Parameter self-tuning method and device of magnetic suspension flexible conveying system and storage medium

Abstract

The application relates to a parameter self-tuning method and device of a magnetic suspension flexible conveying system and a storage medium. Based on a multi-source system triggering mechanism, a self-tuning starting signal is generated. Then, a mover of the magnetic suspension flexible conveying system is controlled to execute a preset target excitation motion track, and motion data of the mover are collected, and in combination with a translational linear motion model, load characteristics of the magnetic suspension flexible conveying system are estimated in real time. Next, target control parameters suitable for the current load are dynamically calculated through a preset tuning rule. Finally, the target control parameters calculated in real time are sent to a driver of the magnetic suspension flexible conveying system, and the driver is controlled to operate according to the target control parameters. In this way, an online and efficient self-tuning method of the magnetic suspension flexible conveying system can be realized, so that the stable operation of the magnetic suspension flexible conveying system is ensured while the accuracy is ensured.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of industrial automation control, in particular to a parameter self-tuning method and device of a magnetic suspension flexible conveying system and a storage medium. BACKGROUND

[0002] The magnetic suspension flexible conveying system is an automatic conveying device integrating magnetic suspension driving technology and flexible manufacturing concept, and the core is to realize the non-contact suspension and driving of the mover and the stator track through electromagnetic force. The magnetic suspension flexible conveying system needs to handle loads of different weights and shapes. In order to achieve optimal motion performance (such as the fastest speed, the smallest vibration, etc.), the servo control parameters of the mover need to be accurately tuned. At present, the industry generally adopts the "recipe" method, that is, the engineers pre-adjust a set of parameters for the known load, and the device calls when running. However, this method is time-consuming and laborious, and cannot cope with unknown loads or changes in load characteristics in the middle of the load (such as material consumption or material switching). If an offline calibration tool is used for self-tuning, it often needs to be stopped and the change of load characteristics caused by the change of load is not considered, which affects the industrial production efficiency and precision. SUMMARY

[0003] The purpose of the present application is to provide a parameter self-tuning method and device of a magnetic suspension flexible conveying system, and a storage medium, to solve the problem of low efficiency of the traditional tuning method of the magnetic suspension flexible conveying system.

[0004] In order to achieve the above-mentioned purpose, the first aspect of the present application provides a parameter self-tuning method of a magnetic suspension flexible conveying system, comprising:

[0005] Generating a self-tuning start signal based on a multi-source system trigger mechanism;

[0006] In response to the self-tuning start signal, controlling the mover of the magnetic suspension flexible conveying system to execute a preset target excitation motion trajectory, and collecting motion data of the mover;

[0007] Based on the motion data, determining a translational linear motion model of the magnetic suspension flexible conveying system, and based on the translational linear motion model, real-time estimating the load characteristics of the magnetic suspension flexible conveying system;

[0008] Based on the load characteristics and a preset tuning rule, calculating target control parameters adapted to the current load, and the tuning rule includes a mapping relationship between the load characteristics and the control parameters;

[0009] Downlinking the target control parameters to the driver of the magnetic suspension flexible conveying system, and controlling the driver to run according to the target control parameters.

[0010] The second aspect of the application provides a parameter self-tuning device of a magnetic levitation flexible conveying system, comprising:

[0011] A triggering module is configured to generate a self-tuning start signal based on a multi-source system triggering mechanism.

[0012] A collection module is configured to control a mover of the magnetic levitation flexible conveying system to perform a preset target excitation motion trajectory and collect motion data of the mover in response to the self-tuning start signal.

[0013] A first determination module is configured to determine a translational linear motion model of the magnetic levitation flexible conveying system based on the motion data and estimate a load characteristic of the magnetic levitation flexible conveying system in real time based on the translational linear motion model.

[0014] A second determination module is configured to calculate a target control parameter adapted to a current load based on the load characteristic and a preset tuning rule, wherein the tuning rule comprises a mapping relationship between the load characteristic and the control parameter.

[0015] A control module is configured to send the target control parameter to a driver of the magnetic levitation flexible conveying system and control the driver to operate according to the target control parameter.

[0016] The third aspect of the application provides a computer readable storage medium, wherein the computer readable storage medium stores a program, and the program can be loaded by a processor and execute the parameter self-tuning method of the magnetic levitation flexible conveying system.

[0017] The application has the following beneficial effects:

[0018] The application generates a self-tuning start signal based on a multi-source system triggering mechanism, without manual judgment or intervention, thereby solving the problem of manual triggering and limited scenarios in the traditional parameter tuning method. Then, the mover of the magnetic levitation flexible conveying system is controlled to perform a preset target excitation motion trajectory, and the motion data of the mover is collected, and the load characteristic of the magnetic levitation flexible conveying system is estimated in real time in combination with a translational linear motion model. Next, a target control parameter adapted to a current load is dynamically calculated through a preset tuning rule. Through dynamic adaptation to load changes, the problem that a traditional fixed parameter control is difficult to adapt to dynamic changes of the load is solved, the mover tracking error can be reduced, and the control precision of the parameter is improved. Finally, the target control parameter calculated in real time is sent to the driver of the magnetic levitation flexible conveying system, and the driver is controlled to operate according to the target control parameter, without the need for shutdown operation, and the online self-tuning process can be completed. In this way, an online and efficient self-tuning method of the magnetic levitation flexible conveying system can be realized, so as to ensure stable operation of the magnetic levitation flexible conveying system while ensuring precision.

[0019] Other features and advantages of the present application will be illustrated in the following detailed description. BRIEF DESCRIPTION OF DRAWINGS

[0020] Figure 1 A flowchart of a parameter self-tuning method of a magnetic levitation flexible conveying system provided in an embodiment of the present application is shown in FIG. 1.

[0021] Figure 2 A schematic diagram of an S-shaped curve complete acceleration and deceleration trajectory planning provided in an embodiment of the present application is shown in FIG. 2.

[0022] Figure 3 A structural schematic diagram of a parameter self-tuning device of a magnetic levitation flexible conveying system provided in an embodiment of the present application is shown in FIG. 3.

[0023] Figure 4 A structural schematic diagram of a controller provided in an embodiment of the present application is shown in FIG. 4. DETAILED DESCRIPTION

[0024] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0025] In the description of the present application, the terms "first", "second", etc. are used only for the purpose of description, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the technical features indicated. Therefore, the features defined with "first", "second" can explicitly or implicitly include one or more of the features. In the description of the present application, the meaning of "multiple" is two or more, unless otherwise specifically limited. In the present application, the word "exemplary" is used to mean "serving as an example, instance, or illustration". Any embodiment described as "exemplary" in the present application is not necessarily construed as preferred or advantageous over other embodiments. In order to enable any person skilled in the art to implement and use the present application, the following description is given. In the following description, details are listed for the purpose of explanation. It should be understood that those skilled in the art can realize the present application without using these specific details. In other examples, well-known structures and processes will not be described in detail to avoid unnecessary details making the description of the present application obscure. Therefore, the present application is not intended to be limited to the embodiments shown, but is consistent with the broadest scope consistent with the principles and features disclosed in the present application.

