter acquisition unit continuously acquires real-time operating parameters including output dead zone data and dynamic response hysteresis. In the initial stage of the test, the raw operating data without phase compensation shows an oscillating and spreading trend. The superposition of communication delay and mechanical climbing inertia causes the characteristic vector of the controlled physical quantity to deviate from the baseline trajectory. The measured initial phase hysteresis reaches 0.35 radians. The state space mapping module maps the real-time state characteristics to the dimensionality-reduced state space image model and calculates the topological displacement of the node migration trajectory relative to the preset steady-state trajectory. Monitoring data feedback shows that when the communication delay reaches 85.4ms, the adaptation deviation jumps to 7.2%, exceeding the preset limit of 5%. The dynamic compensation solution module takes in this adaptation deviation, generates a control increment sequence and inputs it into the dimensionality-reduced state space image model, and calculates that the inherent response hysteresis constant stabilizes in the range of 18.5ms to 19.2ms. Before the dimensionality-reduced state space image model performs parameter calculations, it executes an online physical parameter decoupling procedure based on the principle of multi-source signal frequency domain separation. The data acquisition unit synchronously records the total operating current of the servo motor and the round-trip timestamp of the remote communication link at a sampling period of 1ms. The dynamic compensation and calculation module extracts the low-frequency envelope of the total operating current data within 50 consecutive sampling periods, calculates the first time difference between the low-frequency envelope and the master control position command, and extracts the statistical average of the round-trip timestamp sequence as the second time difference. In order to achieve frequency separation at the physical level, the dynamic compensation and calculation module has a built-in sliding average filter with a length of 50 sampling points. This filter slides in a fixed step of 1ms. By accumulating the 50 current sampling values ​​in the current buffer and dividing by 50, the real-time low-frequency envelope is calculated, thereby filtering out mechanical oscillation interference with frequencies higher than 200Hz. When the value fluctuation of the low-frequency envelope within 10 consecutive filtering periods is less than 0.05mA, the system locks this average value as the extracted pure mechanical hysteresis component and assigns it a value. The dynamic compensation and calculation module subtracts the second time difference from the first time difference to separate the pure mechanical hysteresis component caused only by physical friction and mechanical inertia, and assigns it to the inherent response hysteresis constant. By employing synchronous sampling and differential operation mechanisms, the strong coupling interference between network latency and mechanical resistance is isolated at the objective measurement level. Based on this, the dynamic compensation solution module calculates the solution according to the formula... Calculate the reverse compensation phase angle, and at this moment, the real-time round-trip transmission delay. Recorded 103.8ms, multi-parameter coupling correlation weights were introduced. After completing the orthogonal solution, the output inverse compensation phase angle Δθ is 0.42 radians, generating a dynamic envelope of adjustment parameters that includes the phase lead.

[0038] A comparative sample group using a traditional fixed delay compensation scheme and an experimental sample group using the aforementioned state topology mapping scheme were established. Multi-gradient delay immunity tests were initiated to demonstrate the system performance evolution. When the link jitter delay injection was maintained in the low-order range of 50.0 ms, the inter-axis synchronization difference of both the comparative and experimental sample groups converged to within 0.05 radians. As the jitter delay injection gradually increased beyond 120.0 ms, the system reached a nonlinear performance inflection point. The synchronization difference of the comparative sample group abruptly increased to 1.25 radians, triggering continuous mechanical resonance. This corresponds to its static compensation mechanism in the surface... For logic failures at the mechanical inertia saturation boundary, the timing alignment unit inside the test sample continuously sends control signals with dynamic phase lead based on the time marker in the dynamic envelope of the adjustment parameters. By limiting the amplitude change rate, the logic avoids transient overload of the actuator. Test data confirms that under the extreme jitter delay of 142.7ms and the combined interference of Gaussian noise, the inter-axis synchronous phase difference of the test sample stably converges to 0.12 radians, verifying the ability of the topological displacement calculation mechanism to neutralize nonlinear spatiotemporal hysteresis and maintain the control accuracy of the automated integrated system at the physical level.

