Method and system for motion positioning control based on flexible magnetic drive conveying device

Abstract

The application relates to a motion positioning control method and system based on a flexible magnetic drive conveying device, and belongs to the technical field of magnetic drive conveying positioning. The prior art adopts a fixed-parameter zero-vibration shaper, which cannot adapt to the dynamic change of resonance characteristics under the high-speed multi-mover scene, and is prone to cause positioning overshoot and residual oscillation. The method is divided into a preparation stage and an execution stage. In the preparation stage, an adaptive suppression signal is determined based on historical steady-state motion data, which is used to suppress the self-oscillation and position overshoot of a single mover. A global coupling disturbance prediction model is constructed and trained by combining device structure parameters and the adaptive suppression signal, and a cooperative prediction compensation signal is generated to offset the mechanical coupling vibration between movers. In the execution stage, the two types of compensation signals are sequentially superimposed on the initial positioning command to obtain a cooperative positioning command and are sent to the stator driving unit. The application can simultaneously suppress the self-vibration of the single body and the cross-mover coupling disturbance, and effectively improve the positioning accuracy and running stability of the multi-axis mover under high-speed working conditions.

Description

Technical Field

[0001] This application relates to the field of motion positioning technology for flexible magnetic drive conveying systems, and in particular to a motion positioning control method and system based on a flexible magnetic drive conveying device. Background Technology

[0002] Flexible magnetic drive conveyor is a new type of industrial automation equipment that uses magnetic levitation conveying technology to position several moving parts on a transport track to complete complex conveying tasks such as workpiece processing, inspection, and diversion. During the transport of workpieces, it is necessary to pre-define appropriate operating planning parameters for each moving part, that is, to clarify the running speed, acceleration, and other indicators of each moving part at different times, so as to control all moving parts to run smoothly along the preset trajectory.

[0003] Currently, to achieve mover positioning in flexible magnetic drive conveyor systems, a cascaded PID feedback and feedforward compensation method based on independent control of a single mover is commonly used. This method consists of a position loop, a velocity loop, and a current loop forming a three-loop cascaded structure, supplemented by velocity feedforward and acceleration feedforward based on an ideal model. This suppresses the basic position deviation, velocity fluctuation, and acceleration lag of the mover during magnetic drive conveyor operation, and simultaneously compensates for the system's inertial resistance and steady-state thrust error. The industry commonly uses a zero-vibration shaper, which works by convolving the original step position command with a set of pulse sequences with fixed time intervals and amplitudes to generate a smooth command that can self-cancel out the system's resonant modes. Theoretically, this achieves oscillation-free mover positioning and improves the motion stability of the mover under normal operating conditions.

[0004] However, in the actual application scenario of high-speed multi-mover high-precision positioning, any mover executing the positioning command at high speed will generate a reverse force when it completes the acceleration and deceleration action. This force directly acts on the stator track and the equipment base, thereby exciting structural vibration. Such vibration can be transmitted to other movers along the conveyor track, forming external disturbances with strong randomness and large dynamic changes, causing the system's resonant frequency to change in real time.

[0005] Traditional zero-vibration shapers have fixed built-in control parameters and can only match a single fixed resonant frequency. They cannot adaptively adjust parameters according to the actual working conditions on site. When this method is directly applied to the above-mentioned high-speed and high-precision moving part motion scenario, the moving part at the positioning end is easily affected by the structural vibration transmitted by adjacent moving parts, resulting in overshoot and residual oscillation at the positioning end, which seriously weakens the operational stability of high-precision positioning of moving parts during multi-moving part collaborative conveying. Summary of the Invention

[0006] To improve the poor operational stability of high-precision positioning of movers during multi-motor cooperative conveying in high-speed and high-precision scenarios, this application provides a motion positioning control method and system based on a flexible magnetic drive conveying device.

[0007] Firstly, this application provides a motion positioning control method based on a flexible magnetic drive conveying device, employing the following technical solution:

[0008] A motion positioning control method based on a flexible magnetic drive conveyor, comprising a multi-axis mover positioning control system, the system including a stator drive unit for driving a corresponding mover along a preset stator track to complete positioning according to instructions; the method includes an initial preparation stage and an intermediate execution stage, the initial preparation stage including:

[0009] Obtain the current location requirement information and parse the current location requirement information to obtain the initial location command;

[0010] Acquire steady-state motion data of the target mover in at least one historical mission cycle, and determine the adaptive suppression signal of the target mover based on the steady-state motion data. The adaptive suppression signal is a compensation signal used to suppress the self-positioning oscillation and position overshoot of the single mover.

[0011] Obtain the pre-stored equipment structure parameters, construct and train a global coupling interference prediction model based on the adaptive suppression signal, and generate a collaborative prediction compensation signal according to the global coupling interference prediction model. The collaborative prediction compensation signal is a reverse compensation signal used to cancel the mechanical coupling vibration interference transmitted between several movers along the track.

[0012] The method includes the following in the intermediate execution phase:

[0013] The initial positioning command and the adaptive suppression signal are obtained, and the adaptive suppression signal is superimposed on the initial positioning command to obtain the process positioning command.

[0014] The cooperative prediction compensation signal is acquired, and the cooperative prediction compensation signal is superimposed on the process positioning command to obtain the cooperative positioning command. The cooperative positioning command is then sent to the corresponding stator drive unit.

[0015] By adopting the above technical solution, this application divides the early preparation stage and the mid-term execution stage to achieve layered and precise vibration suppression and positioning control: the early preparation stage generates an adaptive suppression signal offline based on the historical steady-state motion data of the mover, which can specifically suppress the positioning oscillation and end overshoot caused by the resonance of the single mover itself, effectively overcoming the shortcomings of the traditional fixed parameter shaping method that cannot adapt to the dynamic changes of working conditions and the poor single vibration suppression effect.

[0016] Simultaneously, based on the inherent structural parameters of the equipment and combined with adaptive suppression signals, the global coupling interference prediction model is constructed and trained, which can accurately quantify the mechanical coupling vibration disturbance transmitted between multiple movers along the track, providing reliable data and model support for global vibration compensation.

[0017] The intermediate execution stage corrects the initial positioning command by superimposing two-level compensation signals layer by layer. This not only greatly eliminates the motion deviation of the single mover body from the source of the command, but also offsets the coupling vibration interference generated by the interaction of multiple movers to a certain extent. It suppresses the residual oscillation and overshoot phenomenon at the end of the positioning, effectively solving the technical pain point of poor positioning and operation stability caused by the dynamic resonance and cross-axis coupling disturbance of the mover in the high-speed multi-motor cooperative conveying scenario. It can significantly improve the accuracy and stability of high-speed cooperative positioning of multi-axis movers in flexible magnetic drive conveying devices.

[0018] In one specific implementation scheme, the steady-state motion data includes historical position error signals and historical acceleration response signals collected during the final positioning phase of the target mover in a historical mission cycle; determining the adaptive suppression signal of the target mover based on the steady-state motion data includes:

[0019] The historical position error signal is subjected to time-domain feature extraction to obtain the historical resonance response waveform of the target mover at the end of the historical mission cycle positioning stage. The historical resonance response waveform refers to the oscillation curve of the distance deviation between the actual position and the target position of the target mover as a function of time at the end of the historical mission cycle positioning stage.

[0020] The historical resonant response waveform is inverted to obtain the initial suppression signal;

[0021] Obtain the maximum allowable positioning error parameter for the current positioning task of the target mover, and limit the amplitude of the initial suppression signal according to the maximum allowable positioning error parameter to obtain an adaptive suppression signal.