[0026] Figure 1A flowchart of a parameter self-tuning method of a magnetic levitation flexible conveying system provided in an embodiment of the present application is shown in FIG. 1. As shown in FIG. 1, the parameter self-tuning method can include steps 101-105, which will be described in detail below. Figure 1

[0027] Step 101, generating a self-tuning start signal based on a multi-source system triggering mechanism.

[0028] The multi-source system triggering mechanism refers to automatically determining whether the parameter self-tuning step needs to be started through multiple triggering conditions. As an example, the multi-source system triggering mechanism can integrate time dimension, load dimension, performance dimension, and manual intervention as triggering logic to ensure that the self-tuning step is automatically started when necessary, reducing the consumption of system resources caused by invalid self-tuning. The self-tuning start signal is an instruction signal triggering the subsequent self-tuning step of the magnetic levitation flexible conveying system, which can be transmitted to the controller through the system bus. Compared with the traditional single manual triggering tuning method, it can be automatically started on demand, reduce the dependence on manual intervention, and can adapt to different scenarios, reduce invalid tuning, and save computing power consumption.

[0029] Step 102, in response to the self-tuning start signal, controlling the mover of the magnetic levitation flexible conveying system to execute a preset target excitation motion trajectory and collecting motion data of the mover.

[0030] The target excitation motion trajectory is a preset motion trajectory for collecting real-time motion parameters of the current control magnetic levitation flexible conveying system. For the characteristics of the magnetic levitation flexible conveying system, such as the transportation of liquid medicine, in order to reduce suspension oscillation and ensure the stability of data collection, a trajectory with continuous and second-order derivable acceleration, such as an S-shaped trajectory, can be used as the target excitation motion trajectory to reduce rigid impact. The collection of motion data can be synchronized with the execution of the target excitation motion trajectory at a set frequency. As an example, the suspension height can be monitored by a magnetic levitation gap sensor, and the acceleration and / or maximum acceleration can be reduced if the suspension height is greater than the set safety height. After recalculating the target excitation motion trajectory, the motion of the mover is continued. The real-time correction mechanism of the gap fluctuation can reduce the suspension instability caused by too large trajectory parameters, and improve the safety of the system.

[0031] Step 103, determining a translational linear motion model of the magnetic levitation flexible conveying system based on the motion data, and real-time estimating the load characteristics of the magnetic levitation flexible conveying system based on the translational linear motion model.

[0032] ​The motion of the magnetic levitation conveying system is translational motion of the mover without rotation or swing. The core motion parameters are displacement, velocity, and acceleration, without considering rotational inertia, angular acceleration, and other rotational motion parameters. The core force of the magnetic levitation flexible conveying system can be simplified as motor reasoning against inertial force and friction force. In order to ensure the real-time performance of parameter self-tuning, the motion model of the magnetic levitation flexible conveying system can ignore the nonlinear motion (such as edge effect) and be simplified as a linear motion model. The translational linear motion model is a linear model of the relationship between motor reasoning, inertial force, and friction force based on the characteristics of translational linear motion. The final goal of parameter self-tuning is to calculate the control parameters that adapt to the load characteristics. The load characteristics refer to the characteristic parameters related to the load. According to Newton's second law and the friction law, the load characteristics can include total inertia and friction coefficient. The total inertia is the sum of the mover and the load mass, which is constant in a short time and linearly related to acceleration. The friction coefficient is mainly determined based on air resistance and electromagnetic damping, which is linearly related to velocity at medium speed. The total inertia directly reflects the resistance of the load to acceleration and determines the system response speed. The friction coefficient reflects the resistance of the load to velocity and determines the steady-state operation energy consumption. Therefore, as an example, the translational linear motion model can be a linear model with the total inertia and the friction coefficient as the parameters to be identified. Through the linear model, the characteristics of low friction and inertia dominance of the magnetic levitation flexible conveying system can be reflected, the problem of large calculation amount and slow efficiency caused by excessive complex model can be reduced, and the demand for real-time estimation of load characteristics can be met.

[0033] Step 104, based on the load characteristics and the preset tuning rule, calculating the target control parameters that adapt to the current load.

[0034] The target control parameters are parameters used for controlling the magnetic levitation flexible conveying system after self-tuning. The tuning rule is a quantitative calculation logic established by combining control theory and system characteristics. The tuning rule can include the mapping relationship between the load characteristics and the control parameters. Compared with the traditional experience parameter table, the adaptability is poor. When the load characteristics change, the control parameters can be accurately updated in real time.

[0035] Step 105, issuing the target control parameters to the driver of the magnetic levitation flexible conveying system, and controlling the driver to operate according to the target control parameters.

[0036] The driver of the magnetic levitation flexible conveying system is a power amplification device that receives control parameters and outputs motor current, and is a core component connecting the controller and the linear motor. The real-time target control parameters after self-tuning are packaged and sent to the driver of the linear motor through industrial Ethernet, etc. The driver receives the target control parameters and immediately updates the internal register values. Then, the driver drives the mover to move according to the new target excitation motion trajectory based on the updated target control parameters. In order to ensure the accuracy of the updated mover motion, in an example, the mover motion tracking error can be continuously monitored. If the motion tracking error is too large, the self-tuning step such as step 101 can be started, and the parameter correction of the self-tuning is continuously performed. In this way, the online real-time self-tuning step can be realized, and the working efficiency of the system can be improved.

[0037] The embodiment of the present application generates a self-tuning start signal based on a multi-source system triggering mechanism, without manual judgment or intervention, and solves the problem of manual triggering and scene limitation in traditional parameter tuning. By dynamically adapting to load changes, the problem that traditional fixed parameter control is difficult to adapt to dynamic changes in load is solved, the mover tracking error can be reduced, and the control accuracy of the parameters can be improved. At the same time, the target control parameters calculated in real time are sent to the driver of the magnetic levitation flexible conveying system, the driver is controlled to operate according to the target control parameters, without the need for shutdown operation, and the online self-tuning process can be completed. In this way, an online efficient self-tuning method of the magnetic levitation flexible conveying system can be realized, so as to ensure the stable operation of the magnetic levitation flexible conveying system while ensuring the accuracy.

[0038] In step 101, the multi-source system triggering mechanism can generate a self-tuning start signal based on at least one of the following: an external instruction for starting self-tuning, detecting that a load is placed on the mover, and detecting that the motion tracking error of the mover exceeds a set error.

[0039] The external instruction for starting self-tuning refers to an instruction signal that initiates the self-tuning step from the outside of the system. For example, it can include manual instructions of a human-computer interaction interface, integrated instructions of an upper system, or remote control instructions, etc. In an example, the external instruction can carry an instruction verification code. After the controller of the magnetic levitation flexible conveying system verifies that the verification code is passed, the self-tuning start is executed, reducing the interference of misoperation or illegal instructions, etc.

