An online identification method, device, electronic device and storage medium for the moment of inertia of a rotary shaft servo system

Abstract

This application belongs to the field of mechatronics technology. Specifically, it relates to a method, device, electronic device, and storage medium for online identification of the moment of inertia of a rotating shaft servo system. This method first establishes a calculation model for the driving torque and frictional torque of the rotating shaft servo system; inputs a preset motor speed command with periodic speed regulation characteristics to the servo system; sequentially calculates the observation error of the motor speed, the recursive compensation term of the motor speed, and the robust activation coefficient within each sampling period; and finally updates the parameters to be identified in the next sampling period, repeating the sampling period until the preset motor speed command ends. Using this method, it is possible to perform efficient, high-precision, and high-reliability online identification of the moment of inertia of the rotating shaft servo system, perform real-time observation of the load torque, and the identification results can provide important references for on-site engineering technicians when analyzing the dynamic performance of the system, designing high-quality controllers, and tuning the parameters of the motor drive.

Description

Technical Field

[0001] This application belongs to the technical field of mechatronics. Specifically, it relates to a method, device, electronic device and storage medium for on-line identification of the moment of inertia of a rotating shaft servo system. Background Technique

[0002] The rotating shaft servo system, including the tool spindle, the motion joint shaft, the positioning turntable shaft, etc., is one of the basic functional components of precision manufacturing equipment represented by numerically controlled machine tools. Usually, a servo motor is used as the power source, and a set of transmission parts are connected between the servo motor and the working load to achieve high-precision and high-efficiency rotary motion, and it is widely used in the industrial field. The total moment of inertia of the rotating shaft servo system is a necessary parameter for analyzing the dynamic response of the electromechanical system, designing a high-performance controller, and tuning the parameters of the motor driver. Therefore, it is of great research significance and practical value to obtain the total moment of inertia of the rotating shaft electromechanical system efficiently and accurately.

[0003] In the design and application process of the rotating shaft servo system, engineers usually can only obtain the moment of inertia of the motor shaft itself from the motor manufacturer. For the moment of inertia of various components such as various connectors, transmission mechanisms, and load weights installed on the motor shaft, only approximate estimates can be made after a large number of simplifications, and the drive parameters of the motor are adjusted accordingly by trial and error. This approximate estimation method will lead to different degrees of mismatch of drive parameters in the rotating shaft servo system, affecting the running stability of the rotating shaft. For the direct drive rotating shaft system, since there is no transmission mechanism, the moment of inertia at the motor load end will be much larger than that of the motor shaft itself. At this time, the estimation error of the total moment of inertia of the system will further aggravate the degree of mismatch of drive parameters. In addition to the approximate estimation method, for complex mechatronic systems such as rotating shafts, traditional methods often require building a detailed three-dimensional model and using industrial software to calculate the total moment of inertia of the system. However, the assembly relationship between the components in the rotating shaft system is complex, making the software operation process very cumbersome and inefficient. At the same time, there are manufacturing and assembly errors in the actual system, and as the running time increases, wear occurs in each component, and the assembly prestress degrades, all of which will cause the total moment of inertia of the actual system to differ greatly from the value read by the software, and it is not suitable for the controller design of precision electromechanical equipment. Therefore, in order to accurately obtain the current total moment of inertia of the rotating shaft system, it is necessary to identify the moment of inertia according to the actual dynamic response of the system.

[0004] At present, the identification methods of moment of inertia can be divided into two categories: offline and online. The offline method can only be applied in the commissioning stage of the rotating shaft system, which requires storing a large amount of data, cannot effectively track the change of the total moment of inertia of the system, has a high test cost and is not conducive to engineering implementation. The online method usually takes the moment of inertia of the motor shaft of a small servo motor as the identification object. Under the condition of ensuring high modeling accuracy, by reading the speed and current of the servo motor in real time, the moment of inertia of the motor shaft is recursively calculated online. However, the present invention is directed to a rotating shaft servo system. The total moment of inertia of this electromechanical system not only includes the moment of inertia of the motor shaft, but also includes various connecting parts, transmission mechanisms and working loads loaded on the motor shaft, etc. At this time, the physical model for the online identification algorithm of the moment of inertia must be extended from the original servo motor model along the transmission chain to the entire rotating body. With the expansion of the modeling object, the system model will introduce more significant internal modeling errors and external environmental disturbances. The model accuracy will depend to a large extent on the dynamic characteristics of each component in the rotating shaft electromechanical system, rather than the motor components that are easy to describe. For example, the difference in the frictional torque characteristics introduced by bearing components. Small servo motors usually use rolling bearing sets with good lubrication conditions and stable working environments. Therefore, their frictional torque has an obvious linear function relationship with the motor shaft speed, including a constant Coulomb frictional torque and a viscous frictional torque proportional to the speed, and the friction coefficient basically does not change with time. However, in the rotating shaft servo system, the main transmission shaft connected to the motor shaft usually uses a hydrostatic or hydrodynamic bearing set or a rolling bearing set with general lubrication conditions, and its viscous friction coefficient will change continuously with the changes of the motor speed, working temperature and lubricating oil properties, making the frictional torque have an obvious non-linear function relationship with the motor speed, and each coefficient has complex time-varying characteristics, which are difficult to accurately describe and have large modeling errors. At this time, for application objects such as rotating shaft servo systems with complex structures, time-varying characteristics and difficult to accurately model, the existing online identification methods are easily affected by uncertain factors inside and outside the system such as modeling errors, parameter perturbations, unmodeled terms, unknown disturbances, non-linear terms and sampling noises, have poor robustness, are difficult to ensure the convergence of the identification results, and have poor reliability. For example, the literature (Lian C, Xiao F, Gao S, et al. Load Torque and Moment of Inertia Identification for Permanent Magnet Synchronous Motor Drives Based on Sliding Mode Observer[J]. IEEE TRANSACTIONS ON POWER ELECTRONICS, 2019, 34(6): 5675-5683.). Summary of the Invention