[0022] By adopting the above technical solution, this application relies on the historical position error signal and historical acceleration response signal of the mover positioning end in the historical mission cycle to perform data analysis, accurately extract the historical resonance response waveform of the positioning end, and truly reproduce the residual oscillation and overshoot characteristics in the steady-state positioning process of the mover:

[0023] Specifically, the initial suppression signal is obtained by inverting the phase of the resonant response waveform, which can generate a compensation waveform with the same amplitude and opposite direction to the body's resonant vibration, thus achieving vibration cancellation from a mechanistic perspective. At the same time, the amplitude of the initial suppression signal is limited by the maximum allowable positioning error parameter of the equipment to avoid excessive compensation signal amplitude causing out-of-tolerance and jitter problems, ensuring that the compensation process is always effectively executed within the allowable accuracy range.

[0024] This method adaptively generates suppression signals based on the real historical oscillation characteristics of the mover, which is different from the generalized compensation mode with fixed shaping parameters in the traditional method. It can accurately adapt to the changes in the resonance characteristics of a single unit under different loads and operating conditions, greatly improving the pertinence and effectiveness of the self-suppression of the single mover end. It weakens the residual oscillation of the single unit positioning and the end overshoot from the source, laying a high-precision operating foundation for subsequent multi-motor coupled vibration compensation.

[0025] In a specific implementation scheme, the motion process of the target mover in the total travel of a single positioning task is divided into a rapid approach segment and a precise stopping segment. The rapid approach segment refers to the stage in which the target mover moves rapidly from the initial position to the target position with maximum acceleration or a relatively large speed. The precise stopping segment refers to the stage from when the target mover enters the preset tolerance error band until it stops stably at the target position. The end of the rapid approach segment corresponds to the moment when the target mover first enters the tolerance error band centered on the target position and with the maximum allowable positioning error as the radius.

[0026] The steps of superimposing the adaptive suppression signal onto the initial positioning command to obtain the process positioning command, and the steps of superimposing the cooperative prediction compensation signal onto the process positioning command to obtain the cooperative positioning command, are performed in the precision stop segment.

[0027] In one specific implementation, the method further includes, after acquiring steady-state motion data of the target mover in at least one historical mission cycle during the preliminary preparation phase:

[0028] Offline frequency analysis was performed on the steady-state motion data to obtain the estimated resonant frequency band;

[0029] Before determining the adaptive suppression signal of the target mover based on the steady-state motion data in the intermediate execution phase, the following is also included:

[0030] The measured resonant frequency band and the pre-stored estimated resonant frequency band are obtained, and the measured resonant frequency band and the estimated resonant frequency band are compared to obtain the frequency band deviation.

[0031] Determine whether the frequency band deviation is within a preset allowable deviation range;

[0032] If the judgment result is yes, then the pre-stored adaptive suppression signal is superimposed on the initial positioning command to obtain the process positioning command;

[0033] If the judgment result is negative, the pre-stored adaptive suppression signal is corrected in real time according to the frequency band deviation, and the corrected adaptive suppression signal is superimposed on the initial positioning command to obtain the process positioning command.

[0034] By adopting the above technical solution, this application divides the single positioning stroke of the mover into a rapid approach segment and a precision stop segment. Adaptive suppression signal and collaborative prediction compensation signal are superimposed only in the precision stop segment, which is prone to overshoot and residual oscillation. There is no additional compensation signal interference in the rapid approach segment. This not only ensures the operating efficiency of the high-speed transport of the mover, but also focuses on optimizing the positioning end accuracy, achieving a balance between efficiency and accuracy.

[0035] Meanwhile, this application conducts offline frequency analysis on historical steady-state motion data, pre-stores the estimated resonant frequency band for the mover end-effector positioning, and compares the frequency band deviation between the measured resonant frequency band and the estimated resonant frequency band in real time during the mid-term execution. Based on the deviation interval determination result, it adaptively selects to directly reuse the pre-stored suppression signal or correct the suppression signal in real time. This effectively solves the problem that a single pre-stored suppression signal cannot adapt to the slight deviation of the resonance characteristics under the working condition, and makes up for the limitations of the pure offline calibration compensation method. It ensures that the adaptive suppression signal always matches the current real-time load and resonance state of the mover, further improving the accuracy and adaptability of the single mover end-effector vibration suppression, ensuring the positioning stability of the precision stop section, and providing a stable single-unit operation foundation for multi-motor collaborative high-precision positioning.

[0036] In one specific implementation scheme, constructing and training the global coupling interference prediction model based on the adaptive suppression signal includes:

[0037] Based on the equipment structural parameters, establish a parametric dynamics model corresponding to the current stator track;

[0038] Using the adaptive suppression signal as a calibration excitation source, the parameter dynamics model is calibrated using excitation-response method to obtain the calibrated empirical parameters, thereby completing the training and obtaining the global coupling interference prediction model.

[0039] In one specific implementation scheme, generating the collaborative prediction compensation signal based on the global coupling interference prediction model includes:

[0040] The acceleration sequence of the target mover within the precision stopping segment is obtained according to the initial positioning command, and the acceleration sequence is input into the global coupling interference prediction model to obtain the predicted vibration displacement generated by the target mover at the positioning end of each other mover at each sampling time within the precision stopping segment.

[0041] For each sampling moment within the precision stopping segment, a system disturbance matrix for each sampling moment is constructed based on the predicted vibration displacements corresponding to all target movers;

[0042] Based on the system perturbation matrix, a cooperative prediction compensation amount is obtained for each target mover, and a cooperative prediction compensation signal is generated for each target mover based on the cooperative prediction compensation amount.

[0043] By adopting the above technical solution, this application builds a parametric dynamic model corresponding to the stator track based on the inherent structural parameters of the equipment, uses the adaptive suppression signal as the standard calibration excitation source to complete the model excitation-response calibration, and iteratively corrects the key empirical parameters to achieve accurate training of the full-domain coupling interference prediction model, so that the model can truly restore the mechanical vibration coupling characteristics transmitted between multiple movers through the guide rail and the base, and effectively ensure that the vibration disturbance prediction fits the actual working conditions.

[0044] Meanwhile, this application accurately extracts the acceleration sequence of the movers in this stage as the model input, predicts the transmitted vibration displacement between each mover at each sampling time, and accurately solves the coupled disturbance influence at the positioning end of each mover by constructing a system disturbance matrix. Finally, it generates a collaborative prediction and compensation signal that matches the real-time disturbance potential. This modeling and compensation generation method abandons the traditional empirical fixed compensation strategy and realizes quantitative prediction and accurate compensation based on the real structural vibration transmission mechanism. It specifically counteracts the cross-coupled vibration interference in the multi-motor collaborative motion process, effectively solves the end oscillation and positioning offset problems caused by adjacent axis vibration disturbance in high-speed precision positioning, and further improves the overall accuracy and operational stability of multi-axis mover collaborative positioning.

[0045] Secondly, this application provides a smart terminal, which adopts the following technical solution:

[0046] A smart terminal is characterized by comprising a memory and a processor, wherein the memory stores at least one instruction, at least one program, code set, or instruction set, and the at least one instruction, at least one program, code set, or instruction set is loaded and executed by the processor to implement a motion positioning control method based on a flexible magnetic drive conveying device as described in the first aspect.

[0047] Thirdly, this application provides a multi-axis mover positioning control system based on a flexible magnetic drive conveyor, which adopts the following technical solution:

[0048] A multi-axis mover positioning control system based on a flexible magnetic drive conveyor includes:

[0049] Stator track, a magnetic levitation track used to transport the target workpiece;

[0050] The mover assembly includes several movers slidably disposed on the stator track, and the movers are used to carry the target workpiece;

[0051] The position detection unit is used to collect the position detection signals of each mover in real time and obtain the coordinate information of each mover based on the position detection signals.