[0039] Example 3: In a multi-axis collaborative manufacturing pipeline scenario, when constructing a dimensionality-reduced state-space mapping model for the master control center, the direct mapping of the feature vectors of the multi-source heterogeneous controlled physical quantities causes data sparsity and topological distortion in the high-dimensional space, resulting in the lack of a quantitative geometric benchmark for the adaptation deviation calculation. The state parameter acquisition unit acquires a historical operating dataset containing the position feedback time series, speed loop integral saturation, and torque current pulsation amplitude of the synchronous servo drive shaft under calibration conditions. The state-space mapping module extracts the principal components of the historical operating dataset based on the principal component analysis algorithm, retains the orthogonal basis vectors with a cumulative variance contribution rate greater than 95%, and constructs a three-dimensional low-order projection matrix. The state-space mapping module uses this low-order projection matrix to reduce the dimensionality of the feature vectors of the controlled physical quantities and map them to the benchmark state nodes, establishing a manifold topological surface covering the entire envelope operating range of the controlled object.

[0040] The dynamic compensation solution module establishes the coupling correlation weights in the multi-parameter coupled system based on the manifold topological surface. It extracts the directional derivative vector of the eigenvalue of the i-th actuator's principal component in the low-order projection matrix, calculates the absolute value of the inner product between this directional derivative vector and a preset reference vector, and forms the corresponding correlation scalar. The dynamic compensation solution module calculates according to the formula. Determine the coupling association weights, where Let be the coupling correlation weight of the i-th actuator in the multi-parameter coupled system. Let i be the correlation scalar corresponding to the i-th actuator. Let k be the correlation scalar corresponding to the k-th actuator, k be the summation index, and n be the total number of controlled actuators.

[0041] The state space mapping module extracts real-time state features and maps them onto the topological surface of the manifold, generating real-time mapped coordinate points. The state space mapping module calculates the geodesic distance between the mapped coordinate point and the nearest neighbor reference state node on the preset steady-state trajectory, and assigns the geodesic distance as the topological displacement. The above-mentioned dimensionality reduction projection and geodesic distance calculation process transforms the multivariable coupled system state drift into geometric displacement variables in the low-dimensional manifold space, forming the physical quantization input reference for the dynamic compensation solution module to generate the dynamic envelope of the adjustment parameters.

[0042] Example 4: In a newly deployed automated manufacturing production line scenario including multi-axis linkage modules, the main control center initiates a pre-baseline calibration procedure before issuing business control instructions to the controlled object. This establishes the objective geometric reference origin within the reduced-dimensional state space mapping model. The state parameter acquisition unit continuously acquires 5000 cycles of position feedback time series at a 10ms sampling period under no-load operating conditions, outputting dead zone data and inherent dynamic response hysteresis. The state space mapping module projects this no-load historical operating dataset onto a low-order projection matrix using principal component analysis to generate an initial manifold coordinate cluster. A mean-shift clustering algorithm is then used to extract the high-density distribution center lines of this initial manifold coordinate cluster and lock them as a preset steady-state trajectory. This pre-calibration process is based on the physical operation feedback of the controlled object. Feedback data corrects the static deviation caused by the assembly tolerance of mechanical transmission components in the calculation of topological displacement. When constructing the dimension-reduced projection matrix, the system collects 5000 sets of original running vectors containing position feedback, velocity saturation, and torque current during the initialization phase. By accumulating the variance, the weight coefficients of the three orthogonal principal components are determined to be 0.72, 0.19, and 0.09, respectively. The 3D features are then projected onto a 2D plane. The reference origin of the projection space is locked as the cluster center vector of the three execution axes at the theoretical zero point position with a synchronization deviation of 0. When calculating the geodesic distance, the system uses an 8-neighborhood linear interpolation algorithm with a minimum search step of 0.1mm to find the nearest neighbor node. By accumulating the sum of the Euclidean distances between the current mapped coordinate point and all adjacent nodes on the preset steady-state trajectory, the geometric displacement reflecting the drift of the physical trajectory is obtained.