[0005] This application aims to partially solve the problems existing in the prior art. Based on the inventor's understanding and recognition of the following facts and problems, although the existing online identification algorithms for the moment of inertia of permanent magnet synchronous motors propose an accurate model of the motor and correct the identified value of the moment of inertia of the motor shaft through the real-time observation error of the load torque on the premise of knowing the load torque, when this method is applied to a rotary shaft servo system, due to the large modeling error of the physical model used in the algorithm, it is difficult for the algorithm to converge stably to a definite value, obvious numerical oscillations occur in the iterative process, the identification accuracy is seriously affected, and the identification result cannot be used. At the same time, the iterative process of some online algorithms is too complex, which is not conducive to on-site debugging and difficult to ensure calculation real-time and engineering feasibility.

[0006] In view of this, the present disclosure proposes an online identification method, device, electronic device and storage medium for the moment of inertia of a rotary shaft servo system, realizing high-efficiency, high-precision and high-reliability online identification of the moment of inertia of a rotary shaft servo system in actual engineering applications.

[0007] According to the first aspect of the present disclosure, an online identification method for the moment of inertia of a rotary shaft servo system is proposed, including:

[0008] Establish a calculation model for the driving torque and frictional torque of the rotary shaft servo system;

[0009] Using the calculation model, online identify the moment of inertia of the rotary shaft servo system according to the motor speed command of the rotary shaft servo system.

[0010] Optionally, the calculation model of the driving torque and frictional torque is as follows:

[0011]

[0012] In the formula, is the calculated value of the driving torque of the rotary shaft servo system, is the calculated value of the frictional torque of the rotary shaft servo system, K t is the torque coefficient of the motor, i is the sampled value of the motor current, ω is the sampled value of the motor speed, and A, B and C are constants.

[0013] Optionally, using the calculation model, online identify the moment of inertia of the rotary shaft servo system according to the motor speed command with stage speed regulation characteristics of the rotary shaft servo system, including:

[0014] (1) Set the initial sampling period serial number k = 1, and initialize the observed value of the motor speed to 0, the observed value of the load torque to 0, and the identified value of the moment of inertia to J init, the reciprocal identification value of the moment of inertia is

[0015] k is the sampling period serial number in the on-line identification process of the moment of inertia, and J init is the iterative initial value of the moment of inertia of the rotary shaft servo system;

[0016] (2) Input a preset motor speed command ω with stage speed regulation characteristics to the rotary shaft servo system d ;

[0017] (3) During the operation of the rotary shaft servo system, perform periodic sampling at a fixed time interval T s In the k-th sampling period, obtain the sampling value ω(k) of the motor speed, the sampling value i(k) of the motor current, and the sampling value ω d (k) of the motor speed command of the current rotary shaft servo system. Perform low-pass filtering on the sampling value ω(k) of the motor speed and overwrite the original value; use the calculation model of the driving torque and the friction torque to calculate the calculated value of the driving torque of the rotary shaft servo system and the calculated value of the friction torque

[0018] (4) Use the following formula to calculate the observation error e(k) of the motor speed and the recursive compensation term u(k) in the k-th sampling period:

[0019]

[0020] In the formula, is the observed value of the motor speed in the k-th sampling period, and k e is the compensation gain of the observation error of the motor speed, and the value of k e is non-negative; ε is the tolerance threshold of the observed value of the motor speed to error interference, and the value is positive. sat(e) is a saturation function, as shown in the following formula:

[0021]

[0022] In the formula, Δ e is the saturation threshold of the observation error of the motor speed, and the value is positive. sgn(e) is a sign function;

[0023] (5) Use the following formula to calculate the robust activation coefficient l(k) in the k-th sampling period:

[0024]

[0025] In the formula, is the observed value of the load torque in the k-th sampling period, and are the calculated values of the driving torque and the friction torque for the k-th sampling period respectively; g is the convergence speed gain, which takes a negative value, and Δ ω is the activation threshold of the motor speed fluctuation, which takes a positive value; max{x1, x2} is the maximum value function, as shown in the following formula:

[0026]

[0027] (6) Use the following formula to calculate the derivative of the observed value of the motor speed for the k-th sampling period respectively the derivative of the observed value of the load torque and the derivative of the identified value of the reciprocal of the moment of inertia as follows:

[0028]

[0029] (7) Use the following formula to recursively update the observed value of the motor speed the observed value of the load torque the identified value of the reciprocal of the moment of inertia for the (k + 1)-th sampling period as follows:

[0030]

[0031] In the formula, T s is the time interval between two adjacent sampling periods;

[0032] (8) Judge the preset motor speed command ω d with stage speed regulation characteristics. If the preset motor speed command ω d ends, then use the observed value of the motor speed in step (7) the observed value of the load torque the identified value of the reciprocal of the moment of inertia at each sampling period as the ordinate and the online identification time as the abscissa to plot the observed curve of the motor speed, the observed curve of the motor load torque and the convergence curve of the identified value of the reciprocal of the system moment of inertia, and obtain the identification result of the system moment of inertia from the convergence curve of the identified value of the reciprocal of the system moment of inertia If the preset motor speed command ω d has not ended, then return to step (2).