[0040] The embodiments of the present application can also monitor the load state of the mover in real time, and when it is determined that the load is stably placed on the mover, the self-tuning step can be automatically triggered. In one example, the load detection can be performed by a non-contact or contact sensor or a combination thereof. For example, the pressure of the mover can be detected based on a pressure sensor installed on the bearing surface of the mover, and when the pressure value of the mover is greater than a set pressure value, it is determined that the load is placed on the mover. For another example, the image of the bearing surface of the mover can be captured by a visual sensor, and compared with a reference image of the empty load, and when the profile of the workpiece is recognized and the matching degree with the target workpiece is greater than a set matching degree, it is determined that the load is placed on the mover. In order to reduce misjudgment, the judgment can also be based on the combined determination result of the pressure sensor and the visual sensor. When the self-tuning is triggered when the load is detected to be placed on the mover, the production change efficiency can be improved, and for small batch and multi-specification conveying requirements, such as different specifications of reagent bottles in the biomedicine field, the parameters of the current load can be adapted through self-tuning, and the decline in conveying accuracy caused by fixed parameters can be reduced.

[0041] In the operation process of the magnetic suspension flexible conveying system, the motion tracking error of the mover can be monitored in real time. The motion tracking error is the deviation value between the actual motion trajectory and the target motion trajectory, and is a core index for measuring the conveying accuracy. The error threshold is a preset upper limit of the error according to the process requirement, and is a reference for determining whether the parameters need to be re-tuned. In one example, the actual position of the mover can be collected in real time by a magnetic grid encoder or an optical grating ruler, and compared with the preset target position, and the difference is the motion tracking error. In another example, when the number of motion tracking errors exceeds a set number of times, such as 3 times in succession, a self-tuning start signal is generated again, and the false triggering caused by single transient deviation is reduced. By starting the self-tuning step to correct the parameter drift in real time, the accuracy decay problem caused by the parameter drift of the magnetic suspension flexible conveying system in long-term operation can be solved, and the operation accuracy of the system can be improved.

[0042] In step 102, in response to the self-tuning start signal, initial parameters of the first excitation motion trajectory can be obtained. The first excitation motion trajectory is an initial trajectory scheme retrieved by the system without verification. In order to protect the stability of the magnetic levitation flexible conveying system, the acceleration and deceleration of the first excitation motion trajectory can be set to be in a symmetrical relationship. That is, the acceleration change law, time length, parameter size, etc. of the acceleration process and deceleration process of the first excitation motion trajectory are in a mirror image relationship. The initial parameters can include motion displacement, target speed, maximum acceleration, jerk, initial speed and final speed. The motion displacement adapts to the effective stroke of the magnetic levitation flexible conveying system, and ensures that the trajectory completely covers the acceleration and deceleration and constant speed stage. The target speed takes into account the identification accuracy and tuning efficiency. If the speed is too low, it is easy to cause weak friction signal, and if the speed is too high, it is easy to introduce electromagnetic noise. The maximum acceleration is set based on the maximum output thrust of the driver of the magnetic levitation flexible conveying system, to reduce the problem of insufficient thrust caused by excessive acceleration.

[0043] In one example, the first excitation motion trajectory can be set as an S-type excitation motion trajectory. Figure 2 An S-type curve complete acceleration and deceleration trajectory planning schematic diagram provided in a specific embodiment of the present application is shown in FIG. 1. Figure 2 In FIG. 1, P is displacement, V is speed, a is acceleration, j is jerk, and t is time. Figure 2 The variation law of displacement P, speed V, acceleration a, and jerk J with time t of a seven-segment S-type motion trajectory is shown. The S-type curve trajectory planning has continuous acceleration and smooth speed transition, the system runs stably, and the flexible impact is small. The complete speed process can include seven stages of jerk acceleration, uniform acceleration, deceleration, uniform speed, acceleration and deceleration, uniform deceleration, and deceleration (see S1, S2, S3, S4, S5, S6, and S7). S1 to S7 are characteristic points on the displacement curve, corresponding to the displacement state of different motion stages. For the relationship diagram of speed V and time t, it can include initial speed , , , , , , , , , , , , , , , , , indicates a characteristic moment on the time axis, corresponding to the switching point of each motion stage. In the maximum acceleration-a versus time-t graph, the maximum acceleration , and the maximum deceleration . In the maximum jerk-j versus time-t graph, the maximum jerk , and the maximum deceleration .

[0044] In actual applications, the motion displacement S obtained according to the positions of the starting point and the ending point, the initial speed , the final speed , the target speed , the maximum acceleration , and the maximum jerk , etc. are used as constraint conditions to divide the acceleration-deceleration process into four cases: complete seven-segment speed, six-segment speed, five-segment speed, and four-segment speed. Each stage can correspond to a time length T (such as , , , , , and correspond to the time lengths of different stages, respectively), , , , , , , and are the starting or ending time points of different stages. Then, the first excitation motion trajectory generated based on the initial parameters is verified.

[0045] First, it is verified whether the first excitation motion trajectory contains a uniform acceleration segment. Since the acceleration in the uniform acceleration segment is constant, the inertial force dominates the thrust, and thus the uniform acceleration segment is the core stage of total inertia identification. Specifically, based on the target speed, the initial speed, the jerk, and the maximum acceleration, the first verification formula is used to verify whether the initial excitation motion trajectory contains a uniform acceleration segment. The first verification formula is a quantitative basis for judging whether the first excitation motion trajectory contains a constant acceleration segment, i.e., a uniform acceleration segment. The first verification formula verifies whether the system reaches the maximum acceleration during the acceleration process and forms a continuous uniform acceleration segment.

[0046] As an example, the first verification formula satisfies:

[0047] .

[0048] wherein is the target speed, is the initial speed, is the jerk, and is the maximum acceleration.

[0049] The first check formula is to determine whether the maximum acceleration change provided by the jerk during the process of increasing the speed from the initial speed to the target speed is greater than or equal to the minimum acceleration change required by the maximum acceleration. If the initial parameters satisfy the first check formula, it indicates that the jerk is large enough to maintain the maximum acceleration for a period of time before the speed reaches the target speed, and then the first excitation trajectory can be directly determined as the second excitation trajectory. Otherwise, if the initial parameters do not satisfy the first check formula, the jerk is adjusted upward or the maximum acceleration is adjusted downward until the adjusted initial parameters satisfy the first check formula, and the first excitation trajectory after adjusting the initial parameters is determined as the second excitation trajectory.

[0050] Based on the check result of the first check formula, the second excitation trajectory can be obtained, and the total acceleration time and the total deceleration time in the second excitation trajectory can be calculated. The second excitation trajectory is the excitation trajectory obtained after the uniform acceleration segment is checked, and the second excitation trajectory includes the uniform acceleration segment. The total acceleration time is the total duration of the acceleration process, and the total deceleration time is the same as the total acceleration time due to the symmetry of the trajectory.