[0043] Under no-load operation, the dynamic compensation solution module calculates the baseline topological displacement of each reference state node in the initial manifold coordinate family, deviating from the preset steady-state trajectory. The ratio of this baseline topological displacement to the nominal total length of the preset steady-state trajectory is converted into a normalized displacement coefficient sequence. The statistical distribution characteristics of this normalized displacement coefficient sequence are extracted to calibrate the dynamic response threshold baseline. By performing a skewness test on 10,000 sets of historical normalized displacement data, it is confirmed that the distribution exhibits a symmetrical unimodal characteristic with an absolute skewness value less than 0.1, satisfying Gaussian distribution characteristics. This ensures the statistical effectiveness of the 3-standard-deviation criterion in identifying abnormal states. According to the formula... Determine the preset limit for measuring the degree of adaptation deviation, where As a preset limit, This represents the statistical mean of the normalized displacement coefficient sequence. To normalize the standard deviation of the displacement coefficient sequence, this preset limit sets the quantization isolation boundary for mechanical fluctuations and spatiotemporal hysteresis state drift of the underlying hardware of the automated integrated system. The control signal output module switches the system to a parameter adaptive adjustment state based on the locked preset steady-state trajectory and preset limit. During the loaded service phase, the automated integrated system filters out erroneous compensation commands induced by random noise in the communication link according to the pre-calibrated quantization boundary. In actual operation, the actuator experiences mechanical inertia shift due to process load fluctuations. The state space mapping module includes an online baseline reconstruction mechanism. During the loaded service phase, the state parameter acquisition unit continuously monitors the real-time rotation of the actuator for 10 control cycles. When the torque current output value deviates from the baseline torque current during the previous no-load calibration by more than 15%, the online adaptive update procedure is triggered. The dynamic compensation calculation module records the current load torque range and calculates the additional mechanical ramp time caused by the current load compared to the no-load condition based on the linear relationship between torque and angular acceleration in Newton's second law. The state space mapping module multiplies the additional mechanical ramp time as a characteristic scaling factor by the geodesic span of the initial manifold coordinate cluster to generate the corresponding loaded manifold coordinate cluster for the current actual load condition, which replaces the preset steady-state trajectory in real time. This overcomes the physical boundary limitations of a single no-load calibration and maintains dynamic alignment between the compensation baseline and the physical reality during long-term loaded operation.

[0044] Example 5: Before a distributed multi-axis collaborative manufacturing production line deployed across regions is connected to the actual production load, the main control center sets the amplitude variation constraint boundary of the actuator control command. The state parameter acquisition unit continuously injects a sequence of step excitation signals with amplitudes increasing by a preset step size into the input end of the actuator, and simultaneously acquires the mechanical displacement feedback curve at the output end of the actuator. The actual rise time, steady-state error, and overshoot corresponding to each step excitation signal in the mechanical displacement feedback curve are extracted. When it is determined that the overshoot of the mechanical displacement feedback curve is greater than 5% or the steady-state error is greater than the preset process tolerance threshold, the state parameter acquisition unit extracts the amplitude of the previous step excitation signal that triggered the above boundary state and the actual rise time corresponding to the previous step excitation signal. The quotient of the amplitude of the previous step excitation signal and the corresponding actual rise time is calculated and established as the rated dynamic response limit of the actuator. This operation constructs a quantitative data source for adjusting the slope of the constraint remote control command based on the on-site mechanical characteristics of the physical equipment.

[0045] The dynamic compensation solution module obtains the rated dynamic response limit. It is set as the highest-order saturation constraint parameter for generating the control increment sequence. When the dynamic compensation solution module outputs the dynamic envelope of the adjustment parameter, it continuously calculates the quotient of the difference in parameter amplitude between adjacent control points in the command sequence and the time interval. When it detects that the quotient is equal to the rated dynamic response limit... At that time, the dynamic compensation solution module limits the continued increase of the amplitude of the control increment sequence and correspondingly extends the action time window of the compensation parameter. This constraint logic defines the limit adjustment slope of the remote control command based on the objective load feedback of the actuator, so that the parameter adaptive adjustment state suppresses the mechanical excitation and hardware overload induced by the sudden change of parameter amplitude when outputting the phase compensation signal.

