A hysteresis compensation control method for piezoelectric actuators to improve tangential amplitude

Abstract

The present invention relates to the technical field of control systems, and more specifically to a piezoelectric driver hysteresis compensation control method for improving tangential amplitude, comprising: obtaining input voltage data and displacement data of the piezoelectric driver to construct a voltage hysteresis curve; performing segmented processing on the voltage hysteresis curve to obtain local hysteresis curves; obtaining a variation difference coefficient based on the difference in local change trends between each two local hysteresis curves and the position distribution between data points on each two local hysteresis curves; obtaining a curve variation characteristic value based on the distribution range of the variation difference coefficients corresponding to all local hysteresis curves on the voltage hysteresis curve, combined with the distribution of the change trends of all data points on the voltage hysteresis curve; determining the number of operators based on the curve variation characteristic value, constructing a PI model of the piezoelectric driver, and performing hysteresis compensation control on the piezoelectric driver. The present invention can dynamically obtain the most appropriate number of operators and accurately and smoothly describe the variation characteristics of the hysteresis curve.

Description

Technical Field

[0001] The present invention relates to the technical field of control systems, and in particular to a piezoelectric driver hysteresis compensation control method for improving tangential amplitude. Background Art

[0002] A piezoelectric ceramic actuator is a type of actuator that uses the inverse piezoelectric effect of piezoelectric materials to generate mechanical motion. Through the friction between the stator and the mover, the microscopic high-frequency vibration of the stator is converted into the macroscopic linear motion of the mover. It has the advantages of compact structure, high precision, self-locking when power is off, and no electromagnetic interference.

[0003] Existing piezoelectric ceramic actuators generally operate based on the L1B2 dual-vibration mode principle. These actuators operate in the first-order longitudinal vibration mode (L1) and the second-order bending vibration mode (B2). The first-order longitudinal vibration and the second-order bending vibration are combined to produce an elliptical motion, which in turn propels the sliding component into linear motion. These actuators are widely used in precision instruments, micro-electromechanical systems, aerospace, and other fields. However, the generation of second-order bending vibrations in piezoelectric plates requires a relatively high voltage, which results in significant mechanical losses and significantly reduces the maximum motion speed of the driven element. Furthermore, because the electric dipoles within the piezoelectric material cannot return to their normal state after deflection, there is no one-to-one correspondence between the input and output signals, resulting in hysteresis and nonlinear characteristics in the system.

[0004] Existing methods use play operators to construct PI models to implement driver control methods. The number of play operators is generally set to a fixed value based on empirical values. This method lacks consideration of the characteristic degree of the changing trend of the voltage hysteresis curve and cannot dynamically select the appropriate number of play operators for different degrees of change. It is difficult to ensure that the voltage hysteresis curve can be accurately described while reducing the computational cost. Summary of the Invention

[0005] In order to solve the technical problem that the number of play operators in the existing method is a fixed value, resulting in poor hysteresis compensation effect on the piezoelectric driver, the purpose of the present invention is to provide a piezoelectric driver hysteresis compensation control method for improving the tangential amplitude. The technical solution adopted is as follows:

[0006] Obtain input voltage data and displacement data of the piezoelectric actuator and construct a voltage hysteresis curve of the piezoelectric actuator;

[0007] According to the change trend of each data point on the voltage hysteresis curve, the voltage hysteresis curve is segmented to obtain each local hysteresis curve;

[0008] According to the difference of local change trends between each two local hysteresis curves and the position distribution between the data points on each two local hysteresis curves, the change difference coefficients between different local hysteresis curves are obtained;

[0009] According to the distribution range of the change difference coefficients corresponding to all local hysteresis curves on the voltage hysteresis curve, combined with the quantity distribution and range distribution of the change trends of all data points on the voltage hysteresis curve, the curve change characteristic value is obtained;

[0010] The number of operators is determined based on the characteristic value of the curve change, a PI model of the piezoelectric driver is constructed, and hysteresis compensation control is performed on the piezoelectric driver.

[0011] Preferably, obtaining the variation difference coefficients between different local hysteresis curves according to the difference in local variation trends between each two local hysteresis curves and the position distribution between data points on each two local hysteresis curves specifically includes:

[0012] Obtain the slope of each data point on the voltage hysteresis curve according to all data points within the neighborhood of each data point on the voltage hysteresis curve;

[0013] Obtain any two local hysteresis curves with the same change trend, match the two local hysteresis curves based on the dynamic time warping algorithm, and obtain the matching relationship between each data point;

[0014] According to the slope difference and position distribution between any two local hysteresis curves that have a matching relationship, the change difference coefficient between any two local hysteresis curves is obtained.