[0051] After determining that the second excitation trajectory includes the uniform acceleration segment, it is then checked whether the second excitation trajectory includes a uniform speed segment. Since the acceleration of the uniform speed segment is 0, the thrust only overcomes the friction, so the uniform speed segment is the core stage of the friction coefficient identification. Specifically, the initial excitation trajectory can be checked whether it includes a uniform speed segment based on the motion displacement, target speed, total acceleration time, total deceleration time, initial speed, and final speed, combined with the second check formula. The second check formula is a quantitative basis for determining whether the second excitation trajectory includes a uniform speed segment, and the core is to ensure that the duration of the uniform speed segment is greater than a set time, such as 0.1s, through the relationship between the motion displacement, target speed, and total acceleration and deceleration time.

[0052] As an example, the second check formula satisfies:

[0053] .

[0054] wherein, is the motion displacement, is the target speed, is the total acceleration time, is the total deceleration time, is the initial speed, is the final speed.

[0055] The theoretical time required for the mover to move at the target speed throughout the entire process is compared with the total time of the acceleration segment and the deceleration segment. If the initial parameters satisfy the second check formula, the second excitation motion trajectory is determined as the third excitation motion trajectory, that is, the theoretical time is greater than or equal to the total time, which indicates that the motion displacement is long enough, and there is still time to run at the target speed except for the acceleration and deceleration segments, that is, a uniform speed segment is formed. On the contrary, if the initial parameters do not satisfy the second check formula, that is, the theoretical time is less than the total time, it indicates that the displacement is too short, and the acceleration needs to be immediately decelerated after reaching the target speed, without a uniform speed segment. At this time, the motion displacement needs to be increased or the target speed needs to be decreased until the second check formula is satisfied, and the second excitation motion trajectory after adjusting the initial parameters is determined as the third excitation motion trajectory.

[0056] Based on the checking result of the second check formula, the third excitation motion trajectory can be obtained. The third excitation motion trajectory is the trajectory scheme after the uniform speed segment check, which ensures to contain the uniform acceleration segment and the uniform speed segment, and is the basis of the final target excitation motion trajectory. The duration of the uniform speed segment directly determines the data amount of the friction coefficient identification, and is the key to the estimation accuracy of the friction coefficient. Therefore, the uniform speed segment time in the third excitation motion trajectory needs to be calculated. Finally, based on the third excitation motion trajectory and the uniform speed segment time, the target excitation motion trajectory is generated.

[0057] In the embodiments of the present application, the load characteristics can include the total inertia and the friction coefficient. Taking the target excitation motion trajectory as an S-type excitation motion trajectory as an example, the S-type excitation motion trajectory can include seven stages. Table 1 is a specific example of the seven stages of the S-type excitation motion trajectory.

[0058] Table 1

[0059]

[0060] In step 103, the motion data can be obtained at a set time interval first to obtain a plurality of groups of motion data, each group of motion data including motor thrust, acceleration and speed. The motion data is a set of physical quantities describing the motion state of the mover, which can include real-time thrust, real-time acceleration and real-time speed, etc., and is the original input of the load characteristic estimation. The set time interval is the time period for collecting motion data, which is set according to the demand to ensure that the dynamic changes of acceleration and speed are captured. The collected motion data can be subjected to operations such as moving average filtering and outlier rejection to improve the effectiveness of the data. The collection process continues until the end of the target excitation motion trajectory, for example, a plurality of groups of data can be collected, grouped and stored according to the time window of the uniform acceleration segment and the uniform speed segment, and finally the effective data of the uniform acceleration segment and the uniform speed segment are obtained.

[0061] Then, a translational linear motion model is established with total inertia and friction coefficient as the parameters to be identified. This model is based on the force analysis of the magnetic levitation flexible transport system, simplifying the complex dynamic characteristics into a second-order linear system. The parameters to be identified refer to the unknowns in the model that need to be estimated by the algorithm; the values ​​of total inertia and friction coefficient directly reflect the load characteristics. As an example, the translational linear motion model satisfies the following formula:

[0062] .

[0063] in, For the number of collections, For the first The motor thrust collected in this second sampling, where J is the total moment of inertia. For the first The acceleration collected in this instance, where B is the coefficient of friction. For the first The speed of each acquisition. For the first The smaller the systematic error in the initial data acquisition, the higher the model accuracy. Setting only two parameters to be identified reduces parameter coupling and iterative divergence, enabling the subsequent recursive least squares method to run in real time.

[0064] Next, based on multiple sets of motion data, a staged recursive least squares method is used to iteratively update the total inertia and friction coefficient of the translational linear motion model until the model converges. As shown in Table 1, since the uniform acceleration and uniform velocity segments contribute differently to parameter identification, the staged recursive least squares method, by dividing the model into uniform acceleration and uniform velocity segments, allows different iterations to focus on different parameters to be identified, thus reducing parameter coupling interference and improving the estimation accuracy of load characteristics. Once the translational linear motion model has iteratively converged, the iteratively updated total inertia and friction coefficient are output as the real-time estimated load characteristics of the magnetic levitation flexible transport system.

[0065] The following section elaborates on the iterative update process of the total inertia and friction coefficient of the translational linear motion model using the staged recursive least squares method.

[0066] First, considering the multi-segment structure of the target excitation trajectory, two types of key data can be collected based on the trajectory's motion characteristics, such as motion feature matching principles. These two types of key data can include collecting the first motion data from the uniform acceleration segment and the second motion data from the uniform velocity segment. The first motion data is the set of velocities from the uniform acceleration segment, characterized by constant and significant acceleration, used for identifying the dominant total inertia. The second motion data is the set of velocities from the uniform velocity segment, characterized by constant velocity and approximately zero acceleration, used for identifying the dominant friction coefficient.

[0067] Then, the uniform acceleration section is first iteratively updated by the least square method. Specifically, a first initial parameter vector of the uniform acceleration section is obtained, the first initial parameter vector being determined based on a preset initial total inertia and an initial friction coefficient. The first initial parameter vector is a starting parameter of the iteration of the uniform acceleration section, reflecting the initial cognition of the total inertia and the friction coefficient. The first initial parameter vector is then updated by the least square method until a relative variation rate of adjacent iterations of the first initial parameter vector is less than a first set variation rate, it is determined that the iteration of the uniform acceleration section converges, and the first initial parameter vector after the iteration converges is determined as a second initial parameter vector. The second initial parameter vector is a parameter vector after the convergence of the uniform acceleration section, serving as a starting parameter of the iteration of the uniform speed section, so that parameter inheritance is realized. The first set variation rate is a judgment threshold for the convergence of the uniform acceleration section, being set based on the requirement that the total inertia needs to be rapidly approximated to the true value.