[0033] According to the second aspect of the present disclosure, an on-line identification device for the moment of inertia of a rotating shaft servo system is proposed, including:

[0034] A model establishment module for establishing a calculation model of the driving torque and the friction torque of the rotating shaft servo system;

[0035] A calculation module for online identifying the moment of inertia of a rotary shaft servo system according to a motor speed command of the rotary shaft servo system by using the calculation model.

[0036] According to a third aspect of the present disclosure, there is provided an electronic device, including:

[0037] A memory for storing computer-executable instructions;

[0038] A processor configured to execute:

[0039] Establish a calculation model for the driving torque and the friction torque of the rotary shaft servo system;

[0040] Online identify the moment of inertia of the rotary shaft servo system according to a motor speed command of the rotary shaft servo system by using the calculation model.

[0041] According to a fourth aspect of the present disclosure, there is provided a computer-readable storage medium having a computer program stored thereon, the computer program being configured to cause the computer to execute:

[0042] Establish a calculation model for the driving torque and the friction torque of the rotary shaft servo system;

[0043] Online identify the moment of inertia of the rotary shaft servo system according to a motor speed command of the rotary shaft servo system by using the calculation model.

[0044] According to an embodiment of the present disclosure, by using the method for online identifying the moment of inertia of the rotary shaft servo system of the present disclosure, it is possible to perform efficient, high-precision, and high-reliability online identification of the moment of inertia of the rotary shaft servo system, and to perform real-time observation of the load torque, overcoming the problem of insufficient robustness of the existing method under the large modeling error caused by a complex system. The identification result can provide an important reference for on-site engineering technicians when analyzing the dynamic performance of the system, designing a high-quality controller, and tuning the parameters of the motor driver.

[0045] Additional aspects and advantages of the present disclosure will be given in part in the following description, become apparent in part from the following description, or be learned through the practice of the present invention. Description of the Drawings

[0046] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can also obtain other drawings based on these drawings without creative efforts.

[0047] Figure 1It is a flowchart for online identification of the moment of inertia of a rotary shaft servo system shown according to an embodiment of the present disclosure.

[0048] Figure 2 It is a schematic cross-sectional view of a grinding wheel motorized spindle servo system of a precision roll grinder shown according to an embodiment of the present disclosure.

[0049] Figure 2 Among them, 1 is the grinding wheel, 2 is the grinding wheel connection assembly, 3 is the main shaft body, 4 is the rotor permanent magnet, 5 is the stator winding, 6 is the angle encoder, 7 is the hydrostatic and hydrodynamic bearing set, and 8 is the motorized spindle housing.

[0050] Figure 3 It is an observed curve of the motor speed shown according to an embodiment of the present disclosure.

[0051] Figure 4 It is an observed curve of the motor load torque shown according to an embodiment of the present disclosure.

[0052] Figure 5 It is a convergence curve of the reciprocal identification value of the moment of inertia of a rotary shaft servo system shown according to an embodiment of the present disclosure.

[0053] Figure 6 It is a schematic block diagram of an online identification device for the moment of inertia of a rotary shaft servo system shown according to an embodiment of the present disclosure. Detailed implementation manners

[0054] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts shall fall within the protection scope of the present application.

[0055] An online identification method for the moment of inertia of a rotary shaft servo system shown according to an embodiment of the present disclosure may include the following steps:

[0056] In step 1, a calculation model of the driving torque and the frictional torque of the rotary shaft servo system is established.

[0057] In one embodiment, the calculation model of the driving torque and the frictional torque is as follows:

[0058]

[0059] In the formula, is the calculated value of the driving torque of the rotary shaft servo system, is the calculated value of the frictional torque of the rotary shaft servo system, K t$K_t$ is the torque coefficient of the motor, $i$ is the sampled value of the motor current, $\omega$ is the sampled value of the motor speed, and $A$, $B$, and $C$ are constants. By conducting no-load tests on the rotary shaft servo system under different motor speed commands, plotting the scatter diagram of the calculated value of the steady-state driving torque and the sampled value of the steady-state motor speed, and further performing curve fitting on the functional relationship between the two, the values of $A$, $B$, and $C$ can be obtained.

[0060] In step 2, using the calculation model, according to the motor speed command with a phased speed regulation characteristic of the rotary shaft servo system, the moment of inertia of the rotary shaft servo system is identified online.