[0015] Preferably, obtaining the variation difference coefficient between any two local hysteresis curves according to the slope difference and position distribution between any two local hysteresis curves having a matching relationship specifically includes:

[0016] For two local hysteresis curves with the same slope sign, the absolute value of the difference in the slopes of the data points with matching relationship between the two local hysteresis curves is used as the trend difference factor; the Euclidean distance between the corresponding coordinates of the data points with matching relationship between the two local hysteresis curves is used as the distance difference factor;

[0017] Based on the balanced distribution of the ratio between the trend difference factor and the distance difference factor, the change difference coefficient between any two local hysteresis curves is determined.

[0018] Preferably, obtaining the slope of each data point on the voltage hysteresis curve based on all data points within the neighborhood of each data point on the voltage hysteresis curve specifically includes:

[0019] Obtain all data points within a window of a preset length containing each data point for straight line fitting, and obtain the slope of the fitted line as the slope of the corresponding data point.

[0020] Preferably, the curve change characteristic value is obtained based on the distribution range of the change difference coefficients corresponding to all local hysteresis curves on the voltage hysteresis curve, combined with the quantity distribution and range distribution of the change trends of all data points on the voltage hysteresis curve, and specifically includes:

[0021] The first complexity coefficient is obtained according to the distribution range of all the variation difference coefficients;

[0022] A second complex coefficient is obtained according to the distribution quantity of the slope of each data point on the voltage hysteresis curve and the distribution probability of the slope;

[0023] The ratio of the first complex coefficient to the second complex coefficient is rounded down to obtain the curve change characteristic value.

[0024] Preferably, obtaining the first complexity coefficient according to the distribution range of all variation difference coefficients specifically includes:

[0025] The difference between the maximum and minimum values ​​of all variation coefficients is taken as the first complex coefficient.

[0026] Preferably, obtaining the second complex coefficient according to the distribution quantity and distribution probability of the slope of each data point on the voltage hysteresis curve specifically includes:

[0027] The probability value of the slope of each data point on the voltage hysteresis curve is obtained based on the kernel density estimation algorithm; the number of all data points with the same slope as each data point is obtained; and the second complex coefficient is obtained by weighted summing the number of data points using the probability value.

[0028] Preferably, the step of segmenting the voltage hysteresis curve according to the variation trend of each data point on the voltage hysteresis curve to obtain each local hysteresis curve specifically includes:

[0029] Each extreme point of the voltage hysteresis curve is obtained, and each extreme point is used as a segmentation point to segment the voltage hysteresis curve to obtain each local hysteresis curve.

[0030] Preferably, determining the number of operators based on the curve change characteristic value and constructing the PI model of the piezoelectric driver specifically includes:

[0031] The sum of the curve change characteristic value and the preset initial operator number is used as the number of unilateral play operators to construct the PI model of the piezoelectric actuator.

[0032] Preferably, the step of acquiring input voltage data and displacement data of the piezoelectric driver and constructing a voltage hysteresis curve of the piezoelectric driver specifically includes:

[0033] The input voltage of the piezoelectric driver is used as the horizontal coordinate and the displacement of the piezoelectric driver is used as the vertical coordinate, and curve fitting is performed to construct the voltage hysteresis curve of the piezoelectric driver.

[0034] The embodiments of the present invention have at least the following beneficial effects:

[0035] The present invention first constructs a voltage hysteresis curve, and performs local segmentation processing based on the change trend of each data point, so that the same change trend is obtained on each local hysteresis curve, which facilitates the subsequent analysis of the simplicity and complexity of the change characteristics of the voltage hysteresis curve. Then, from the two aspects of the difference in local change trends and the position distribution between data points, the change difference coefficients between different local hysteresis curves are comprehensively quantified to reflect the degree of difference in the change trend. Furthermore, the complexity of the overall change trend of the voltage hysteresis curve is measured, taking into account the distribution range of the change difference coefficient and the distribution of the local change trend of the data points. It is possible to apply a smaller number of operators to voltage hysteresis curves with lower change trend complexity, and apply a larger number of operators to voltage hysteresis curves with higher change trend complexity, dynamically obtain the most appropriate number of operators, accurately and smoothly describe the change characteristics of the hysteresis curve, and construct a PI model and a PI inverse model based on the play operator, thereby realizing hysteresis compensation control and stable drive control of the piezoelectric driver, which can accurately describe the voltage hysteresis curve while reducing the computational cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0037] Figure 1 This is a flowchart of the steps of a piezoelectric driver hysteresis compensation control method for improving tangential amplitude provided by the present invention;