[0068] Next, the uniform speed section is iteratively updated by the least square method to realize convergence. Specifically, the second initial parameter vector is updated by the least square method until a relative variation rate of adjacent iterations of the second initial parameter vector is less than a second set variation rate, it is determined that the iteration of the uniform speed section converges. The second set variation rate is a judgment threshold for the convergence of the uniform speed section, being set based on the requirement that the friction coefficient needs to be stably accurate, and being more stringent than the first set variation rate, the second set variation rate being less than the first set variation rate. Therefore, the uniform speed section directly uses the parameters of the uniform acceleration section, without re-initialization, and the convergence time is greatly reduced, which meets the requirement of online real-time estimation.

[0069] Finally, the total inertia and the friction coefficient after the iteration are extracted based on the second initial parameter vector after the iteration converges. The total inertia and the friction coefficient obtained after the stage-by-stage iteration meet the actual physical law, and are the optimal estimation of the true load characteristics, which can be directly used for the calculation of control parameters and for the analysis of load variation trends. Compared with the traditional full-stage iteration, the stage-by-stage recursive least square method improves the calculation accuracy of the load characteristics by decoupling the parameters, and shortens the iteration time by inheriting the parameters, which meets the requirement of online real-time scene and provides a reliable load characteristic input for the parameter self-tuning of the magnetic suspension flexible conveying system.

[0070] In the embodiments of the present application, the target control parameters can include a target feedback gain parameter and a target feedforward gain parameter. In the magnetic suspension flexible conveying system, the target feedback gain parameter and the target feedforward gain parameter are complementary control parameters, respectively assuming the functions of error correction and anti-interference and advance compensation. In the scene of dynamic load variation, the stability, anti-interference ability and high-precision tracking performance of the system can be simultaneously realized.

[0071] The target feedback gain parameter is a gain parameter for suppressing disturbances, ensuring system stability and tracking accuracy, is the core parameter of proportional-integral-derivative (PID) control, and can include proportional gain, integral gain and derivative gain. There are various unknown disturbances in the running of the magnetic levitation flexible conveying system, such as track unevenness, electromagnetic noise and air turbulence, etc., so that the actual motion deviates from the target excitation motion trajectory. The three can realize dynamic correction of the magnetic levitation flexible conveying system through a closed-loop feedback mechanism.

[0072] The target feedforward gain parameter is used to compensate for known load resistance in advance and can include acceleration feedforward parameters and speed feedforward parameters. The two can actively offset known load resistance before error occurs through an open-loop feedforward mechanism, reduce error accumulation, and improve dynamic tracking accuracy.

[0073] Through the closed-loop error correction of the target feedback gain parameter, the disturbance rejection and stability of the magnetic levitation flexible conveying system are ensured, and through the open-loop compensation of the target feedforward gain parameter, the tracking accuracy and response speed are improved, which can perfectly adapt to the demand for high precision and fast response of the magnetic levitation flexible conveying system under the dynamic change of load.

[0074] Specifically, in step 104, the closed-loop characteristic equation of the magnetic levitation flexible conveying system can be established based on the translational linear motion model. The closed-loop characteristic equation describes an algebraic equation of the dynamic characteristics of the system. The closed-loop characteristic equation can include closed-loop poles, which are solutions of the closed-loop characteristic equation and reflect the modal of the free motion of the system. From the translational linear motion model to the closed-loop characteristic equation, the feedback gain calculation is bound to the dynamic characteristics of the system, solving the blindness of the traditional experience trial and error.

[0075] Then, according to the current estimated total inertia and friction coefficient, the expected poles of the expected characteristic equation are set. The expected poles are the closed-loop poles that meet the performance index and need to be matched with the total inertia and friction coefficient. By adapting the load characteristics through the expected poles, the problem of overshoot or slow response caused by fixed poles when the load changes is solved.

[0076] Based on the pole placement algorithm, the target feedback gain parameter is solved, which makes the closed-loop poles equal to the expected poles. The pole placement algorithm is a control design method that makes the closed-loop poles equal to the expected poles by designing the feedback gain, so as to directly configure the system dynamic characteristics according to the performance index.

[0077] Specifically, based on Newton's second law, the Laplace transformation is performed on the motion equation of the magnetic levitation flexible conveying system, and the open-loop transfer function, i.e. the linear motor system model of the magnetic levitation flexible conveying system is obtained as follows:

[0078] .

[0079] where J is the total inertia calculated, B is the friction coefficient calculated, and s is the Laplace transform.

[0080] Meanwhile, feedback control can be introduced, for example, to calculate the force with error. The most commonly used can be PID control, and the PID controller transfer function is satisfies:

[0081] .

[0082] where, is the proportional gain, is the integral gain, is the derivative gain.

[0083] The closed-loop characteristic equation can satisfy:

[0084] .

[0085] The roots of the closed-loop characteristic equation, i.e. the closed-loop poles, directly determine the system stability and response speed. For example, the real part of the root is negative, which indicates that the system is stable, and the farther the root is from the imaginary axis, the faster the response speed.

[0086] In order to make the system meet the requirements of fast response and no overshoot, the expected poles can be set, and the node configuration equation of the expected second-order system is configured, and the poles of the expected second-order system satisfy the characteristic equation:

[0087] .

[0088] where, is the expected damping ratio (usually 0.7-1.0), is the expected natural frequency, which is related to the system bandwidth. Specifically, .

[0089] By comparing the coefficients of the closed-loop characteristic equation and the expected characteristic equation, the target feedback gain parameters can be solved.

[0090] The empirical formula of the proportional gain satisfies: , which matches the position response strength.

[0091] The empirical formula of the derivative gain satisfies: , which matches the damping and suppresses the overshoot.

[0092] The empirical formula of the integral gain satisfies: , which eliminates the steady-state error of the system.

[0093] The system PID parameter values are obtained by the J, B and the system expected damping ratio and the known system bandwidth. The embodiments of the application can solve the problems of insufficient gain or overshoot of the traditional fixed PID control when the load changes by precisely matching the target feedback gain parameter with the load characteristics. By setting the expected pole, the system response time can be reduced and the system response speed can be improved.

[0094] Then, the acceleration feedforward parameter is calculated based on the total inertia and the inertia force law, and the velocity feedforward parameter is calculated based on the friction coefficient and the friction force law. The acceleration feedforward parameter and the velocity feedforward parameter are determined as the target feedforward gain parameter. By compensating the inertia force and the friction force in advance, the tracking error of the mover can be reduced, especially for the acceleration section dominated by the inertia force and the uniform speed section dominated by the friction force. The feedforward bears the main load resistance compensation, and the feedback only needs to handle small disturbances, reducing the shock caused by excessive feedback gain. Based on the physical formula calculation, the matching degree of the feedforward gain parameter and the load characteristics can be enhanced, and the dynamic changes of the load can be adapted.

[0095] As an example, the feedforward gain optimization based on the total inertia and the friction coefficient can include the following steps.