[0046] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0047] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. An intelligent remote control parameter adaptive adjustment system for an automated integrated system, characterized in that, include: The status parameter acquisition unit is used to acquire the real-time operating parameters of the controlled object and extract the real-time status features that characterize the operating conditions of the controlled object. The state space mapping module, whose input end is connected to the state parameter acquisition unit, is used to map real-time state features to a preset dimensionality-reduced state space image model, and calculate the topological displacement of the node migration trajectory relative to the preset steady-state trajectory based on the node migration trajectory of the real-time state features in the dimensionality-reduced state space image model, so as to determine the degree of adaptation deviation of the remote control parameters relative to the operating conditions. The dynamic compensation solution module, which is connected to the state space mapping module, is used to generate a control increment sequence based on the adaptation deviation when the adaptation deviation exceeds the preset limit and input it into the dimensionality-reduced state space mapping model. By calculating the state evolution displacement of the control increment sequence under the constraints of transmission delay and physical inertia of the actuator, the corresponding inverse compensation phase angle is calculated, and a dynamic envelope of adjustment parameters with time gradient is generated. The control signal output module is used to convert the dynamic envelope of the adjustment parameters into a control signal with a phase lead and send it to the actuator. The phase lead is used to offset the lag of the actuator and parameter coupling interference.

2. The intelligent remote control parameter adaptive adjustment system for an automated integrated system according to claim 1, characterized in that, Real-time status characteristics include: the output dead zone data of the actuator, the dynamic response hysteresis, and the time delay jitter sequence characterizing the transmission fluctuation of the remote communication link.

3. The intelligent remote control parameter adaptive adjustment system for an automated integrated system according to claim 1, characterized in that, The state space mapping module establishes a dimensionality-reduced state space mapping model in the following way: obtain the controllable physical quantity feature vector of the actuator, and map the controllable physical quantity feature vector to the reference state node in the dimensionality-reduced state space mapping model; perform real-time offset modulation on the reference state node according to the real-time state characteristics to generate a topology evolution vector characterizing the system state drift trend.

4. The intelligent remote control parameter adaptive adjustment system for an automated integrated system according to claim 1, characterized in that, When calculating the reverse compensation phase angle, the dynamic compensation solution module follows these judgment rules: Where Δθ is the reverse compensation phase angle, To control the real-time round-trip transmission delay of signals in a remote communication link, Let be the inherent response hysteresis constant of the i-th actuator. Let represent the coupling association weight of the i-th actuator in the multi-parameter coupled system, and n be the total number of controlled actuators.

5. The intelligent remote control parameter adaptive adjustment system for an automated integrated system according to claim 1, characterized in that, The dynamic envelope of the adjustment parameters consists of a set of time-related instruction sequences. The phase of each control point in the instruction sequence is within the leading range determined by the inverse compensation phase angle, and the rate of change of the amplitude of the instruction sequence is limited by the rated dynamic response limit of the actuator.

6. The intelligent remote control parameter adaptive adjustment system for an automated integrated system according to claim 1, characterized in that, The dynamic compensation solution module is also used to identify multi-parameter coupling relationships in automated integrated systems by orthogonalizing the evolution trajectories of different controlled variables in a dimensionality-reduced state-space mapping model.

7. The intelligent remote control parameter adaptive adjustment system for an automated integrated system according to claim 1, characterized in that, Before issuing control signals, the control signal output module is also used to acquire real-time feedback signals transmitted back by the status parameter acquisition unit, and to correct the feedback gain of the dynamic envelope of the adjustment parameters based on the deviation between the real-time feedback signals and the preset control target.

8. The intelligent remote control parameter adaptive adjustment system for an automated integrated system according to claim 1, characterized in that, The control signal output module includes a timing alignment unit, which adjusts the timing of the phase lead applied to the actuator according to the time marker in the dynamic envelope of the adjustment parameters, so that the phase lead cancels out the mechanical inertia of the actuator in situ.

9. The intelligent remote control parameter adaptive adjustment system for an automated integrated system according to claim 1, characterized in that, The system also includes a health assessment module, which is used to obtain the performance degradation parameters of the actuator and the historical packet loss characteristics of the remote communication link, and update the calculation step size in the dimensionality reduction state space mapping model based on the performance degradation parameters and historical packet loss characteristics.

10. The intelligent remote control parameter adaptive adjustment system for an automated integrated system according to claim 1, characterized in that, The system also includes a redundant safety unit, which is used to shield the output of the dynamic compensation solution module when the packet loss rate of the remote communication link exceeds the safety threshold, and lock the operating state of the actuator within the preset safety envelope range based on the historical steady-state parameters stored in the dimensionality-reduced state space image model.

Images