[0038] Figure 2 It is a structural schematic diagram of the electric ceramic driver provided by the present invention;

[0039] Figure 3 It is a structural diagram of the simulation experiment platform provided by the present invention;

[0040] Figure 4 is a flowchart of the steps of the method for obtaining the coefficient of variation provided by the present invention;

[0041] Figure 5 It is a flowchart of the steps of the method for obtaining the characteristic value of curve change provided by the present invention. DETAILED DESCRIPTION

[0042] To further illustrate the technical means and effects of the present invention to achieve the intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, describes in detail the specific implementation, structure, features, and effects of a piezoelectric driver hysteresis compensation control method for improving the tangential amplitude proposed by the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable form.

[0043] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0044] The following describes in detail a specific solution of a hysteresis compensation control method for a piezoelectric driver for improving the tangential amplitude provided by the present invention with reference to the accompanying drawings.

[0045] See also Figure 1 , which shows a flowchart of a piezoelectric driver hysteresis compensation control method for improving tangential amplitude provided by one embodiment of the present invention, the method comprising the following steps:

[0046] Step S100 , obtaining input voltage data and output displacement data of the piezoelectric driver, and constructing a voltage hysteresis curve of the piezoelectric driver.

[0047] As a specific example, the present embodiment adopts an electro-ceramic driver structure such as Figure 2 As shown in the figure, the piezoelectric ceramic actuator consists of two piezoelectric ceramic plates, a metal substrate and a friction plate. In order to obtain the displacement-voltage driving change of the piezoelectric ceramic actuator, a piezoelectric ceramic actuator is constructed as shown in the figure. Figure 3 The simulation experimental platform shown. The normal displacement resolution of the piezoelectric ceramic driver is 0.01µm, and the sampling frequency is 1000Hz. This embodiment uses the software in the computer to process the piezoelectric ceramic driver signal and form a digital voltage signal. The signal is modulated by simulation control and connected to the piezoelectric ceramic driver to produce a series of actions. When the piezoelectric ceramic driver is deformed under voltage excitation, its output displacement is dynamically sampled by the sensor. The sensor converts the physical displacement signal into an electrical signal in real time, forms a digital input after analog-to-digital conversion, and synchronously feeds back to the computer software system for subsequent processing.

[0048] Based on this, each displacement signal corresponding to each input signal is collected. Specifically, the input voltage of the piezoelectric driver is used as the horizontal coordinate and the displacement of the piezoelectric driver is used as the vertical coordinate to perform curve fitting and construct the voltage hysteresis curve of the piezoelectric driver.

[0049] More specifically, this embodiment uses the acquired piezoelectric ceramic driver voltage-displacement signal as the horizontal and vertical axes to construct a voltage hysteresis curve for the piezoelectric ceramic driver. This hysteresis curve, with the horizontal axis representing the piezoelectric ceramic driver's input voltage and the vertical axis representing the piezoelectric ceramic driver's output displacement, provides a data foundation for subsequent analysis of hysteresis characteristic trends.

[0050] Step S200 , segmenting the voltage hysteresis curve according to the change trend of each data point on the voltage hysteresis curve to obtain each local hysteresis curve.

[0051] The piezoelectric curve of a piezoelectric ceramic actuator exhibits a distinctly asymmetric characteristic. Specifically, the voltage hysteresis curve exhibits a first phase with a large slope change, exhibiting a concave function. This means that the displacement gradually slows down with increasing voltage, and a second phase with a more gradual, linear slope change. This means that the displacement becomes more consistent with increasing voltage. Therefore, the voltage hysteresis curve must first be segmented according to its varying trends. Local segments with the same trend can then be grouped together to facilitate further characterization of the degree of variation within these segments.