[0096] In uniform motion, the motor thrust mainly overcomes the friction force and the external load, and the role of the velocity feedforward parameter is to compensate the speed-related resistance (mainly the friction force). In actual application, the value of the velocity feedforward parameter is close to 1 (for example, it can be 0.95-0.98). The role of the acceleration feedforward parameter is to accurately compensate the inertia force, which is the key to improving performance.

[0097] According to Newton's second law: The calculation formula of the acceleration feedforward parameter satisfies: In order to perfectly compensate the inertia force: Therefore, the final acceleration feedforward is equal to the system inertia force, that is, .

[0098] Through the collaborative optimization of the feedback gain parameter and the feedforward gain parameter, the tracking accuracy and stability of the magnetic suspension flexible conveying system can be considered, the precise matching of the control parameter and the load characteristics is realized, and the problem of poor load adaptability caused by the traditional fixed parameter control is solved.

[0099] In step 105, the target control parameter can be written into the parameter configuration storage area of the driver through the communication interface of the driver of the magnetic levitation flexible conveying system first. The communication interface is the interface connecting the controller and the driver, which needs to meet the real-time and reliability, for example, the industrial real-time Ethernet interface can be used. The parameter configuration storage is a special storage area in the driver for storing control parameters, which can include a real-time effective running memory and a long-term saved configuration storage area. Through real-time transmission, the quick effect of the parameter can be guaranteed. And even if the system is powered off, the parameter can still be reused, reducing the initialization time.

[0100] Then, a hot update verification operation is performed to perform range verification and logic verification on the target control parameter to verify the validity of the target control parameter. The hot update verification refers to verifying the validity of the written target control parameter without interrupting the running of the driver, reducing the situation that illegal parameters cause system failure. The range verification is to verify whether the target control parameter is within the physical range allowed by the hardware capability of the driver, reducing the hardware damage caused by parameter over-limit. The logic verification is to verify whether the relevance of the target control parameter meets the control algorithm logic, reducing the poor system performance caused by improper parameter matching.

[0101] If the hot update verification passes, the control parameter of the driver is updated to the target control parameter for running, and a parameter effective signal is fed back to the magnetic levitation flexible conveying system. The driver synchronizes the target control parameter of the parameter configuration storage area to the running memory, and the update process time is short, which can ensure seamless switching of the parameter. If the hot update verification does not pass, the control parameter of the driver is not updated, and the system continues to run according to the current control parameter, reducing the system out of control caused by invalid new control parameter, and feeding back a hot update abnormal signal to the magnetic levitation flexible conveying system. The abnormal signal can include alarm information and abnormal reason, facilitating the rapid positioning of faults and reasons by the operation and maintenance personnel.

[0102] The embodiments of the present application improve the safety and update efficiency of the target control parameter through real-time writing and hot update verification of the target control parameter. Through the feedback of the abnormal signal, the rapid positioning of the fault is realized. In this way, the stable and high-precision running of the magnetic levitation flexible conveying system can be realized without shutdown.

[0103] Figure 3 A structure diagram of a parameter self-tuning device of a magnetic levitation flexible conveying system provided in the embodiments of the present application is shown in FIG. 3. Figure 3 As shown in FIG. 3, the parameter self-tuning device 300 of the magnetic levitation flexible conveying system can include a triggering module 301, a collecting module 302, a first determining module 303, a second determining module 304, and a control module 305.

[0104] The triggering module 301 is configured to generate a self-tuning start signal based on a multi-source system triggering mechanism.

[0105] The acquisition module 302 is configured to control the mover of the magnetic suspension flexible conveying system to perform a preset target excitation motion trajectory and acquire motion data of the mover in response to the self-tuning start signal.

[0106] The first determination module 303 is configured to determine a translational linear motion model of the magnetic suspension flexible conveying system based on the motion data, and estimate the load characteristics of the magnetic suspension flexible conveying system in real time based on the translational linear motion model.

[0107] The second determination module 304 is configured to calculate a target control parameter adapted to the current load based on the load characteristics and a preset tuning rule, the tuning rule including a mapping relationship between the load characteristics and the control parameter.

[0108] The control module 305 is configured to issue the target control parameter to a driver of the magnetic suspension flexible conveying system, and control the driver to operate according to the target control parameter.

[0109] In the embodiments of the present application, the triggering module 301 can include at least one of a first start unit, a second start unit and a third start unit.

[0110] The first start unit is configured to generate the self-tuning start signal in response to an external instruction for self-tuning start.

[0111] The second start unit is configured to generate the self-tuning start signal in response to detecting that a load is placed on the mover.

[0112] The third start unit is configured to generate the self-tuning start signal in response to detecting that a motion tracking error of the mover exceeds a set error.

[0113] In the embodiments of the present application, the acquisition module 302 can include a first acquisition unit, a first verification unit, a first calculation unit, a second verification unit, a second calculation unit and a generation unit.

[0114] The first acquisition unit is configured to acquire initial parameters of the first excitation motion trajectory in response to the self-tuning start signal, the acceleration and the deceleration of the first excitation motion trajectory being in a symmetrical relationship, and the initial parameters including a motion displacement, a target speed, a maximum acceleration, a jerk, an initial speed and a final speed.

[0115] The first verification unit is configured to verify whether the initial excitation motion trajectory has a uniform acceleration segment based on the target speed, the initial speed, the jerk and the maximum acceleration, and in combination with a first verification formula.

[0116] The first calculation unit is configured to obtain a second excitation motion track based on a check result of the first check formula, and calculate total acceleration time and total deceleration time in the second excitation motion track.

[0117] The second check unit is configured to check, based on the motion displacement, the target speed, the total acceleration time, the total deceleration time, the initial speed and the final speed, whether the initial excitation motion track has a uniform speed segment in combination with a second check formula.

[0118] The second calculation unit is configured to obtain a third excitation motion track based on a check result of the second check formula, and calculate a uniform speed segment time in the third excitation motion track.

[0119] The generation unit is configured to generate a target excitation motion track based on the third excitation motion track and the uniform speed segment time.

[0120] In the embodiments of the present application, the first check formula satisfies:

[0121] ;

[0122] wherein, the target speed is V, the initial speed is V0, the jerk is j, and the maximum acceleration is a max.

[0123] The first calculation unit is further configured to, when the initial parameters satisfy the first check formula, determine the first excitation motion track as the second excitation motion track, and when the initial parameters do not satisfy the first check formula, increase the jerk or decrease the maximum acceleration until the adjusted initial parameters satisfy the first check formula, and determine the first excitation motion track after the initial parameters are adjusted as the second excitation motion track.

[0124] In the embodiments of the present application, the second check formula satisfies:

[0125] ;

[0126] wherein, the motion displacement is S, the target speed is V, the total acceleration time is T1, the total deceleration time is T2, the initial speed is V0, and the final speed is Vf.