[0052] The intelligent terminal described in the second aspect is communicatively connected to the position detection unit and the moving part group, and is used to obtain positioning requirement information to generate an initial positioning instruction, and to obtain an adaptive shaping instruction based on the coordinate information to adaptively shape and adjust the initial positioning instruction to obtain a process positioning instruction.

[0053] The intelligent terminal is also used to obtain a collaborative prediction compensation signal based on the pre-stored equipment structure parameters, and to correct the process positioning command based on the collaborative prediction compensation signal to obtain a collaborative positioning command. Then, the collaborative positioning command is sent to the stator track to drive the corresponding mover to complete point-to-point positioning.

[0054] Fourthly, this application provides a computer-readable storage medium, which adopts the following technical solution:

[0055] A computer-readable storage medium storing at least one instruction, at least one program, code set, or instruction set, wherein the at least one instruction, at least one program, code set, or instruction set is loaded and executed by a processor to implement a motion positioning control method based on a flexible magnetic drive conveying device as described in the first aspect.

[0056] In summary, this application includes at least one of the following beneficial technical effects:

[0057] 1. This application adopts a dual-layer compensation mode that combines adaptive suppression and coupling prediction compensation. It eliminates the end resonance and positioning overshoot of a single mover by adaptively suppressing the signal, and compensates for the track coupling vibration interference between multiple movers by relying on the global coupling interference prediction model. It can simultaneously suppress the inherent oscillation of a single unit and the cross disturbance of multiple axes, and solves the problem that the traditional fixed parameter shaping method cannot adapt to the dynamic resonance changes in high-speed working conditions and has weak anti-disturbance capability. It greatly improves the accuracy and stability of mover positioning under high-speed and high-precision working conditions.

[0058] 2. This application adopts a segmented control strategy, superimposing dual compensation signals only in the precision stop segment, which effectively improves the overall accuracy and operational stability of high-speed collaborative positioning of multi-axis movers while ensuring the efficiency of high-speed conveying operation. Attached Figure Description

[0059] Figure 1 This is a schematic diagram of the structure of a multi-motor cooperative positioning control system based on a flexible magnetic drive conveying device according to an embodiment of this application.

[0060] Figure 2 This is a schematic diagram of a multi-motor cooperative positioning control system according to an embodiment of this application, illustrating the composition and interaction of the various parts.

[0061] Figure 3This is a flowchart illustrating a multi-motor cooperative positioning control method based on a flexible magnetic drive conveying device, according to another embodiment of this application.

[0062] Figure 4 This is a schematic diagram of another embodiment of the present application used to illustrate a historical resonant response waveform.

[0063] Figure 5 This is a flowchart illustrating the initial positioning command and adaptive suppression signal processing logic of a smart terminal in the early preparation stage, according to another embodiment of this application.

[0064] Figure 6 This is a flowchart illustrating the logic of constructing and training a global coupling interference prediction model for a smart terminal in the early preparation stage, according to another embodiment of this application. Detailed Implementation

[0065] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be further described in detail below with reference to the accompanying drawings.

[0066] The following description, in conjunction with the accompanying drawings, provides a more detailed account of an embodiment of a multi-axis mover positioning control system based on a flexible magnetic drive conveyor.

[0067] One embodiment of this application discloses a multi-axis mover positioning control system based on a flexible magnetic drive conveyor.

[0068] Reference Figure 1 A multi-axis mover positioning control system based on a flexible magnetic drive conveyor includes a stator track, a mover assembly, a position detection unit, and an intelligent terminal, wherein:

[0069] Reference Figure 1 and Figure 2 The stator track is composed of several stator modules spliced ​​together to form a magnetic levitation track for transporting the target workpiece. Electromagnetic coils are arranged inside the stator modules. Each stator module includes a stator drive unit corresponding to the electromagnetic coil in that stator module. The stator drive unit is specifically a drive circuit inside the stator module. The stator drive unit is used to precisely control the magnitude and direction of the current in the electromagnetic coil inside the stator module. In this way, the electromagnetic coil can generate the traveling wave magnetic field required to drive the mover located on the stator module according to the current change.

[0070] Reference Figure 2 The mover assembly includes several movers, each of which is slidably mounted on the stator track to support the workpiece. Each mover has a permanent magnet array installed at its bottom. The permanent magnet array is used to couple with the traveling wave magnetic field to generate a non-contact electromagnetic force between the stator track and the mover. The electromagnetic force is used to directly drive the mover to achieve a specific movement of the mover on the stator track.

[0071] Reference Figure 2 The position detection unit is arranged along the stator track to collect the absolute position and velocity information of each mover in real time, obtain the position detection signal, and obtain the coordinate information of each mover based on the position detection signal. The position detection unit in this embodiment includes, but is not limited to, magnetic grating ruler and read head assembly, optical grating ruler and read head assembly, Hall sensor array integrated in the stator module, etc. The position detection of the mover is the prior art. This embodiment takes the Hall sensor array integrated in the stator module as an example.

[0072] Reference Figure 2 The intelligent terminal is communicatively connected to the stator drive unit and position detection unit of each stator module. The intelligent terminal includes an instruction shaping submodule, a coupling prediction compensation submodule, and a cooperative instruction distribution submodule. The instruction shaping submodule is used to obtain the positioning requirement information input by the user at the input terminal to generate an initial positioning instruction. At the same time, it obtains the coordinate information of each mover in the position detection unit and obtains an adaptive shaping instruction based on all coordinate information to adaptively shape and adjust the initial positioning instruction to obtain a process positioning instruction. Then, the process positioning instruction is sent to the coupling prediction compensation submodule. The coupling prediction compensation submodule is used to obtain the cooperative prediction compensation signal based on the pre-stored equipment structure parameters, and to correct the process positioning command based on the cooperative prediction compensation signal to obtain the cooperative positioning command. Then, the cooperative positioning command is sent to the cooperative command distribution submodule. The cooperative command distribution submodule is used to parse the received cooperative positioning command to obtain the target positioning command corresponding to each mover, and to send all target positioning commands to the stator drive unit corresponding to the stator module where the corresponding mover is currently located. Each stator drive unit precisely controls the magnitude and direction of the current in its electromagnetic coil according to the received cooperative positioning command to drive the corresponding mover to complete point-to-point positioning.

[0073] It should be noted that the input terminal in this embodiment is communicatively connected to the instruction shaping submodule in the smart terminal. The input terminal is used to receive positioning requirement information set by the user based on the current workpiece and the conveying task. The positioning requirement information includes at least the target position, target speed, motion sequence, and maximum positioning allowable error parameter of each moving part carrying the current workpiece. The input terminal sends the positioning requirement information to the instruction shaping submodule so that the instruction shaping submodule can generate the corresponding initial positioning instruction. In this embodiment, the specific implementation form of the input terminal is not limited. The input terminal can be a human-machine interface integrated into the smart terminal, or it can be an external device such as a host computer, programmable logic controller, or manufacturing execution system, as long as it can provide formatted positioning requirement information to the smart terminal.

[0074] The smart terminal includes a processor and a memory, the memory being used to store at least one instruction, at least one program, code set, or instruction set; the processor performs the following steps S100-S600 when executing at least one instruction, at least one program, code set, or instruction set.

[0075] The implementation of the method will be explained in detail below with reference to the above system:

[0076] Reference Figure 3 Another embodiment of this application provides a motion positioning control method based on a flexible magnetic drive conveying device. The method mainly includes three stages: a pre-preparation stage before the mover performs the current positioning task, a mid-execution stage before the mover performs the current positioning task, and a post-optimization stage after the mover completes the current positioning task.