[0052] Specifically, each extreme point of the voltage hysteresis curve is obtained, and each extreme point is used as a segmentation point to segment the voltage hysteresis curve to obtain each local hysteresis curve. It should be understood that the method of obtaining extreme points on the voltage hysteresis curve is a well-known technology and will not be further described here. Dividing the voltage hysteresis curve into intervals with relatively consistent characteristics, namely local hysteresis curves, facilitates subsequent accurate fitting using methods such as the PI model and the play operator, reducing modeling complexity.

[0053] Step S300 , obtaining a variation difference coefficient between different local hysteresis curves according to the difference in local variation trends between each two local hysteresis curves and the position distribution between data points on each two local hysteresis curves.

[0054] Since the piezoelectric material used in piezoelectric ceramic actuators is difficult to restore to its original state after deflection, piezoelectric ceramic actuators will exhibit a certain degree of hysteresis, which is manifested as a multi-value mapping feature on the hysteresis curve of the piezoelectric ceramic actuator. That is, for the same input voltage value, the output displacement may correspond to multiple different values. Typically, the PI model is used to characterize the piezoelectric characteristics of piezoelectric ceramic actuators, but existing PI models mostly use the form of equally spaced thresholds, ignoring the current-displacement change characteristics of the piezoelectric material of the piezoelectric ceramic actuator itself. The traditional PI model fits the actual displacement curve by constructing a PI model by weighted linear superposition of multiple thresholds, but the characteristics exhibited by the piezoelectric ceramic actuator at different positions are different.

[0055] Each of the two adjacent local hysteresis curves is located on both sides of an extreme point, and the change trend of their local curve segments changes, that is, the overall slope of the two adjacent local hysteresis curves is different, and each local hysteresis curve represents a local curve segment with the same change trend. The change trend reflects, to a certain extent, the inherent piezoelectric change characteristics of the piezoelectric material of the piezoelectric drive motor. By calculating the slope change of the local hysteresis curve, the mutation position of the piezoelectric characteristics in the hysteresis curve can be obtained. By calculating the curve difference coefficients of different local hysteresis curves, the mutation state between different local hysteresis curves can be effectively analyzed. If the piezoelectric characteristics of the local hysteresis curve mutate, the curve difference coefficient calculated at this time will also be large. When constructing the PI model of the piezoelectric drive motor, the curve difference coefficients of different hysteresis curve segments are used for adaptive adjustment.

[0056] As a specific example, Figure 4 The method for obtaining the change difference coefficients between different local hysteresis curves can be implemented by steps S301 to S303.

[0057] Step S301 : obtaining the slope of each data point on the voltage hysteresis curve according to all data points in the neighborhood of each data point on the voltage hysteresis curve.

[0058] To avoid excessive computational cost due to an excessive number of data point samples, this embodiment calculates the slope of each data point on the voltage hysteresis curve by setting a local window. Specifically, all data points within a window of a preset length containing each data point are obtained for linear fitting, and the slope of the fitted line is obtained as the slope of the corresponding data point.

[0059] As a specific example, a window of a preset length centered around each data point is obtained for slope calculation. The preset length can be 3 or 5, and the implementer can select this value based on the specific implementation scenario. In particular, if the local window cannot be obtained for the data point at the initial position on the voltage hysteresis curve, the slope calculation can be performed by obtaining a window of the data point and the N-1 data points closest to it, where N is the preset length. For example, when the preset length is 5, the first data point on the voltage hysteresis curve and the four adjacent data points to its left are obtained to form a window. For the second data point of the voltage hysteresis curve, the first to fifth data points form the window of the second data point.

[0060] It should be noted that the calculation method of the slope is a well-known technology and will not be introduced in detail here.

[0061] Step S302 : obtaining any two local hysteresis curves with the same change trend, matching the two local hysteresis curves based on a dynamic time warping algorithm, and obtaining a matching relationship between each data point.

[0062] Considering that the two local hysteresis curves for which variance analysis is needed have different lengths, meaning that the number of points contained in different local hysteresis curves cannot correspond one to another, the dynamic time warping algorithm is used to obtain the corresponding relationship between the data points of the two local hysteresis curves. For ease of description, in the subsequent steps, two data points with a matching relationship will be referred to as a matching pair.

[0063] It should be noted that the same change trend means that the two local hysteresis curves exhibit the same concavity and convexity, that is, both exhibit concavity or both exhibit convexity. When the local hysteresis curve changes linearly, the two local hysteresis curves both exhibit an upward trend or both exhibit a downward trend.