[0127] The second computing unit is further configured to determine the second excitation motion track as a third excitation motion track when the initial parameters satisfy the second check formula, and to increase the motion displacement or decrease the target speed until the second check formula is satisfied, and then determine the second excitation motion track after the initial parameters are adjusted as the third excitation motion track when the initial parameters do not satisfy the second check formula.

[0128] In the embodiment of the present application, the load characteristics include total inertia and friction coefficient, and the first determining module 303 can include a second obtaining unit, a first establishing unit, an iteration unit and an output unit.

[0129] The second obtaining unit is configured to obtain the motion data at a set time interval to obtain a plurality of groups of motion data, each group of motion data including motor thrust, acceleration and speed.

[0130] The first establishing unit is configured to establish a translational linear motion model with the total inertia and the friction coefficient as to-be-identified parameters.

[0131] The iteration unit is configured to update the total inertia and the friction coefficient of the translational linear motion model based on the plurality of groups of motion data by using a stage-by-stage recursive least square method until the translational linear motion model converges.

[0132] The output unit is configured to output the total inertia and the friction coefficient after the iteration update as real-time estimated load characteristics of the magnetic levitation flexible conveying system.

[0133] The translational linear motion model satisfies the following formula:

[0134] ;

[0135] wherein, is the number of collection times, is the motor thrust collected for the i-th time, J is the total inertia, is the acceleration collected for the i-th time, B is the friction coefficient, is the speed collected for the i-th time, is the system error collected for the i-th time.

[0136] ​​​​The iterative unit is specifically used to collect first motion data of the uniform acceleration segment and second motion data of the uniform velocity segment based on the motion characteristics of the target excitation trajectory; obtain the first initial parameter vector of the uniform acceleration segment, which is determined based on a preset initial total inertia and initial friction coefficient; update the first initial parameter vector using the least squares method until the relative change rate of the first initial parameter vector in adjacent iterations is less than a first set change rate, determine that the uniform acceleration segment iteration has converged, and determine the first initial parameter vector after iteration convergence as the second initial parameter vector; update the second initial parameter vector using the least squares method until the relative change rate of the second initial parameter vector in adjacent iterations is less than a second set change rate, determine that the uniform velocity segment iteration has converged, and the second set change rate is less than the first set change rate; determine the updated total inertia and friction coefficient based on the second initial parameter vector after iteration convergence.

[0137] In this embodiment, the target control parameters include target feedback gain parameters and target feedforward gain parameters. The second determination module 304 may include a second establishment unit, a setting unit, a first solution unit, a third calculation unit, and a second solution unit.

[0138] The second unit is used to establish the closed-loop characteristic equation of the magnetic levitation flexible transport system based on the translational linear motion model. The closed-loop characteristic equation includes the closed-loop poles.

[0139] The setting unit is used to set the desired pole of the desired characteristic equation based on the currently estimated total inertia and the friction coefficient.

[0140] The first solving unit is used to solve for the target feedback gain parameter, which is equal to the desired pole, based on the pole placement algorithm.

[0141] The third calculation unit is used to calculate acceleration feedforward parameters based on total inertia and the law of inertial force, and to calculate velocity feedforward parameters based on friction coefficient and the law of frictional force.

[0142] The second solver unit is used to determine the acceleration feedforward parameters and velocity feedforward parameters as the target feedforward gain parameters.

[0143] Figure 4 This is a schematic diagram of the structure of a controller provided in an embodiment of this application. Figure 4 As shown, the controller 400 may include a memory 401 and a processor 402. The memory 401 is configured to store instructions. The processor 402 is configured to retrieve instructions from the memory 401 and, when executing instructions, to implement the parameter self-tuning method of the magnetic levitation flexible transport system described above.

[0144] The embodiment of the present application further provides a computer readable storage medium, which stores a program capable of being loaded by a processor and executing the parameter self-tuning method of any one of the magnetic suspension flexible conveying systems.

[0145] The parameter self-tuning device, the controller and the computer readable storage medium store instructions, so that the steps in the parameter self-tuning method of any one of the magnetic suspension flexible conveying systems provided by the embodiment of the present application can be executed, and thus the beneficial effects of the parameter self-tuning method of any one of the magnetic suspension flexible conveying systems provided by the embodiment of the present application can be achieved. Details are described above, and thus will not be described here.

[0146] Those skilled in the art can understand that all or part of the functions of the various methods in the above embodiments can be realized by hardware or by a computer program. When all or part of the functions in the above embodiments are realized by a computer program, the program can be stored in a computer readable storage medium, which can include a read-only memory, a random access memory, a magnetic disk, an optical disk, a hard disk, etc. The above functions are realized by executing the program by a computer. For example, the program is stored in a memory of a device, and when the program in the memory is executed by a processor, the above functions are realized. In addition, when all or part of the functions in the above embodiments are realized by a computer program, the program can also be stored in a server, another computer, a disk, an optical disk, a flash disk or a mobile hard disk, etc. The program is downloaded or copied into a memory of a local device, or the system of the local device is updated, and when the program in the memory is executed by a processor, the above functions are realized.

[0147] The above application is described by using specific examples, which is only used to help understand the present application and does not limit the present application. According to the idea of the present application, those skilled in the art can make several simple deductions, modifications or replacements.

Claims

1. A method for parameter self-tuning of a magnetic levitation flexible transport system, characterized in that, include: A self-tuning start signal is generated based on a multi-source system triggering mechanism. In response to the self-tuning start signal, the mover of the magnetic levitation flexible conveyor system is controlled to execute a preset target excitation motion trajectory, and the motion data of the mover is collected; Based on the motion data, a translational linear motion model of the magnetic levitation flexible transport system is determined, and based on the translational linear motion model, the load characteristics of the magnetic levitation flexible transport system are estimated in real time, including the total inertia and the coefficient of friction. Based on the load characteristics and preset tuning rules, target control parameters adapted to the current load are calculated, and the tuning rules include the mapping relationship between load characteristics and control parameters. The target control parameters are sent to the driver of the magnetic levitation flexible conveyor system, and the driver is controlled to operate according to the target control parameters. The step of determining the translational linear motion model of the magnetic levitation flexible transport system based on the motion data, and estimating the load characteristics of the magnetic levitation flexible transport system in real time based on the translational linear motion model, includes: The motion data is acquired at set time intervals to obtain multiple sets of motion data, each set of motion data including motor thrust, acceleration and velocity; A translational linear motion model is established using the total inertia and the friction coefficient as parameters to be identified; Based on multiple sets of motion data, the total inertia and friction coefficient of the translational linear motion model are iteratively updated using a staged recursive least squares method until the translational linear motion model converges. The total inertia and friction coefficient, updated iteratively, are output as the load characteristics of the magnetic levitation flexible transport system obtained in real time. The translational linear motion model satisfies the following formula: ； in, For the number of collections, For the first The motor thrust collected in this instance, J represents the total inertia. For the first The acceleration collected in this instance, where B is the coefficient of friction. For the first The speed of the next acquisition. For the first System error during the first acquisition.