[0077] The preliminary preparation section includes the following steps S100-S300:

[0078] S100: Obtain the current positioning requirement information and parse the current positioning requirement information to obtain the initial positioning command;

[0079] Among them, the current positioning requirement information corresponds to the current positioning task, and the current positioning requirement information is set by the user through the input terminal. The current positioning requirement information includes at least the initial position of the target mover, the target position, the target velocity, and the maximum acceleration.

[0080] Specifically, the steps for parsing the current positioning requirement information to obtain the initial positioning command include the following:

[0081] S110, determine the total travel distance of the target mover for this positioning task based on the initial position and the target position, and determine the velocity curve type of the target mover for this positioning task based on the total travel distance, target velocity and maximum acceleration;

[0082] Wherein, the initial position refers to the absolute starting coordinates of the target mover in this positioning task, the target position refers to the absolute ending coordinates of the target mover in this positioning task, and the total travel refers to the absolute distance between the target position and the initial position; the velocity curve type includes, but is not limited to, trapezoidal velocity curve and triangular velocity curve; in this embodiment, the existing uniformly accelerated linear motion acceleration formula is used to determine the velocity curve type: calculate the minimum travel required for the target mover to accelerate from rest to the target velocity. If twice the minimum travel is greater than or equal to the total travel, it means that the target mover needs to start decelerating immediately after accelerating to the target velocity in the total travel, and the velocity curve type is triangular velocity curve. At this time, the target mover goes through the acceleration segment and then directly enters the deceleration segment; if twice the minimum travel is less than the total travel, it means that the total travel is sufficient, and the velocity curve type is trapezoidal velocity curve. At this time, the target mover goes through the acceleration segment, the uniform speed segment and the deceleration segment.

[0083] S120, Generate the initial positioning command for the target mover based on the velocity curve type;

[0084] The initial positioning command refers to the step position command sequence obtained without any adjustment or processing, based solely on the current positioning requirements. This command sequence specifies the position that the target mover should reach at each moment of the positioning task, and the final step in the sequence is the target position.

[0085] The following example of a linear conveyor belt provides a simple explanation of the parameters and calculations in S100:

[0086] Assuming the target mover is currently positioned at coordinate x=0mm on the stator track, the user sends a positioning task to the smart terminal via the input terminal, setting the current positioning requirements as follows: target position x=200mm, target velocity v1=1000mm / s, motion timing: no time limit, maximum allowable positioning error ±0.05mm, and maximum acceleration a=5000mm / s². 2 (This value is usually a system preset, but can be adjusted manually according to the actual situation.)

[0087] After receiving the aforementioned positioning request information, the intelligent terminal (specifically referring to the instruction shaping submodule within the terminal) calculates the total travel x of one of the target movers. max =200mm, minimum time t required to accelerate to the target speed a =0.2s, minimum required travel distance s during acceleration phase a =100mm, total stroke required for complete acceleration and deceleration x 总 =200mm=x max Since the uniform velocity time period is 0, the positioning task is determined to be a triangular velocity curve, and the corresponding initial positioning command, i.e., the step position command sequence, is shown in Table 1 below:

[0088] Table 1

[0089] 0 Start 0 1 acceleration phase 0.0025 ... ... ... 200 End of acceleration phase / Beginning of deceleration phase 100 ... ... ... 400 termination 200

[0090] S200: Acquire steady-state motion data of the target mover in at least one historical mission cycle, and determine the adaptive suppression signal of the target mover based on the steady-state motion data.

[0091] Among them, the historical task cycle refers to the complete point-to-point positioning task execution process that the target mover has completed and recorded and stored by the intelligent terminal before the current positioning task is executed. Each historical task cycle includes the entire process operation data of the corresponding task mover starting from the initial position, passing through the acceleration phase, the constant speed phase, and finally completing the positioning and stopping at the target position; the steady-state motion data includes the historical position error signal and historical acceleration response signal collected by the position detection unit at the end stage of the positioning of the target mover in the corresponding historical task cycle; the current positioning task also includes the maximum positioning allowable error parameter; the adaptive suppression signal is a compensation signal used to suppress the positioning oscillation and position overshoot of the single mover itself.

[0092] It should be noted that, in this embodiment, steady-state motion data is used to extract the historical resonance response characteristics exhibited by the target mover at the positioning end during the early preparation stage, and then generate an adaptive suppression signal that can be directly applied in the subsequent intermediate execution step S400 to suppress end overshoot and residual oscillation. The underlying logic is that, in the actual production of the flexible magnetic drive conveyor, the workpiece load carried by the target mover usually remains relatively stable over multiple consecutive task cycles, or is switched in an orderly manner according to the batch pattern of the production process. Therefore, the resonance response waveform exhibited by the target mover at the positioning end in the historical task cycle adjacent to or similar to the current positioning task can more accurately reflect the end vibration behavior characteristics under the current load condition.

[0093] By extracting the resonant response waveform from the data of historical mission cycles and inverting it, a set of adaptive suppression signals matching the current load conditions can be generated. By superimposing this signal onto the initial positioning command at a suitable positioning end, a reverse suppression force can be actively applied before vibration occurs, thereby offsetting end overshoot and residual oscillation to a certain extent from the source.

[0094] In addition, the steady-state motion data in this embodiment mainly selects the data collected at the end of the positioning process when the target mover is in a low-speed and medium-speed motion state. At this time, the target mover is in the transition range of deceleration and stopping, and its actual motion speed is in the low-to-medium speed range. The position error and acceleration response signal signal-to-noise ratio are higher at this stage, which can effectively avoid the interference of high-frequency noise of the sensor on the resonance characteristic analysis under high-speed conditions. At the same time, the micro-vibration at the end of the positioning process is the key target to be suppressed by the adaptive suppression signal in this embodiment. The resonance response characteristics extracted based on the data at this stage are completely consistent with the control target of the adaptive suppression signal in physical essence, thereby effectively ensuring that the adaptive suppression signal generated accordingly can accurately cancel the end vibration of the target mover under the current load.

[0095] In practice, the selection of historical task cycles follows the principle of prioritizing similarity of operating conditions. That is, steady-state motion data corresponding to historical task cycles that are more than similar to the current positioning task in key operating condition parameters such as load quality, target speed, and motion range are retrieved first. If there is no historical task cycle that meets the similarity threshold, the data of the most recently completed historical task cycle of the target mover is retrieved as the default sample, or the general calibration data of other movers of the same model under the same load conditions is called as a substitute.

[0096] Specifically, the steps in S200 for determining the adaptive suppression signal of the target mover based on steady-state motion data include the following:

[0097] S210, extract time-domain features from the historical position error signal in the steady-state motion data to obtain the historical resonance response waveform of the target mover at the end of the historical mission cycle positioning stage.

[0098] Among them, the historical resonance response waveform refers to the oscillation curve of the distance deviation between the actual position and the target position of the target mover at the end of the historical mission cycle, which changes over time. The waveform is extracted as follows:

[0099] Extract the position error signal segment from the steady-state motion data after the target mover enters the tolerance error band centered on the target position and with the maximum allowable positioning error as the radius. Record the distance deviation and time point corresponding to each segment as the historical resonance response waveform for this historical mission cycle. The specific historical resonance response waveform can be found in [reference needed]. Figure 4 An example is shown in the example (in the example, the horizontal axis represents time t, with the moment when the target mover enters the tolerance zone as the zero point, the vertical axis represents the position deviation e, positive values ​​represent the actual position of the target mover exceeding the target position, i.e., the overshoot direction, negative values ​​represent the actual position of the target mover not reaching the target position, and the curve is the fitting curve of the connecting line between corresponding points of multiple segments).