[0064] Step S303 : obtaining a variation difference coefficient between any two local hysteresis curves according to the slope difference and position distribution between any two local hysteresis curves that have a matching relationship.

[0065] Specifically, for two local hysteresis curves with the same slope sign, the absolute value of the difference in the slopes of the data points with a matching relationship between the two local hysteresis curves is used as the trend difference factor; the Euclidean distance of the corresponding coordinates of the data points with a matching relationship between the two local hysteresis curves is used as the distance difference factor; based on the balanced distribution of the ratio between the trend difference factor and the distance difference factor, the change difference coefficient between any two local hysteresis curves is determined.

[0066] It should be noted that the change trend in each local hysteresis curve is the same, that is, the slope signs of all data points in the same local hysteresis curve are the same, for example, both positive or both negative. Two local hysteresis curves with the same change trend can be directly obtained through the sign of the slope, for example, any two local hysteresis curves with the same positive slope sign can be obtained, or any two local hysteresis curves with the same negative slope sign can be obtained.

[0067] As a specific example, take any two local hysteresis curves with the same slope sign as an example, and record them as local hysteresis curve A and local hysteresis curve B respectively. Then, the change difference coefficient between local hysteresis curve A and local hysteresis curve B can be expressed as:

[0068]

[0069] in, Indicates the coefficient of variation between the local hysteresis curve A and the local hysteresis curve B, It represents the number of matching pairs between local hysteresis curve A and local hysteresis curve B, that is, the number of matching relationships. represents the slope of the data point in the i-th matching pair on the local hysteresis curve A, represents the slope of the data point in the ith matching pair on the local hysteresis curve B, represents the Euclidean distance between the corresponding coordinates of the data point in the i-th matching pair on the local hysteresis curve A and the data point in the i-th matching pair on the local hysteresis curve B.

[0070] It is the trend difference factor, which reflects the slope difference between the corresponding data points in the matching relationship on the two local hysteresis curves. The smaller its value is, the smaller the difference in the change trend between the two is. The distance difference factor reflects the straight-line distance between matching data points on two local hysteresis curves. It simultaneously considers both trend analysis and spatial position, measuring the difference in variation between two local hysteresis curves. When the distance difference factor is small, the slope differences of closely spaced points have a greater impact on the overall difference coefficient, reflecting sensitivity to local details. When the distance difference factor is large, it avoids pseudo-differences caused by large coordinate distances dominating the calculation (such as the natural deviation of the loop endpoints).

[0071] Step S400 , obtaining a curve change characteristic value based on the distribution range of the change difference coefficients corresponding to all local hysteresis curves on the voltage hysteresis curve and the quantity distribution and range distribution of the change trends of all data points on the voltage hysteresis curve.

[0072] Because the basic play operator is symmetric about the origin, which is inconsistent with the asymmetric characteristics of the voltage hysteresis curve, the piezoelectric characteristics of the PI model constructed using the basic play operator will have a large error compared to the actual model. Furthermore, in actual applications, the setting of the number of thresholds in the play operator currently relies heavily on experimental experience. More thresholds will produce a smoother and more accurate description of the hysteresis curve, but the calculation time will increase. Therefore, this embodiment dynamically determines the number of play operators based on the changing trend of the voltage hysteresis curve, ensuring an accurate description of the voltage hysteresis curve while reducing the calculation cost.

[0073] In the first step, the unilateral play operator is used as the basic play operator. The unilateral play operator has good asymmetric characteristics and is consistent with the changes in the hysteresis curve of the piezoelectric ceramic driver, thereby more accurately describing the changing characteristics of the voltage hysteresis curve.

[0074] In the second step, the slope of the voltage hysteresis curve reflects the hysteresis change characteristics of the piezoelectric ceramic driver. The number of unilateral play operators is dynamically adjusted based on the change in the slope of the voltage hysteresis curve. First, the degree of dynamic adjustment of the number of unilateral play operators is quantified.

[0075] As a specific example, Figure 5 As shown, the method for obtaining the curve change characteristic value can be implemented by steps S401 to S403.

[0076] Step S401: obtaining a first complexity coefficient according to the distribution range of all variation difference coefficients.

[0077] Specifically, the difference between the maximum value and the minimum value of all variation difference coefficients is taken as the first complex coefficient.