[0061] In one embodiment, using the calculation model, according to the motor speed command with a phased speed regulation characteristic of the rotary shaft servo system, the moment of inertia of the rotary shaft servo system is identified online. The online identification process is as Figure 1 shown and may include the following steps:

[0062] (1) Set the initial sampling period number $k = 1$, and initialize the observed value of the motor speed to 0, the observed value of the load torque to 0, the identified value of the moment of inertia as $J$ init , the identified value of the reciprocal of the moment of inertia as

[0063] $k$ is the sampling period number in the online identification process of the moment of inertia, and $J$ init is the iterative initial value of the moment of inertia of the rotary shaft servo system. At initialization, set $J$ init as the nominal moment of inertia of the motor shaft. For the convenience of formula derivation, in one embodiment of the present disclosure, the online identification algorithm directly performs recursive update on the identified value of the reciprocal of the moment of inertia in each sampling period, and the identified value of the moment of inertia can be obtained by taking the reciprocal of this value

[0064] (2) Input a preset motor speed command $\omega$ d with a phased speed regulation characteristic to the rotary shaft servo system; the numerical value of the motor speed command $\omega$ d changes no less than twice, and the time interval between two adjacent numerical value changes should be greater than the time required for the motor speed to adjust to the numerical value of the previous motor speed command;

[0065] (3) During the operation of the rotary shaft servo system, at a fixed time interval $T$ sPerform periodic sampling. During the k-th sampling period, obtain the sampling value ω(k) of the motor speed, the sampling value i(k) of the motor current, and the sampling value ω d (k) of the motor speed command of the current rotary shaft servo system. Perform low-pass filtering on the sampling value ω(k) of the motor speed and overwrite the original value; use the calculation model of the driving torque and the frictional torque to calculate the calculated value of the driving torque of the rotary shaft servo system and the calculated value of the frictional torque The time interval T of the sampling period s , supported by the hardware performance of the rotary shaft servo system, should be taken as small as possible;

[0066] (4) Use the following formula to calculate the observation error e(k) of the motor speed and the recursive compensation term u(k) of the motor speed in the k-th sampling period:

[0067]

[0068] In the formula, is the observed value of the motor speed in the k-th sampling period. After initialization in step (1), the value of the next sampling period is recursively updated in each sampling period according to step (7); k e is the compensation gain of the observation error of the motor speed, and the value of k e is non-negative and should not be too large. As long as it can ensure that the observed value of the motor speed can approach and track its sampling value, it is okay. The larger k e is, the greater the observation sensitivity of the motor speed, but it will increase the amplitude of the numerical fluctuation in the online identification process; ε is the tolerance threshold of the observed value of the motor speed to error interference, and the value is positive. It needs to be adjusted on-site. As long as it can ensure that the observed value of the motor speed can approach and track its sampling value, it is okay and should not be too large. The larger ε is, the greater the observation sensitivity of the motor speed, but it will increase the amplitude of the numerical fluctuation in the online identification process; sat(e) is the saturation function, as shown in the following formula:

[0069]

[0070] In the formula, Δ e is the saturation threshold of the observation error of the motor speed, and the value is positive. It needs to be adjusted on-site. As long as it can ensure that the observed curve of the motor speed is smooth and there is no chattering, it is okay and should not be too large. An overly large Δ e will affect the convergence speed of this algorithm; sgn(e) is the sign function;

[0071] (5) Use the following formula to calculate the robust activation coefficient l(k) in the k-th sampling period:

[0072]

[0073] In the formula, is the observed value of the load torque in the k-th sampling period. After the initialization in step (1), the value for the next sampling period is recursively updated in each sampling period according to step (7); and are respectively the calculated values of the driving torque and the friction torque in the k-th sampling period calculated in step (3); g is the convergence speed gain, which takes a negative value and its absolute value should not be too large, ensuring that the observed value of the load torque can converge stably to its actual value. The larger g is, the higher the observation sensitivity of the load torque, but the larger the amplitude of the numerical fluctuation of the observed value of the load torque, affecting the identification accuracy of the moment of inertia; Δ ω is the activation threshold of the motor speed fluctuation, which takes a positive value and needs to be adjusted on-site. Try to take a small value under the condition that the identified value of the reciprocal of the moment of inertia can converge stably. The larger Δ ω is, the stronger the ability of this algorithm to resist error interference, but the corresponding decrease in the identification accuracy of the moment of inertia; max{x1, x2} is the maximum value function, as shown in the following formula:

[0074]

[0075] (6) Using the following formula, calculate the derivative of the observed value of the motor speed, the derivative of the observed value of the load torque and the derivative of the identified value of the reciprocal of the moment of inertia in the k-th sampling period respectively as follows:

[0076]

[0077] (7) Using the following formula, recursively update the observed value of the motor speed the observed value of the load torque the identified value of the reciprocal of the moment of inertia for the (k + 1)-th sampling period as follows:

[0078]

[0079] In the formula, T s is the time interval between two adjacent sampling periods;

[0080] (8) Judge the preset motor speed command ω d with stage speed regulation characteristics. If the preset motor speed command ω d ends, then use the observed value of the motor speed in step (7) the observed value of the load torque the identified value of the reciprocal of the moment of inertia Taking the value in each sampling period as the ordinate and the online identification time as the abscissa, plot the observation curve of the motor speed, the observation curve of the motor load torque, and the convergence curve of the reciprocal identification value of the system moment of inertia. Obtain the identification result of the system moment of inertia from the convergence curve of the reciprocal identification value of the system moment of inertia. End the online identification algorithm; if the preset motor speed command ω d has not ended, then return to step (2). Continue with the recursive identification of the moment of inertia in the next sampling period, update the sampling period serial number k = k + 1, and repeat steps (2) to (8) until the identification result of the system moment of inertia is obtained.