[0100] S220 inverts the historical resonant response waveform to obtain the initial suppression signal.

[0101] The inversion process involves multiplying the amplitude of the historical resonant response waveform at each moment by -1, so that the peak becomes the trough and the trough becomes the peak, thus obtaining a reverse suppression waveform that is opposite in phase to the original vibration. When the target mover is about to vibrate due to its own inertial motion, if the reverse waveform is applied to the target mover as a superimposed signal of the position command, the motion generated by the target mover will be equal in amplitude and opposite in direction to the original resonant vibration. The two will cancel each other out after being superimposed in the time domain.

[0102] S230: Obtain the maximum allowable positioning error parameter for the current positioning task of the target mover, and limit the amplitude of the initial suppression signal according to the maximum allowable positioning error parameter to obtain the adaptive suppression signal.

[0103] To ensure that the actual position deviation of the target mover remains within the required positioning accuracy range after the adaptive suppression signal is superimposed on the initial positioning command, the amplitude of the initial suppression signal is limited. The absolute value of the signal's peak value is restricted to within a preset proportion of the maximum allowable positioning error. In this embodiment, half of the maximum allowable positioning error is used as the upper limit of the amplitude limit. If the absolute value of the peak value of the initial suppression signal exceeds this upper limit, it is scaled proportionally to within the upper limit; otherwise, the original amplitude remains unchanged. Specifically, the formula for the adaptive suppression signal is as follows:

[0104]

[0105] ,

[0106] ,

[0107] in, This represents the time-amplitude curve corresponding to the adaptive suppression signal; Indicates the peak value of the initial suppression signal; This represents the time-amplitude curve corresponding to the initial suppression signal; This represents the maximum allowable positioning error for the current positioning task. Indicates the amplitude limiting threshold; This indicates the upper limit of the amplitude, in this embodiment. Pick .

[0108] It should be noted that this embodiment divides the motion process corresponding to the total travel of the target mover in a single positioning task into a rapid approach phase and a precise stopping phase. The rapid approach phase refers to the stage in which the target mover moves rapidly from the initial position to the target position with maximum acceleration or a relatively large speed. The end of this rapid approach phase corresponds to the moment when the target mover first enters the tolerance error zone centered on the target position and with the maximum allowable positioning error as the radius. The precise stopping phase refers to the stage from when the target mover enters the tolerance error zone until it finally stops stably at the target position. In this stage, the target mover is in a low-speed micro-motion state, which is the main interval where end overshoot and residual oscillation occur. It is also the target interval in this embodiment where the adaptive suppression signal is superimposed on the initial positioning command.

[0109] Based on the above division, in this embodiment, the adaptive suppression signal determined by S230 is entirely allocated to the precision stop segment. That is, the adaptive suppression signal is only superimposed on the initial positioning command in the initial stage of the precision stop segment, and no adaptive suppression signal is superimposed in the rapid approach segment. Therefore, the rapid approach segment can still minimize position deviation and pursue motion efficiency to the maximum extent without any interference from additional signals. In the precision stop segment, the adaptive suppression signal actively cancels the end-effector vibration of the target mover with a preset reverse waveform, thereby achieving feedforward suppression of overshoot and residual oscillation while ensuring end-effector positioning accuracy. Specifically, in this embodiment, the above operations are implemented using a switching function. To achieve this, in which This indicates that the adaptive suppression signal is in the precision stop segment; otherwise, it is 0. The subsequent process positioning signal superimposed by S400 onto the initial positioning command is... .

[0110] S240 performs offline frequency analysis on steady-state motion data to obtain the estimated resonant frequency band.

[0111] The estimated resonant frequency band refers to the dominant resonant frequency band at the end of the target mover's positioning during the historical mission cycle (which can be considered as the historical load conditions corresponding to the mission). Specifically, this embodiment can use existing offline frequency domain analysis methods such as Fast Fourier Transform or power spectral density estimation to extract frequency components with significant amplitudes that have a major impact on the stability of the end-positioning from the historical position error signal and historical acceleration response signal, and determine the lower limit of the resonant frequency distribution during the historical mission cycle. and frequency limit In this embodiment, the frequency range is taken as the estimated resonant frequency band.

[0112] Frequency analysis is a relatively mature existing technology, and only a brief calculation description is given here:

[0113] S241, Extract a discrete signal sequence from the steady-state motion data. Performing a Fast Fourier Transform on the discrete signal sequence yields the corresponding frequency domain representation. .

[0114] Among them, discrete signal sequence The length is , The sampling period is sampling frequency In this embodiment, the discrete signal sequence can represent the historical position error signal or historical acceleration response signal in the steady-state motion data; specifically, the fast Fourier transform formula involved in S241 is as follows:

[0115]

[0116] in, For discrete frequency indices in the frequency domain; Represents the first in a discrete signal sequence 1 sample point.

[0117] S242, Calculate the index for each discrete frequency. corresponding physical frequency .

[0118] Among them, physical frequency The calculation formula is as follows:

[0119]

[0120] S243, in frequency domain representation Corresponding amplitude spectrum Extract the lower frequency limit. and frequency limit .

[0121] First, determine the amplitude spectrum. Several peaks with the largest amplitude ;

[0122]

[0123] in, The preset scaling factor is 0.2 in this embodiment, which means that frequency components with amplitudes greater than 20% of the maximum amplitude are retained as valid data.

[0124] Then filter out all the peak values ​​that meet the criteria. Discrete frequency index And utilize the physical frequency in S242 Formula for calculating the selected discrete frequency index The corresponding physical frequency values ​​yield the candidate frequency set. Then, the minimum value in the candidate frequency set is used as the lower frequency limit. The maximum value in the candidate frequency set is used as the upper limit of frequency. .

[0125] S244, the interval It has been determined to be the predicted resonant frequency band.

[0126] In another embodiment, the amplitude spectrum in S243 Alternatively, a smoother power spectral density can be used. , .

[0127] S300 acquires pre-stored device structure parameters, constructs and trains a global coupling interference prediction model based on the adaptive suppression signal, and generates a collaborative prediction compensation signal based on the global coupling interference prediction model.

[0128] The global coupling interference prediction model refers to a mathematical model used to describe the mechanical vibration coupling relationship between a target mover and any other mover transmitted through the stator, track, and base. The global coupling interference prediction model takes the acceleration sequence of the target mover as input and the predicted vibration displacement generated by the target mover at the positioning end of other target movers on the track as output. In this embodiment, the input of the global coupling interference prediction model is specifically the acceleration value of a mover in the system at a certain moment; the output is the predicted vibration displacement caused at the positioning end of another mover after the reaction force generated by the acceleration of the mover at a certain moment is transmitted through the stator, track, and base; the cooperative prediction compensation signal is a reverse compensation signal used to cancel the mechanical coupling vibration interference transmitted along the track between several movers.

[0129] Equipment structural parameters refer to the inherent parameters that describe the physical structure and material properties of the current flexible magnetic drive conveyor. These mainly include geometric parameters such as the total length of the stator track and the length of each stator module; material parameters such as the elastic modulus and shear modulus between the stator track and the base; connection parameters such as the connection stiffness and connection damping between adjacent stator modules; and mass parameters such as the mass of a single mover and the equivalent mass of the stator module.

[0130] It should be noted that since the stator track is composed of several standardized stator modules spliced ​​together end to end, its mechanical structural parameters are determined when the equipment leaves the factory and remain unchanged during normal use. Therefore, the equipment structural parameters on which the global coupling interference prediction model depends have high stability and reusability. Once the model is built and calibrated, it can be repeatedly called throughout its entire life cycle without the need to frequently execute S300 for retraining.