[0078] It should be understood that the first complexity coefficient is the range of the variation difference coefficients corresponding to all local hysteresis curves on the voltage hysteresis curve, which represents the distribution range of the variation difference coefficients. The variation difference coefficient represents the degree of difference in the variation trends between local hysteresis curves with the same variation trend. The greater the difference, the more complex the local data on the voltage hysteresis curve.

[0079] Based on this, the greater the range of the coefficient of variation between two local hysteresis curves with the same variation trend on the voltage hysteresis curve, the more complex the variation trend of the local data on the voltage hysteresis curve is, and the larger the value of the corresponding first complexity coefficient. In other words, the first complexity coefficient represents the difference and complexity of the variation trend between local data on the voltage hysteresis curve. The greater the difference, the more complex the curve, and the more operators should be used.

[0080] Step S402 : obtaining a second complexity coefficient according to the distribution quantity and distribution probability of the slope of each data point on the voltage hysteresis curve.

[0081] Specifically, a probability value of the slope of each data point on the voltage hysteresis curve is obtained based on a kernel density estimation algorithm; the number of all data points with the same slope as each data point is obtained; and the second complex coefficient is obtained by weighted summing the number of data points using the probability value.

[0082] It should be understood that the slope of each data point on the voltage hysteresis curve can be considered a sample of a random variable. The probability value of the slope at each data point can be obtained using a kernel density estimation algorithm. This technique is well known and will not be further described here. The probability value of each data point on the voltage hysteresis curve represents the density distribution of the corresponding slope value.

[0083] By statistically calculating the frequency and corresponding probability of the data slope values ​​in the hysteresis curve, the slope change characteristics of the voltage hysteresis curve can be calculated. If the slope value distribution is relatively concentrated, it means that the linear characteristics of the hysteresis curve are more obvious. At this time, the calculated number of slope values ​​of the hysteresis curve is distributed more densely and the corresponding probability value is also larger.

[0084] By statistically analyzing the frequency of occurrence and corresponding probability distribution of the slope values ​​in the voltage hysteresis curve, the slope change characteristics of the voltage hysteresis curve can be quantitatively characterized: when the slope value distribution is more concentrated, there are more data points with the same slope value, indicating that the linear characteristics of the hysteresis curve are significant. At this time, the distribution frequency of the slope values ​​is dense and the corresponding probability value is higher, and the corresponding second complexity coefficient has a larger value.

[0085] Based on this, the second complexity coefficient characterizes the density of the slope distribution of data points on the voltage hysteresis curve. When the slope distribution density of data points is greater, the corresponding second complexity coefficient value is larger, indicating that the change trend on the voltage hysteresis curve is relatively simple, and therefore fewer operators should be used. When the slope distribution density of data points is less, the corresponding second complexity coefficient value is smaller, indicating that the change trend on the voltage hysteresis curve is more complex, and therefore more operators should be applied.

[0086] Step S403: rounding down the ratio of the first complex coefficient to the second complex coefficient to obtain a curve change characteristic value.

[0087] As a specific example, the calculation formula of the curve change characteristic value can be expressed as:

[0088]

[0089] in, Indicates the characteristic value of the curve change, Indicates the maximum value of all the variation coefficients on the voltage hysteresis curve. Indicates the minimum value of all the variation coefficients on the voltage hysteresis curve, represents the number of all data points on the voltage hysteresis curve, Represents the total number of data points on the voltage hysteresis curve that have the same slope as the nth data point, Represents the probability value of the slope of the nth data point on the voltage hysteresis curve, Indicates the floor symbol.

[0090] When the slope of the data points on the voltage hysteresis curve is more concentrated, the corresponding second complexity coefficient is larger, the voltage hysteresis curve's changing trend is more uniform, and the variation between local hysteresis curves tends to be more consistent. In other words, the smaller the value of the first complexity coefficient, the smaller the corresponding curve change characteristic value, and the smaller the number of unilateral play operators should be. This ensures an accurate description of the hysteresis curve's changing characteristics while also taking into account computational efficiency.

[0091] When the slope distribution of the data points on the voltage hysteresis curve is relatively dispersed, the corresponding second complexity coefficient has a smaller value. At this time, the change trend of the voltage hysteresis curve is more complex. At the same time, the distribution range of the degree of difference in the change between the local hysteresis curves is larger. That is, the value of the first complexity coefficient is larger, the corresponding curve change characteristic value is larger, and the value of the number of unilateral play operators should be larger.