[0081] According to the method for online identification of the moment of inertia for a rotary shaft servo system proposed in the embodiments of the present disclosure, it is possible to perform efficient, high-precision, and high-reliability online identification of the moment of inertia for the rotary shaft servo system of existing precision electromechanical equipment, and at the same time, perform real-time observation of the load torque of the servo motor. Under the condition that the established model is difficult to accurately describe the complex time-varying characteristics of the rotary shaft servo system, it overcomes the problems that the existing methods for obtaining the moment of inertia are easily affected by errors and lack robustness, effectively suppresses the adverse effects of model uncertainty factors on the algorithm accuracy and convergence, and the identified total system moment of inertia provides an important basis for on-site engineering and technical personnel to analyze the dynamic performance of the rotary shaft servo system, design high-quality controllers, and tune the parameters of the motor drive.

[0082] Corresponding to the above method for online identification of the moment of inertia of a rotary shaft servo system, the present disclosure also proposes an online identification device for the moment of inertia of a rotary shaft servo system, as Figure 6 shown, including:

[0083] A model establishment module, configured to establish a calculation model for the driving torque and frictional torque of the rotary shaft servo system;

[0084] A calculation module, configured to use the calculation model to perform online identification of the moment of inertia of the rotary shaft servo system according to the motor speed command of the rotary shaft servo system.

[0085] Embodiments of the present disclosure also propose an electronic device, which may include:

[0086] A memory, configured to store computer-executable instructions;

[0087] A processor, the processor being configured to execute:

[0088] Establish a calculation model for the driving torque and frictional torque of the rotary shaft servo system;

[0089] Use the calculation model to perform online identification of the moment of inertia of the rotary shaft servo system according to the motor speed command of the rotary shaft servo system.

[0090] Embodiments of the present disclosure also propose a computer-readable storage medium, on which a computer program is stored, and the computer program is used to cause the computer to execute:

[0091] Establish a calculation model for the driving torque and the frictional torque of the rotary shaft servo system;

[0092] Using the calculation model, online identify the moment of inertia of the rotary shaft servo system according to the motor speed command of the rotary shaft servo system.

[0093] The following combines the drawings to introduce in detail the flow of the online identification method for the moment of inertia of the rotary shaft servo system according to an embodiment of the present disclosure. For a clearer understanding of the content of the present disclosure, the example of the grinding wheel motorized spindle servo system of a precision roll grinder is used for illustration.

[0094] Figure 2 Shown is the grinding wheel motorized spindle servo system of a precision roll grinder, which is a typical rotary shaft servo system. The main components include: grinding wheel 1, grinding wheel connection assembly 2, spindle main body 3, rotor permanent magnet 4, stator winding 5, angle encoder 6, hydrostatic and hydrodynamic bearing group 7, motorized spindle housing 8, etc. This rotary shaft servo system consists of a motor module composed of a rotor permanent magnet 4 and a stator winding 5, and its electrical and mechanical characteristics are the same as those of a permanent magnet synchronous motor. The components of the total moment of inertia of the system specifically include: grinding wheel 1, grinding wheel connection assembly 2, spindle main body 3, and rotor permanent magnet 4. Apply the proposed online identification method for the moment of inertia of the rotary shaft servo system to the grinding wheel motorized spindle servo system, and conduct online identification of the moment of inertia of the rotary shaft servo system. The specific method steps are as follows:

[0095] In step 1, establish the calculation model for the driving torque and the frictional torque of the rotary shaft servo system as follows:

[0096]

[0097] In the formula, is the calculated value of the driving torque of the rotary shaft servo system; is the calculated value of the frictional torque of the rotary shaft servo system; K t is the torque coefficient of the motor. In this embodiment, the rotor permanent magnet 4 and the stator winding 5 form a motor module similar to a permanent magnet synchronous motor, and K is read in the motor driver t is 4.2 N·m / A. Due to complex factors such as magnetic saturation in the motor module, structural asymmetry between the rotor permanent magnet 4 and the stator winding 5, iron loss, and magnet eddy current loss, the calculated value of the driving torque obtained from the calculation model There is a modeling error between it and its actual value; i is the sampled value of the motor current; ω is the sampled value of the motor speed, which is calculated from the sampled value of the angle encoder 6; A, B, and C are constants. By conducting no-load tests on the rotary shaft servo system under different motor speed commands, plotting the scatter diagram of the calculated value of the steady-state driving torque and the sampled value of the steady-state motor speed, and further performing curve fitting on the functional relationship between the two, the values of A, B, and C are obtained. In this embodiment, the frictional torque actually received by the rotary shaft servo system is directly related to the hydrostatic and hydrodynamic bearing set 7. Since the viscous friction coefficient of the hydrostatic and hydrodynamic bearing set 7 will change continuously with the changes in motor speed, operating duration, working temperature, and lubricant properties, A, B, and C have complex time-varying characteristics and theoretically cannot be described by constant values. Specifically, it is manifested that under the same motor speed command, the calculated values of the steady-state driving torque obtained from different groups of no-load tests are different. To simplify the modeling and avoid repeatedly performing the above-mentioned curve fitting process on the calculation model of the frictional torque during the online identification process, a calculated value of the steady-state driving torque corresponding to each motor speed command is selected respectively, and a scatter diagram is further plotted. After fitting, the constant values A = -0.0008, B = 0.2495, and C = 0.4958 are obtained. Obviously, the calculated value of the frictional torque obtained from the calculation model will have a large modeling error with its actual value;