[0131] Specifically, the steps for constructing and training a global coupled interference prediction model in S300 include the following:

[0132] S310, based on the equipment structural parameters, establish the parametric dynamic model corresponding to the current stator track.

[0133] The parametric dynamics model serves as the initial physical framework for the global coupling interference prediction model in S320. Specifically, it simplifies the stator track and base into a multi-degree-of-freedom system composed of several spring-damped-mass units. When each mover moves on the multi-degree-of-freedom system, its reaction force on the stator track acts as an excitation source on the corresponding node. The displacement response of each node represents the coupled vibration experienced by the mover at the corresponding position. In this embodiment, the parametric dynamics model takes a lumped parameter model, which is more suitable for online operation in industrial controllers, as an example. In other embodiments, the parametric dynamics model can also adopt a finite element model, but the finite element model has a larger computational load than the lumped parameter model.

[0134] Specifically, in this embodiment, the total number of movers is set to be The stator track is discretized into L equivalent nodes (each node can correspond to a stator module or a segment of the stator track), and each node has a lumped mass. When the mover p moves on the stator track, its reaction force It acts on the node where the mover is currently located.

[0135] The parametric dynamics model formula in this embodiment is in matrix form, as follows:

[0136] ,

[0137] ,

[0138] ,

[0139] in, This represents the displacement response vector at each node, i.e., the function of displacement changing with time; This represents the first derivative, i.e., velocity; This represents the second derivative, i.e., acceleration; The mass matrix is ​​a diagonal matrix used in this embodiment, with elements of: ; Multiplying the mass matrix by the acceleration vector corresponds to the calculation of inertial forces in the dynamic equations; The damping matrix is ​​given by the damping coefficients to be calibrated. composition; The stiffness matrix is ​​composed of the equipment structural parameters and the contact stiffness to be calibrated. composition; This is the position coupling matrix, used to distribute the reaction force generated by the mover to the corresponding node.

[0140] S320 uses the adaptive suppression signal as the calibration excitation source to perform excitation-response calibration on the parametric dynamics model, obtains the empirical parameters for calibration, completes the training, and obtains the global coupling interference prediction model.

[0141] Among them, empirical parameters refer to physical parameters that cannot be accurately calculated directly from the equipment structural parameters and must be determined through experimental calibration. In this embodiment, empirical parameters are the energy dissipation capacity of the mechanical structure of stator track and base during vibration, i.e., the damping coefficient, and the equivalent stiffness of the contact interface between adjacent stator module connection surfaces, between guide rail and mover, and between base and ground, i.e., contact stiffness.

[0142] Specifically, during system debugging or initial power-on initialization, the coupling prediction compensation submodule in the intelligent terminal sequentially applies the adaptive suppression signal from step S200 as an active excitation to each mover individually. Simultaneously, the position detection unit collects the acceleration response data of the remaining movers under the action of the adaptive suppression signal to obtain measured response data. Then, the adaptive suppression signal is used as input, and the measured response data is used as a calibration benchmark. This is compared with the model prediction response data of the parametric dynamics model in S310 under the same input conditions. The deviation between the two is used to construct a calibration error function, and the empirical parameters in the parametric dynamics model are iteratively corrected using the least squares method or gradient descent method until the value of the calibration error function converges to within a preset threshold. The damping coefficient and contact stiffness obtained at this time constitute the empirical parameters for completing the calibration. Finally, the empirical parameters for completing the calibration are substituted into the parametric dynamics model constructed in S310 to obtain the trained global coupling interference prediction model.

[0143] The steps for generating the cooperative prediction compensation signal based on the global coupled interference prediction model in S300 include the following:

[0144] S330: Based on the initial positioning command, obtain the acceleration sequence of the target mover in the precision stopping section, and input the acceleration sequence into the global coupling interference prediction model to obtain the predicted vibration displacement generated at the positioning end of each other mover at each sampling time in the precision stopping section.

[0145] The acceleration sequence is obtained by taking the position-time sequence contained in the process positioning command and performing two numerical differentiations to obtain the acceleration-time sequence, and extracting the part corresponding to the precision stop segment as the acceleration sequence. Then, the acceleration values ​​at each moment in the acceleration sequence are sequentially input into the trained global coupled interference prediction model. The output of the model is the predicted vibration displacement caused at the positioning end of each other mover after the acceleration excitation of the target mover is transmitted through the stator, track and base at the corresponding moment.

[0146] It should be noted that during the period when multiple movers complete the designated positioning task, each mover moves at a relatively high speed in the rapid approach segment. The position deviation is mainly dominated by the tracking error, and the relative influence of the coupled vibration is relatively small. Therefore, it is not included in the calculation scope of the cooperative prediction compensation. In this embodiment, only the cooperative vibration suppression in the precision stop segment (the main range in which each mover is in low-speed micro-motion, end overshoot and residual oscillation) needs to be considered.

[0147] S340 constructs the system disturbance matrix for each sampling moment within the precision stop segment based on the predicted vibration displacements corresponding to all target movers.

[0148] Specifically, let the total number of moving parts performing positioning tasks within the precision stopping segment in the system be... These movers constitute the target mover set. Assume there are a total of [number] precision stop sections. Each sampling time For each sampling time Perform the following sub-steps:

[0149] S341, Extract the sampling time from all predicted vibration displacements. The corresponding predicted vibration displacements of all moving parts , constitute the target dynamic subset .

[0150] in, Number the excitation source mover (i.e., the current target mover). The vibration receiving mover is numbered (i.e., one of the other movers besides the target mover). The target moving subset is the sampling time. All corresponding predicted vibration displacements A set of.

[0151] S342, Construct sampling time based on the target moving subset. Corresponding system disturbance matrix .

[0152] Among them, the system disturbance matrix Given an M*M nonsymmetric matrix, the first... Line number Column elements are defined as: when hour, This indicates that the mover itself does not generate base coupling self-disruption; when hour, , indicating the mover At sampling time Opposite to Predicted vibration displacement generated at the positioning end.

[0153] S350, based on system perturbation matrix The cooperative prediction compensation amount for each target mover is obtained, and a cooperative prediction compensation signal for each target mover is generated based on the cooperative prediction compensation amount.

[0154] Specifically, S350 includes the following steps:

[0155] S351, regarding the system disturbance matrix The Summing all elements in the list yields the mover. At sampling time Net positioning offset .

[0156] Among them, net positioning offset Represents the mover At sampling time The net disturbance caused by the coupled vibrations from all other movers in the precision stop segment.

[0157] Specifically, net positioning offset The calculation formula is as follows:

[0158] ,

[0159] S352, for the moving part At sampling time Net positioning offset Inverting the result yields the opposite motion. At sampling time Collaborative prediction compensation amount .

[0160] Among them, collaborative prediction compensation amount Used to counteract the movement At sampling time The net disturbance caused by the coupled vibration is a compensation displacement that needs to be superimposed on its position command; specifically, the collaboratively predicted compensation amount. The calculation formula is as follows:

[0161] ,

[0162] S353, traverse all At each sampling time, S341-S352 are executed repeatedly, and the same mover is... The co-predictive compensation values ​​at all sampling times are arranged in chronological order to obtain the mover. Collaborative prediction compensation signal .

[0163] Among them, collaborative prediction compensation signal It manifests as a sequence, specifically. .