2. The parameter self-tuning method according to claim 1, characterized in that, The generation of a self-tuning start signal based on the multi-source system triggering mechanism includes: In response to an external command for self-tuning startup, the self-tuning startup signal is generated; and / or In response to detecting that a load is placed on the mover, the self-tuning start signal is generated; and / or In response to the detection that the motion tracking error of the mover exceeds a set error, the self-tuning start signal is generated.

3. The parameter self-tuning method according to claim 1, characterized in that, The step of controlling the mover of the magnetic levitation flexible transport system to execute a preset target excitation motion trajectory in response to the self-tuning start signal includes: In response to the self-tuning start signal, the initial parameters of the first excitation motion trajectory are obtained. The acceleration and deceleration of the first excitation motion trajectory are symmetrical. The initial parameters include motion displacement, target velocity, maximum acceleration, jerk, initial velocity, and final velocity. Based on the target velocity, the initial velocity, the jerk, and the maximum acceleration, the initial excitation trajectory is checked for a uniform acceleration segment using the first verification formula. Based on the verification result of the first verification formula, the second excitation motion trajectory is obtained, and the total acceleration time and total deceleration time in the second excitation motion trajectory are calculated. Based on the motion displacement, the target velocity, the total acceleration time, the total deceleration time, the initial velocity, and the final velocity, the second verification formula is used to verify whether the initial excitation motion trajectory has a uniform velocity segment. Based on the verification result of the second verification formula, the third excitation motion trajectory is obtained, and the time of the uniform velocity segment in the third excitation motion trajectory is calculated. The target excitation trajectory is generated based on the third excitation trajectory and the uniform motion segment time.

4. The parameter self-tuning method according to claim 3, characterized in that, The first verification formula satisfies: ； in, The target speed, Let the initial velocity be... For the jerk, The maximum acceleration is mentioned above; The second excitation motion trajectory is obtained based on the verification result of the first verification formula, including: If the initial parameters satisfy the first verification formula, then the first excitation motion trajectory is determined as the second excitation motion trajectory; If the initial parameters do not satisfy the first verification formula, then the jerk is increased or the maximum acceleration is decreased until the adjusted initial parameters satisfy the first verification formula, and the first excitation motion trajectory after adjusting the initial parameters is determined as the second excitation motion trajectory.

5. The parameter self-tuning method according to claim 3, characterized in that, The second verification formula satisfies: ； in, The displacement is the motion displacement. The target speed, For the total acceleration time, The total deceleration time is... The initial velocity, The final velocity; The third excitation motion trajectory is obtained based on the verification result of the second verification formula, including: If the initial parameters satisfy the second verification formula, then the second excitation motion trajectory is determined as the third excitation motion trajectory; If the initial parameters do not satisfy the second verification formula, the motion displacement is increased or the target velocity is decreased until the second verification formula is satisfied, and the second excitation motion trajectory after adjusting the initial parameters is determined as the third excitation motion trajectory.

6. The parameter self-tuning method according to claim 5, characterized in that, Based on multiple sets of motion data, the total inertia and friction coefficient of the translational linear motion model are iteratively updated using a staged recursive least squares method until the translational linear motion model converges, including: Based on the motion characteristics of the target excitation trajectory, the first motion data of the uniform acceleration segment and the second motion data of the uniform velocity segment are collected respectively. Obtain the first initial parameter vector of the uniform acceleration segment, which is determined based on a preset initial total inertia and initial friction coefficient; The first initial parameter vector is updated by the least squares method until the relative rate of change of the first initial parameter vector in adjacent iterations is less than the first set rate of change. The uniform acceleration segment is then determined to have converged, and the first initial parameter vector after iterative convergence is determined as the second initial parameter vector. The second initial parameter vector is updated by the least squares method until the relative rate of change of the second initial parameter vector in adjacent iterations is less than the second set rate of change. The uniform speed segment iteration is then determined to be converged, and the second set rate of change is less than the first set rate of change. The total inertia and the friction coefficient are determined based on the second initial parameter vector after iterative convergence.

7. The parameter self-tuning method according to claim 1, characterized in that, The target control parameters include target feedback gain parameters and target feedforward gain parameters. The calculation of target control parameters adapted to the current load based on the load characteristics and preset tuning rules includes: Based on the translational linear motion model, a closed-loop characteristic equation for the magnetic levitation flexible transport system is established, and the closed-loop characteristic equation includes closed-loop poles. Based on the currently estimated total inertia and friction coefficient, the desired poles of the desired characteristic equation are set; Based on the pole placement algorithm, the target feedback gain parameter, in which the closed-loop pole is equal to the desired pole, is solved; Acceleration feedforward parameters are calculated based on the total inertia and the law of inertial force, and velocity feedforward parameters are calculated based on the friction coefficient and the law of frictional force. The acceleration feedforward parameter and the velocity feedforward parameter are determined as the target feedforward gain parameter.

8. A parameter self-tuning device for a magnetic levitation flexible conveyor system, characterized in that, include: The trigger module is used to generate a self-tuning start signal based on the multi-source system triggering mechanism; The acquisition module is used to respond to the self-tuning start signal, control the mover of the magnetic levitation flexible conveyor system to execute a preset target excitation motion trajectory, and acquire the motion data of the mover; The first determining module is used to determine the translational linear motion model of the magnetic levitation flexible transport system based on the motion data, and to estimate the load characteristics of the magnetic levitation flexible transport system in real time based on the translational linear motion model, wherein the load characteristics include total inertia and friction coefficient. The second determining module is used to calculate the target control parameters adapted to the current load based on the load characteristics and the preset tuning rules, wherein the tuning rules include the mapping relationship between the load characteristics and the control parameters. The control module is used to send the target control parameters to the driver of the magnetic levitation flexible conveyor system and control the driver to operate according to the target control parameters; The first determining module 303 may include: The second acquisition unit is used to acquire the motion data at a set time interval to obtain multiple sets of motion data, each set of motion data including motor thrust, acceleration and velocity; The first establishment unit is used to establish a translational linear motion model with the total inertia and the friction coefficient as the parameters to be identified. An iterative unit is used to iteratively update the total inertia and friction coefficient of the translational linear motion model based on multiple sets of motion data using a staged recursive least squares method until the translational linear motion model converges. The output unit is used to output the iteratively updated total inertia and friction coefficient as the load characteristics of the magnetic levitation flexible transport system obtained in real time. The translational linear motion model satisfies the following formula: ； in, For the number of collections, For the first The motor thrust collected in this instance, J represents the total inertia. For the first The acceleration collected in this instance, where B is the coefficient of friction. For the first The speed of the next acquisition. For the first System error during the first acquisition.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a program that can be loaded by a processor and executed as the parameter self-tuning method for the magnetic levitation flexible transport system as described in any one of claims 1 to 7.

Images