[0098] In step 2, using the calculation model, according to the motor speed command with a staged speed regulation characteristic of the rotary shaft servo system, the moment of inertia of the rotary shaft servo system is identified online. The online identification process is as Figure 1 shown, and specifically includes the following steps:

[0099] (1) Set the initial sampling period number k = 1, and initialize the observed value of the motor speed to 0, the observed value of the load torque to 0, the identified value of the moment of inertia to J init , and the reciprocal identified value of the moment of inertia is as shown in the following formula:

[0100]

[0101] where k is the sampling period number in the online identification process of the moment of inertia; J init is the iterative initial value of the moment of inertia of the rotary shaft servo system. When initializing, set J init to the nominal moment of inertia of the motor shaft. In this embodiment, the motor shaft is specifically composed of the main shaft body 3 and the rotor permanent magnet 4, and the nominal value of the total moment of inertia of the two is 0.0129 kg·m 2; For the convenience of formula derivation, in this embodiment, the online identification algorithm directly performs recursive update on the reciprocal identification value of the moment of inertia within each sampling period and then takes the reciprocal of this value to obtain the identification value of the moment of inertia

[0102] (2) Input a preset motor speed command ω with a phased speed regulation characteristic to the rotating shaft servo system d ; In this embodiment, keep the rotating shaft servo system always in an unloaded state, that is, the actual value of the load torque is zero; the selected motor speed command ω d has a phased acceleration characteristic. Specifically, the initial motor speed command is 200 rpm, and the motor speed command is increased by 100 rpm every 42 s until the motor speed command reaches 1000 rpm and remains for 42 seconds, and then the motor speed command ω d ends;

[0103] (3) During the operation of the rotating shaft servo system, perform periodic sampling at a fixed time interval T s . In the k-th sampling period, obtain the sampling value ω(k) of the motor speed, the sampling value i(k) of the motor current, and the sampling value ω d (k) of the motor speed command of the current rotating shaft servo system. Perform low-pass filtering on the sampling value ω(k) of the motor speed and overwrite the original value. The low-pass filtering cut-off frequency is set to 10 Hz; use the calculation model of the driving torque and the friction torque to calculate the calculated value of the driving torque of the rotating shaft servo system and the calculated value of the friction torque The time interval T of the sampling period s , supported by the hardware performance of the rotating shaft servo system, should be taken as small as possible. In this embodiment, set T s = 0.0005 s;

[0104] (4) Use the following formula to calculate the observation error e(k) of the motor speed and the recursive compensation term u(k) of the motor speed in the k-th sampling period:

[0105]

[0106] In the formula, is the observed value of the motor speed in the k-th sampling period. After initialization in step (1), the value of the next sampling period is recursively updated in each sampling period according to step (7); k e is the compensation gain of the observation error of the motor speed, and the value of k e is non-negative and should not be too large, as long as it can ensure that the observed value of the motor speed can approach and track its sampling value. In this embodiment, set k e= 10; ε is the tolerance threshold of the observed value of the motor speed to the error interference, and it takes a positive value. As long as it is ensured that the observed value of the motor speed can approach and track its sampled value, it should not be too large. In this embodiment, ε = 6000 is set; sat(e) is a saturation function, as shown in the following formula:

[0107]

[0108] In the formula, Δ e is the saturation threshold of the observed error of the motor speed, and it takes a positive value. As long as it is ensured that the observed curve of the motor speed is smooth and there is no chattering, it should not be too large. In this embodiment, Δ e = 1000 is set; sgn(e) is a sign function;

[0109] (5) Use the following formula to calculate the robust activation coefficient l(k) at the k-th sampling period:

[0110]

[0111] In the formula, is the observed value of the load torque at the k-th sampling period. After the initialization in step (1), the value for the next sampling period is recursively updated in each sampling period according to step (7); and are respectively the calculated values of the driving torque and the friction torque at the k-th sampling period calculated in step (3); g is the convergence speed gain, and it takes a negative value, and its absolute value should not be too large. As long as it is ensured that the observed value of the load torque can converge stably to its actual value, in this embodiment, g = -10 is set; Δ ω is the activation threshold of the motor speed fluctuation, and it takes a positive value. Under the condition of ensuring that the identified value of the reciprocal of the moment of inertia can converge stably, it should be taken as small as possible. In the embodiment, Δ ω = 1.0 rad / s is set; max{x1, x2} is a maximum value function, as shown in the following formula:

[0112]

[0113] (6) Use the following formula to calculate the derivative of the observed value of the motor speed the derivative of the observed value of the load torque and the derivative of the identified value of the reciprocal of the moment of inertia respectively as follows:

[0114]

[0115] (7) Use the following formula for the observed value of the motor speed at the (k + 1)-th sampling period the observed value of the load torque Reciprocal identification value of moment of inertia Perform recursive update as follows:

[0116]

[0117] In the formula, T s is the time interval between two adjacent sampling periods. In this embodiment, T s is set to 0.0005 s;

[0118] (8) Judge the preset motor speed command ω d with stage speed regulation characteristics. If the preset motor speed command ω d ends, then use the observed value of the motor speed in step (7) the observed value of the load torque the reciprocal identification value of the moment of inertia Take the value at each sampling period as the ordinate and the online identification time as the abscissa, and plot the observed curve of the motor speed, the observed curve of the motor load torque, and the convergence curve of the reciprocal identification value of the system moment of inertia. Obtain the identification result of the system moment of inertia from the convergence curve of the reciprocal identification value of the system moment of inertia End the online identification algorithm; if the preset motor speed command ω d has not ended, then return to step (2). Continue to perform the recursive identification of the moment of inertia in the next sampling period, update the sampling period serial number k = k + 1, and repeat steps (2) to (8) until the identification result of the system moment of inertia is obtained

[0119] In this embodiment, after the preset motor speed command ω d ends, the observed curve of the motor speed, the observed curve of the motor load torque, and the convergence curve of the reciprocal identification value of the system moment of inertia are plotted as shown in Figure 3 、 Figure 4 and Figure 5 respectively; the abscissa in the three figures represents the change range of the online identification time, with the unit of s, Figure 3 the ordinate of Figure 4 represents the change range of the observed value of the motor speed, with the unit of rad / s, Figure 4 the ordinate of Figure 4 represents the change range of the motor load torque, with the unit of N·m, Figure 5 the ordinate of Figure 5 represents the change range of the reciprocal identification value of the system moment of inertia, with the unit of kg -1 m -2 ; It can be clearly seen from Figure 3 that the method proposed by the present invention can effectively reconstruct the real-time change of the motor speed; Figure 4The dashed line and the solid line in it respectively represent the actual value of the motor load torque and the observed value obtained by using this method. It can be seen from the results that the method proposed by the present invention can effectively observe the real-time load torque of the motor, while Figure 4 the error between the solid line and the dashed line in it reflects the system uncertainties mainly composed of the calculation error of the frictional torque. This result also shows that the proposed online identification method of the moment of inertia can effectively identify the modeling error of the physical model in step 1; from Figure 5 it can be seen that after a finite number of transient speed regulations of the motor speed, the reciprocal identification value of the system moment of inertia can stably converge from the initial value to the final identification value; further reading the reciprocal identification value of the system moment of inertia is 1.745 kg -1 m -2 , then the identification result of the system moment of inertia is 1 / 1.745 = 0.5731 kg·m 2 , specifically, the identified value of the system moment of inertia is the total moment of inertia of the rotating body composed of components such as the grinding wheel 1, the grinding wheel connection assembly 2, the main shaft body 3, and the rotor permanent magnet 4. It can be seen from the results that this method can achieve efficient, high-precision, and high-reliability online identification of the moment of inertia of the rotary shaft servo system under the condition that there is a large modeling error in the calculation models of the driving torque and the frictional torque in step 1, and has strong model robustness.

[0120] It should be noted that in the embodiments of the present disclosure, the so-called processor may be a central processing unit (CPU), or may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor, or the processor may also be any conventional processor, etc. The memory can be used to store the computer program and / or module. By running or executing the computer program and / or module stored in the memory, and by calling the data stored in the memory, the processor realizes various functions of the online identification method for the moment of inertia of the rotary shaft servo system. The memory mainly includes a program storage area and a data storage area. Among them, the program storage area can store an operating system, application programs required for at least one function (such as a sound playback function, an image playback function, etc.); the data storage area can store the data created by the operating system during the running of the application program (such as audio data, graphic data, etc.). In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, a storage device of at least one magnetic disk, or a flash device.

[0121] Based on such understanding, all or part of the processes in the above-described method embodiments of the present disclosure can also be completed by instructing relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above-described method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file, or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disc, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc. It should be noted that the above-described device embodiments are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. In addition, in the attached drawings of the device embodiments provided by the present disclosure, the connection relationship between the modules indicates that they have a communication connection, which can be specifically implemented as one or more communication buses or signal lines. Those of ordinary skill in the art can understand and implement it without creative efforts.

[0122] The above is the preferred implementation manner of the present disclosure. It should be pointed out that for those of ordinary skill in the art in the technical field of the present disclosure, without departing from the principle of the present disclosure, several improvements and refinements can be made, and these improvements and refinements are also regarded as the protection scope of the present disclosure.