[0164] It should also be noted that the adaptive suppression signal in this embodiment is a small-amplitude micro-motion signal superimposed on the initial positioning command. The additional acceleration caused by it is much smaller than the main motion acceleration of the initial positioning command itself. The amplitude of the coupled vibration excited by this small additional quantity is a second-order small quantity, and its impact on the accuracy of the coupled collaborative compensation is negligible in engineering. In this embodiment, the adaptive suppression signal mainly performs time-domain reverse cancellation of the resonant mode of a single mover, which is a single-axis feedforward; the collaborative prediction compensation signal cancels the base vibration transmission between multiple movers, which is a multi-axis collaborative feedforward, and the two action levels physically satisfy linear superposition.

[0165] Reference Figure 5 Before the control unit executes the current positioning task, the intelligent terminal completes the offline design and pre-storage of the adaptive suppression signal through S100-S200. This allows the intermediate execution stage to directly retrieve the adaptive suppression signal that matches the current working condition as a feedforward compensation quantity for superposition. Based on historical operating data, it achieves the pre-suppression of end residual oscillation and positioning overshoot.

[0166] Reference Figure 6 Through S300, the intelligent terminal builds a basic model of guide rail dynamics based on the inherent structural parameters of the device, and completes model excitation calibration and parameter iteration training by combining the adaptive suppression signal obtained offline. On the basis of adaptively suppressing the end resonance and residual oscillation of a single mover, a global disturbance prediction model that can characterize the coupling law of track-transmitted mechanical vibration between multiple movers is further established. This allows the system to quantitatively predict the coupled vibration disturbance between different movers in advance based on the real-time motion acceleration excitation of each mover in the subsequent mid-term execution phase, thereby performing reverse suppression.

[0167] The intermediate execution phase includes the following steps S400-S500:

[0168] S400: Obtain the initial positioning command and the adaptive suppression signal, and superimpose the adaptive suppression signal onto the initial positioning command to obtain the process positioning command.

[0169] The initial positioning command is generated by the pre-preparation section S120 and pre-stored in the command shaping submodule; the adaptive suppression signal is generated by the pre-preparation section S230 and pre-stored in the command shaping submodule.

[0170] It should be noted that, in order to ensure that the adaptive suppression signal maintains effective end-vibration suppression as much as possible under changing load conditions, and to effectively avoid compensation failure caused by mismatch between the pre-stored signal and the real-time operating conditions, S400 in this embodiment specifically includes the following before superimposing the adaptive suppression signal onto the initial positioning command:

[0171] S410: At the start of the precision stop segment, acquire the current measured resonant frequency band and the pre-stored estimated resonant frequency band, and compare the measured resonant frequency band with the estimated resonant frequency band to obtain the frequency band deviation.

[0172] The estimated resonant frequency band is calculated by the preparation section S240 and pre-stored in the instruction shaping submodule; the measured resonant frequency band is mainly based on the online spectrum analysis of the real-time position error signal of the end buffer of the rapid approach section of the mover to extract the measured resonant frequency of the target mover under the current load. In this embodiment, the online spectrum analysis can use existing mature methods such as recursive fast Fourier transform or sliding window FFT, which will not be elaborated here.

[0173] S420 determines whether the frequency band deviation is within the preset allowable deviation range.

[0174] If the judgment result is yes, it is determined that the frequency band deviation is within the allowable range, indicating that the current load condition is basically consistent with the historical condition, and the pre-stored adaptive suppression signal waveform can effectively match the current end vibration characteristics. At this time, S430 is executed directly. If the judgment result is no, it is determined that the frequency band deviation exceeds the allowable range, indicating that the current load condition is significantly different from the historical condition, and the pre-stored adaptive suppression signal may not be able to effectively suppress the current end vibration. At this time, S440 is executed.

[0175] Specifically, the instruction shaping submodule also divides the motion process corresponding to the total travel of this positioning task into a rapid approach segment and a precise stopping segment. In the rapid approach segment, the target mover is driven by the initial positioning command and rapidly approaches the target position with a preset maximum acceleration and maximum speed, without superimposing any adaptive suppression signal.

[0176] S430 transmits the pre-stored adaptive suppression signal through a switching function. Superimposed on the initial positioning command, the process positioning command is obtained. .

[0177] in, The time-position sequence corresponding to the initial positioning command; process positioning command. ; This is the time-amplitude sequence corresponding to the adaptive suppression signal generated in S230.

[0178] S440 corrects the adaptive suppression signal in real time based on the frequency band deviation, and then superimposes the corrected adaptive suppression signal onto the initial positioning command to obtain the process positioning command.

[0179] In the real-time position error signal acquired at the start of the precision stop segment, the response waveform of the current measured resonance is extracted. This waveform is then inverted according to step S220 to obtain the real-time initial suppression signal. Next, according to step S230, the amplitude of this real-time initial suppression signal is limited based on the maximum allowable positioning error parameter of the current positioning task to obtain the corrected adaptive suppression signal. At this time, the process positioning command .

[0180] It should be noted that both S420 and S430 are executed through the instruction shaping submodule after the target mover enters the tolerance error zone centered on the target position and with the maximum allowable positioning error as the radius, i.e., after entering the precision stop segment.

[0181] S500: Obtain the cooperative prediction compensation signal, superimpose the cooperative prediction compensation signal onto the process positioning command to obtain the cooperative positioning command, and send the cooperative positioning command to the corresponding stator drive unit.

[0182] The collaborative prediction compensation signal is generated by the pre-preparation segment S350 and pre-stored in the coupled prediction compensation submodule. Specifically, it is a sequence of collaborative prediction compensation signals for each mover at each sampling time within the precision stop segment. Specifically, the coupling prediction compensation submodule receives the process positioning instructions for each mover sent by the instruction shaping submodule, and determines the position based on the mover number. Retrieve the corresponding collaborative prediction compensation signal at each sampling time within the precision stop segment. The collaborative prediction compensation amount Superimposed on the mover From the process positioning instructions, the moving part is obtained. At sampling time Cooperative positioning commands , .

[0183] The post-optimization phase includes the following steps:

[0184] S600 acquires the end-point positioning deviation data for this positioning task, evaluates the compensation effect, and iteratively optimizes the adaptive suppression signal and the collaborative prediction compensation signal based on the evaluation results.

[0185] Specifically, the intelligent terminal retrieves the actual position deviation data of each mover within the precision stopping segment from the memory, and calculates positioning stability evaluation indicators such as the maximum end-effector overshoot amplitude and the root mean square value of residual oscillation. If all evaluation indicators meet the preset stability threshold, the operation data of this positioning task is directly stored in the database as valid historical data. If the evaluation indicator of any mover exceeds the threshold, the end-effector deviation data collected this time is used to regenerate the adaptive suppression signal according to the method of the previous preparation segment, and the cooperative prediction compensation signal is residually corrected. The updated adaptive suppression signal and cooperative prediction compensation signal will replace the original pre-stored values ​​in the previous preparation segment for use in the next positioning task. Finally, all the operation data of this task is stored in the memory as newly added historical task cycle data to continuously enrich the historical database.

[0186] The specific implementation methods of acquiring end-positioning deviation data, calculating positioning stability evaluation index, and iteratively updating compensation signal based on evaluation results involved in the above S600 have mature solutions in the prior art. For example, deviation data can be acquired in real time through the position detection unit, residual oscillation level can be evaluated by using the root mean square value calculation formula, and feedforward compensation amount can be iteratively corrected by using gradient descent or least squares method. This embodiment can directly adopt the above existing means, and will not be elaborated here.

[0187] Based on the same inventive concept described above, this application also discloses an intelligent terminal, which includes a memory and a processor. The memory stores at least one instruction, at least one program, code set, or instruction set. The at least one instruction, at least one program, code set, or instruction set is loaded and executed by the processor to implement a motion positioning control method based on a flexible magnetic drive conveying device as provided in the above method embodiments.