[0092] This mechanism achieves adaptive optimization of the number of operators by coupling the slope distribution characteristics with the curve difference coefficient. It automatically increases the operator density in areas with prominent nonlinear characteristics, while maintaining streamlined calculations in linear areas, forming an intelligent adjustment strategy that takes into account both model accuracy and computational efficiency.

[0093] Step S500 : determining the number of operators based on the curve change characteristic value, constructing a PI model of the piezoelectric driver, and performing hysteresis compensation control on the piezoelectric driver.

[0094] Specifically, the sum of the curve change characteristic value and the preset initial operator number is used as the number of unilateral play operators to construct the PI model of the piezoelectric actuator. In this embodiment, the preset initial operator number is 10, and the implementer can set it according to the specific implementation scenario.

[0095] Furthermore, hysteresis compensation control of the piezoelectric actuator is performed based on the improved PI model of this embodiment. This process is well known in the art and will only be briefly described here. Specifically, the PI model is inverted to obtain a PI inverse model. The PI model and the PI inverse model are symmetrical about 45 degrees in a rectangular coordinate system. It should be noted that the specific process of solving the PI inverse model is well known in the art and will not be repeated here.

[0096] The PI inverse model can be used to effectively offset the hysteresis effect of the metal-based piezoelectric ceramic driver. The feedforward compensation voltage of the inverse PI model is used as the input of the driving power supply to eliminate the nonlinear mapping relationship of the piezoelectric ceramic driver in advance, thereby realizing hysteresis compensation control of the piezoelectric ceramic driver. At the same time, the PID algorithm is used to calculate the gap between the actual output voltage and the expected output voltage to achieve displacement stability control of the piezoelectric ceramic driver.

[0097] To sum up, in this embodiment, first of all, the metal-based piezoelectric ceramic driver used adds a metal substrate between the two piezoelectric plates, which can be used as a common electrode, and can also increase the tangential amplitude of the friction plate under the same driving voltage, thereby increasing the movement speed of the mover. At the same time, the width of the driver has no obvious effect on the vibration characteristics, and its application scenarios are broader.

[0098] Then, the hysteresis characteristics of the metal-based piezoelectric ceramic driver were calculated and analyzed. By analyzing the slope values ​​and signs of different curve segments in the hysteresis curve, the curve difference coefficient of the hysteresis curve was calculated, which effectively characterized the change law of the hysteresis curve of the piezoelectric ceramic driver in different stages and provided theoretical support for the subsequent optimization of the play operator.

[0099] Furthermore, the defect of the basic play operator that is symmetrical about the center point and cannot accurately describe the characteristics of the hysteresis curve is improved. A unilateral play operator is selected, and the number of unilateral play operators is calculated based on the curve difference coefficient of the hysteresis curve. The most appropriate number of operators is used to accurately and smoothly describe the changing characteristics of the hysteresis curve. The PI model and PI inverse model are constructed based on the play operator, realizing hysteresis compensation control and stable drive control of the piezoelectric actuator.

[0100] The above-described embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present application, and should all be included in the scope of protection of the present application.

Claims

1. A piezoelectric actuator hysteresis compensation control method for improving tangential amplitude, characterized in that: The method comprises the following steps: Obtain input voltage data and displacement data of the piezoelectric actuator and construct a voltage hysteresis curve of the piezoelectric actuator; According to the change trend of each data point on the voltage hysteresis curve, the voltage hysteresis curve is segmented to obtain each local hysteresis curve; According to the difference of local change trends between each two local hysteresis curves and the position distribution between the data points on each two local hysteresis curves, the change difference coefficients between different local hysteresis curves are obtained; According to the distribution range of the change difference coefficients corresponding to all local hysteresis curves on the voltage hysteresis curve, combined with the quantity distribution and range distribution of the change trends of all data points on the voltage hysteresis curve, the curve change characteristic value is obtained; The number of operators is determined based on the characteristic value of the curve change, a PI model of the piezoelectric driver is constructed, and hysteresis compensation control is performed on the piezoelectric driver.