Claims

1. An online identification method for the moment of inertia of a rotary shaft servo system, characterized in that, Comprising: Establish a calculation model for the driving torque and frictional torque of the rotary shaft servo system; Utilize the calculation model to perform online identification of the moment of inertia of the rotary shaft servo system according to the motor speed command of the rotary shaft servo system; The calculation model of the driving torque and frictional torque is as follows: In the formula, is the calculated value of the driving torque of the rotary shaft servo system, is the calculated value of the frictional torque of the rotary shaft servo system, is the torque coefficient of the motor, is the sampled value of the motor current, is the sampled value of the motor speed, , and are constants; According to the motor speed Calculate the calculated value of the driving torque of the rotary shaft servo system And the calculated value of the frictional torque of the rotary shaft servo system , and then according to the motor speed , the calculated value of the driving torque of the rotary shaft servo system And the calculated value of the frictional torque of the rotary shaft servo system Obtain the identified value of the reciprocal of the moment of inertia , for the identified value of the reciprocal of the moment of inertia Take the reciprocal to obtain the identified value of the moment of inertia .

2. The on-line identification method of moment of inertia according to claim 1, characterized in that Utilize the calculation model to perform online identification of the moment of inertia of the rotary shaft servo system according to the motor speed command with stage speed regulation characteristics of the rotary shaft servo system, including: (1) Set the initial sampling period number , initialize the observed value of the motor speed to 0, the observed value of the load torque to 0, the identified value of the moment of inertia is , the identified value of the reciprocal of the moment of inertia is ; k is the sampling period serial number in the on-line identification process of the moment of inertia, is the initial iteration value of the moment of inertia of the rotary shaft servo system; (2)Input a motor speed command with preset stage speed regulation characteristics to the rotary shaft servo system ; (3) During the operation of the rotary shaft servo system, at fixed time intervals perform periodic sampling. In the k th sampling period, obtain the sampling value of the motor speed , the sampling value of the motor current and the sampling value of the motor speed command . Perform low-pass filtering on the sampling value of the motor speed and overwrite the original value; use the calculation model of the driving torque and the friction torque to calculate the calculated value of the driving torque and the calculated value of the friction torque ; (4) Using the following formula, calculate the observation error k of the motor speed and the recursive compensation term of the motor speed for the wherein, is the observed value of the motor speed at the k -th sampling period, is the compensation gain for the observed error of the motor speed, and its value is non-negative; is the tolerance threshold of the observed value of the motor speed to error interference, and its value is positive, is a saturation function, as shown in the following formula: wherein, is the saturation threshold of the observation error of the motor speed, and the value is positive; is the sign function; (5) Using the following formula, calculate the robust activation coefficient for the k th sampling period : In the formula, is the observed value of the load torque at the k -th sampling period, and are respectively the calculated values of the driving torque and the friction torque at the k -th sampling period; g is the convergence speed gain, which takes a negative value, is the activation threshold of the motor speed fluctuation, which takes a positive value; is the maximum value function, as shown in the following formula: (6) Using the following formula, calculate the derivatives of the motor speed observation value, the load torque observation value, and the reciprocal of the moment of inertia identification value for the k th sampling period respectively as follows: Derivative of the motor speed observation value, Derivative of the load torque observation value, and Derivative of the reciprocal of the moment of inertia identification value are as follows: (7) Using the following formula, the observed value of the motor speed in the th sampling period, the observed value of the load torque and the identified value of the reciprocal of the moment of inertia are recursively updated as follows: In the formula, is the time interval between two adjacent sampling periods; (8) Judge the preset motor speed command with stage speed regulation characteristics If the preset motor speed command ends, then use the observed value of the motor speed in step (7) and the observed value of the load torque and the identified value of the reciprocal of the moment of inertia with the values in each sampling period as the ordinate and the online identification time as the abscissa, plot the observed curve of the motor speed, the observed curve of the motor load torque, and the convergence curve of the identified value of the reciprocal of the system moment of inertia, and obtain the identification result of the system moment of inertia from the convergence curve of the identified value of the reciprocal of the system moment of inertia If the preset motor speed command has not ended, then return to step (2).

3. An online identification device for the moment of inertia of a rotary shaft servo system, characterized in that, Comprising: A model establishment module for establishing a calculation model for the driving torque and frictional torque of the rotary shaft servo system; A calculation module for performing online identification of the moment of inertia of the rotary shaft servo system according to the motor speed command of the rotary shaft servo system by using the calculation model; The calculation model of the driving torque and frictional torque is as follows: Wherein, is the calculated value of the driving torque of the rotary shaft servo system, is the calculated value of the frictional torque of the rotary shaft servo system, is the torque coefficient of the motor, is the sampled value of the motor current, is the sampled value of the motor speed, , and are constants; According to the motor speed Calculate the calculated value of the driving torque of the rotary shaft servo system And the calculated value of the frictional torque of the rotary shaft servo system , and then according to the motor speed , the calculated value of the driving torque of the rotary shaft servo system And the calculated value of the frictional torque of the rotary shaft servo system Obtain the identified value of the reciprocal of the moment of inertia , for the identified value of the reciprocal of the moment of inertia Take the reciprocal to obtain the identified value of the moment of inertia .

4. An electronic device, characterized in that, Comprising: A memory for storing computer-executable instructions; A processor, the processor being configured to execute the online identification method for the moment of inertia of the rotary shaft servo system according to any one of claims 1-2.

5. A computer-readable storage medium, characterized in that, A computer program is stored on the computer-readable storage medium, and the computer program is used to cause the computer to execute the online identification method for the moment of inertia of the rotary shaft servo system according to any one of claims 1-2.

Images