[0188] Based on the same inventive concept described above, this application also discloses a computer-readable storage medium storing at least one instruction, at least one program, code set, or instruction set. The at least one instruction, at least one program, code set, or instruction set can be loaded and executed by a processor to implement the motion positioning control method based on a flexible magnetic drive conveying device provided in the above method embodiments.

[0189] It should be understood that "multiple" as used in this article refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. The character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0190] Those skilled in the art will understand that all or part of the steps of the above embodiments can be implemented by hardware or by a program instructing related hardware. The program can be stored in a computer-readable storage medium, such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, and other media capable of storing program code.

[0191] The above description is merely an optional embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A motion positioning control method based on a flexible magnetic drive conveyor, comprising a multi-axis mover positioning control system, the system including a stator drive unit for driving the corresponding mover along a preset stator track to complete positioning according to instructions; characterized in that, The method includes a preliminary preparation phase and an intermediate execution phase, wherein the preliminary preparation phase includes: Obtain the current location requirement information and parse the current location requirement information to obtain the initial location command; Acquire steady-state motion data of the target mover during at least one historical mission cycle, the steady-state motion data including the historical position error signal collected during the final positioning stage of the target mover in the historical mission cycle; The historical position error signal is subjected to time-domain feature extraction to obtain the historical resonance response waveform of the target mover at the end of the historical mission cycle positioning stage. The historical resonance response waveform refers to the oscillation curve of the distance deviation between the actual position and the target position of the target mover as a function of time at the end of the historical mission cycle positioning stage. The historical resonant response waveform is inverted to obtain the initial suppression signal; Obtain the maximum allowable positioning error parameter for the current positioning task of the target mover, and limit the amplitude of the initial suppression signal according to the maximum allowable positioning error parameter to obtain an adaptive suppression signal; Obtain the pre-stored equipment structure parameters, and establish the parametric dynamics model corresponding to the current stator track based on the equipment structure parameters; The adaptive suppression signal is used as the calibration excitation source to perform excitation-response calibration on the parameter dynamics model, and the empirical parameters for calibration are obtained to complete the training and obtain the global coupling interference prediction model. In the total travel of a single positioning task, the motion process corresponding to the target mover includes a precise stopping segment, which refers to the stage from when the target mover enters a preset tolerance band until it stabilizes and stops at the target position. After obtaining the global coupling interference prediction model, the method further includes: The acceleration sequence of the target mover within the precision stopping segment is obtained according to the initial positioning command, and the acceleration sequence is input into the global coupling interference prediction model to obtain the predicted vibration displacement generated by the target mover at the positioning end of each other mover at each sampling time within the precision stopping segment. For each sampling moment within the precision stopping segment, a system disturbance matrix is ​​constructed based on the predicted vibration displacements corresponding to all target movers at each sampling moment; Based on the system perturbation matrix, a cooperative prediction compensation amount is obtained for each target mover, and a cooperative prediction compensation signal is generated for each target mover based on the cooperative prediction compensation amount. The method includes performing the following steps in the precision stop phase during the intermediate execution phase: Obtain the initial positioning command and the adaptive suppression signal, and superimpose the adaptive suppression signal onto the initial positioning command to obtain the process positioning command; The cooperative prediction compensation signal is acquired, and the cooperative prediction compensation signal is superimposed on the process positioning command to obtain the cooperative positioning command. The cooperative positioning command is then sent to the corresponding stator drive unit.

2. The motion positioning control method according to claim 1, characterized in that, The method further includes, after acquiring steady-state motion data of the target mover in at least one historical mission cycle during the preliminary preparation phase: Offline frequency analysis was performed on the steady-state motion data to obtain the estimated resonant frequency band; Before determining the adaptive suppression signal of the target mover based on the steady-state motion data in the intermediate execution phase, the following is also included: The measured resonant frequency band and the pre-stored estimated resonant frequency band are obtained, and the measured resonant frequency band and the estimated resonant frequency band are compared to obtain the frequency band deviation. Determine whether the frequency band deviation is within a preset allowable deviation range; If the judgment result is yes, then the pre-stored adaptive suppression signal is superimposed on the initial positioning command to obtain the process positioning command; If the judgment result is negative, the pre-stored adaptive suppression signal is corrected in real time according to the frequency band deviation, and the corrected adaptive suppression signal is superimposed on the initial positioning command to obtain the process positioning command.

3. A smart terminal, characterized in that, The device includes a memory and a processor, wherein the memory stores at least one instruction, at least one program, code set, or instruction set, and the at least one instruction, at least one program, code set, or instruction set is loaded and executed by the processor to implement the motion positioning control method based on a flexible magnetic drive conveying device as described in any one of claims 1 to 2.

4. A multi-axis mover positioning control system based on a flexible magnetic drive conveyor, characterized in that, include: Stator track, a magnetic levitation track used to transport the target workpiece; The mover assembly includes several movers slidably disposed on the stator track, and the movers are used to carry the target workpiece; The position detection unit is used to collect the position detection signals of each mover in real time and obtain the coordinate information of each mover based on the position detection signals. In the total travel of a single positioning task, the motion process corresponding to the target mover includes a precision stopping segment. The smart terminal as described in claim 3, communicatively connected to the position detection unit and the moving part group, is used for: Obtain the current location requirement information and parse the current location requirement information to obtain the initial location command; Acquire steady-state motion data of the target mover during at least one historical mission cycle, the steady-state motion data including the historical position error signal collected during the final positioning stage of the target mover in the historical mission cycle; The historical position error signal is subjected to time-domain feature extraction to obtain the historical resonance response waveform of the target mover at the end of the historical mission cycle positioning stage; The historical resonant response waveform is inverted to obtain the initial suppression signal; Obtain the maximum allowable positioning error parameter for the current positioning task of the target mover, and limit the amplitude of the initial suppression signal according to the maximum allowable positioning error parameter to obtain an adaptive suppression signal; Obtain the pre-stored equipment structure parameters, and establish the parametric dynamics model corresponding to the current stator track based on the equipment structure parameters; The adaptive suppression signal is used as the calibration excitation source to perform excitation-response calibration on the parameter dynamics model, and the empirical parameters for calibration are obtained to complete the training and obtain the global coupling interference prediction model. The acceleration sequence of the target mover within the precision stopping segment is obtained according to the initial positioning command, and the acceleration sequence is input into the global coupling interference prediction model to obtain the predicted vibration displacement generated by the target mover at the positioning end of each other mover at each sampling time within the precision stopping segment. For each sampling moment within the precision stopping segment, a system disturbance matrix is ​​constructed based on the predicted vibration displacements corresponding to all target movers at each sampling moment; Based on the system perturbation matrix, a cooperative prediction compensation amount is obtained for each target mover, and a cooperative prediction compensation signal is generated for each target mover based on the cooperative prediction compensation amount. In the intermediate execution phase, the smart terminal is also used to perform the following steps in the precision stop phase: Obtain the initial positioning command and the adaptive suppression signal, and superimpose the adaptive suppression signal onto the initial positioning command to obtain the process positioning command; The cooperative prediction compensation signal is acquired, and the cooperative prediction compensation signal is superimposed on the process positioning command to obtain the cooperative positioning command. The cooperative positioning command is then sent to the corresponding stator drive unit.

5. A computer-readable storage medium, characterized in that, The readable storage medium stores at least one instruction, at least one program, code set, or instruction set, wherein the at least one instruction, at least one program, code set, or instruction set is loaded and executed by a processor to implement the motion positioning control method based on a flexible magnetic drive conveying device as described in any one of claims 1 to 2.

Images