2. The method for controlling hysteresis compensation of a piezoelectric driver for improving tangential amplitude according to claim 1, characterized in that: The variation difference coefficients between different local hysteresis curves are obtained based on the difference in local variation trends between each two local hysteresis curves and the position distribution between data points on each two local hysteresis curves, specifically including: Obtain the slope of each data point on the voltage hysteresis curve according to all data points within the neighborhood of each data point on the voltage hysteresis curve; Obtain any two local hysteresis curves with the same change trend, match the two local hysteresis curves based on the dynamic time warping algorithm, and obtain the matching relationship between each data point; According to the slope difference and position distribution between any two local hysteresis curves that have a matching relationship, the change difference coefficient between any two local hysteresis curves is obtained.

3. The method for controlling hysteresis compensation of a piezoelectric driver for improving tangential amplitude according to claim 2, wherein: The variation difference coefficient between any two local hysteresis curves is obtained based on the slope difference and position distribution between any two local hysteresis curves that have a matching relationship, specifically including: For two local hysteresis curves with the same slope sign, the absolute value of the difference in the slopes of the data points with matching relationship between the two local hysteresis curves is used as the trend difference factor; the Euclidean distance between the corresponding coordinates of the data points with matching relationship between the two local hysteresis curves is used as the distance difference factor; Based on the balanced distribution of the ratio between the trend difference factor and the distance difference factor, the change difference coefficient between any two local hysteresis curves is determined.

4. The method for controlling hysteresis compensation of a piezoelectric driver for improving tangential amplitude according to claim 2, wherein: Obtaining the slope of each data point on the voltage hysteresis curve based on all data points within the neighborhood of each data point on the voltage hysteresis curve specifically includes: Obtain all data points within a window of a preset length containing each data point for straight line fitting, and obtain the slope of the fitted line as the slope of the corresponding data point.

5. The method for controlling hysteresis compensation of a piezoelectric driver for improving tangential amplitude according to claim 2, wherein: The curve change characteristic value is obtained based on the distribution range of the change difference coefficients corresponding to all local hysteresis curves on the voltage hysteresis curve, combined with the quantity distribution and range distribution of the change trends of all data points on the voltage hysteresis curve, specifically including: The first complexity coefficient is obtained according to the distribution range of all the variation difference coefficients; A second complex coefficient is obtained according to the distribution quantity of the slope of each data point on the voltage hysteresis curve and the distribution probability of the slope; The ratio of the first complex coefficient to the second complex coefficient is rounded down to obtain the curve change characteristic value.

6. The method for controlling hysteresis compensation of a piezoelectric driver for increasing tangential amplitude according to claim 5, characterized in that: Obtaining the first complexity coefficient according to the distribution range of all variation difference coefficients specifically includes: The difference between the maximum and minimum values ​​of all variation coefficients is taken as the first complex coefficient.

7. The method for controlling hysteresis compensation of a piezoelectric driver for improving tangential amplitude according to claim 5, characterized in that: The second complex coefficient is obtained according to the distribution quantity and distribution probability of the slope of each data point on the voltage hysteresis curve, specifically including: The probability value of the slope of each data point on the voltage hysteresis curve is obtained based on the kernel density estimation algorithm; the number of all data points with the same slope as each data point is obtained; and the second complex coefficient is obtained by weighted summing the number of data points using the probability value.

8. The method for controlling hysteresis compensation of a piezoelectric driver for improving tangential amplitude according to claim 1, characterized in that: The step of segmenting the voltage hysteresis curve according to the change trend of each data point on the voltage hysteresis curve to obtain each local hysteresis curve specifically includes: Each extreme point of the voltage hysteresis curve is obtained, and each extreme point is used as a segmentation point to segment the voltage hysteresis curve to obtain each local hysteresis curve.

9. The method for controlling hysteresis compensation of a piezoelectric driver for improving tangential amplitude according to claim 1, characterized in that: The determining of the number of operators based on the curve change characteristic value and constructing the PI model of the piezoelectric actuator specifically includes: The sum of the curve change characteristic value and the preset initial operator number is used as the number of unilateral play operators to construct the PI model of the piezoelectric actuator.

10. The method for controlling hysteresis compensation of a piezoelectric driver for improving tangential amplitude according to claim 1, characterized in that: The step of obtaining input voltage data and displacement data of the piezoelectric driver and constructing a voltage hysteresis curve of the piezoelectric driver specifically includes: The input voltage of the piezoelectric driver is used as the horizontal coordinate and the displacement of the piezoelectric driver is used as the vertical coordinate, and curve fitting is performed to construct the voltage hysteresis curve of the piezoelectric driver.

Images