A control method of a full life cycle stick-slip piezoelectric driver

Abstract

This application provides a control method for a stick-slip piezoelectric actuator throughout its entire lifecycle, comprising: obtaining a contact force reference value using the equation of the driving characteristic surface; acquiring contact force data from a database; calculating a contact force compensation value and converting it into a compensation voltage for the vertical piezoelectric stack; acquiring step displacement and actual contact force data; obtaining the total wear volume using the Archard adhesive wear model based on the step displacement and voltage data of the vertical piezoelectric stack; proportionally allocating the total wear volume to the slider and driving foot using a wear distribution method based on material hardness; storing the obtained slider wear data, driving foot wear data, and actual contact force data in a contact force database; and predicting contact force data for non-cyclic working ranges based on slider wear data and driving foot wear data within the cyclic working range, and storing this prediction in the database. This method can effectively improve the driving stability of the stick-slip piezoelectric actuator throughout its entire working stroke and lifecycle.

Description

Technical Field

[0001] This application belongs to the field of piezoelectric drive and control technology, and in particular relates to a control method for a stick-slip piezoelectric actuator with a full life cycle. Background Technology

[0002] Piezoelectric actuators have significant advantages such as small size, simple structure, and fast response speed, and have been widely researched and applied in fields such as precision positioning, biomedicine, ultra-precision machining, and robotics. However, during long-term continuous operation, these actuators are prone to performance stability degradation and decreased consistency of step displacement. This problem is particularly prominent in stick-slip piezoelectric actuators, which severely limits their further promotion in scenarios with stringent long-term reliability requirements.

[0003] To overcome the aforementioned problems, researchers have successfully improved the stability of the contact force between the drive foot and the slider in stick-slip piezoelectric actuators by controlling the contact force actively or passively, thus ensuring the performance stability of the actuator throughout its entire working stroke. However, due to the stick-slip driving principle, wear continues to occur between the drive foot and the slider. This causes the driving performance of traditional actuators to decline significantly over time, thereby limiting their lifespan.

[0004] Therefore, there is an urgent need for a method to improve the driving stability of stick-slip piezoelectric actuators throughout their entire life cycle. Summary of the Invention

[0005] In view of this, the purpose of this application is to provide a control method for a stick-slip piezoelectric actuator with a complete life cycle, which can solve the problem that the driving performance of traditional actuators decreases over time.

[0006] This application provides a control method for a stick-slip piezoelectric actuator throughout its entire lifecycle, including: Based on the current horizontal load data, the reference value of the contact force data is obtained by using the pre-calibrated driving characteristic surface equation. The contact force data of the current cycle is obtained from the database to obtain the contact force compensation value, and the contact force compensation value is converted into the compensation voltage of the vertical piezoelectric stack. A laser displacement sensor is used to collect the step displacement of the slider, and a pressure sensor is used to collect the actual contact force data between the slider and the drive foot. Based on step displacement and voltage data of vertical voltage stack, the total wear volume is obtained using the Archard adhesive wear model; A wear volume allocation method based on material hardness is used to proportionally distribute the total wear volume to the slider and the drive foot. Then, the obtained slider wear data, drive foot wear data, and actual contact force data are stored in the contact force database. Based on the wear data of the slider and drive foot within the working range in the database, the contact force data in the non-working range is predicted and stored in the database; wherein, the working range and the non-working range are fixed displacement coordinate intervals divided based on the wear state of the slider surface.

[0007] Furthermore, the contact force data for the non-cyclic working range is predicted using the following method: Control the drive foot to complete one single movement within the full stroke of the slider, and record the initial contact force reference data for the entire stroke; Conduct controlled wear simulation: Complete a set number of cyclic loading cycles within a specified working range. After completion, record the contact force data of the last cycle within the working range. Since the contact force degradation during the cyclic phase is caused by both slider wear and drive foot wear, the total contact force degradation value is first calculated based on the difference between the initial contact force and the contact force of the last cycle within the working range. Then, the total contact force degradation value is decomposed using a pre-calibrated wear distribution coefficient to obtain the average contact force degradation value of the slider during the cyclic phase and the average contact force degradation value of the drive foot throughout the entire stroke. Since the contact force degradation outside the working range is caused solely by wear on the drive foot itself, a spatial equidistant offset method is used. The average contact force degradation of the drive foot throughout the entire stroke is subtracted from the initial contact force baseline data for the full stroke to obtain the contact force data for the non-working range. Furthermore, the equations of the driving feature surface are pre-calibrated in the following manner: Data sets of contact force, horizontal load, and step displacement were obtained through orthogonal experiments. A scatter plot was plotted based on this dataset, and a quadratic surface was used to fit all sample points. When the coefficient of determination R² = 0.9647, the equation of the driving characteristic surface was obtained as follows:

[0008] In the formula, For step displacement, This is the resistance / horizontal load (i.e., the voltage data applied to the horizontal piezoelectric stack). This is contact force data.

[0009] Furthermore, the Archard adhesive wear model is as follows:

[0010] In the formula, Q represents the total wear volume, and k is the wear distribution coefficient, which is related to the material and contact state of the contact pair. Where L is the normal load (voltage data of the vertical voltage stack), H is the sliding distance, and H is the material hardness.

[0011] Furthermore, the wear distribution coefficient is pre-calibrated in the following manner: The contact force compensation function of the vertical piezoelectric stack is not enabled, and the contact force between the slider and the drive foot is set to 2.5N through the micro-displacement platform; Before and after the drive cycle, the contact force data between the slider and the drive foot is collected by the pressure sensor within the displacement coordinate range of -3mm to 3mm, so as to obtain the amount of contact force degradation. Based on the displacement coordinate range of the working range being -1mm to 1mm, and considering that only the drive foot experiences wear outside the working range, the changes in contact force caused by slider wear were obtained. Changes in contact force caused by wear of the drive feet as follows:

[0012]

[0013] In the formula, This represents the average contact force degradation within the displacement coordinate range of -1mm to 1mm. The average contact force degradation within the combined non-cyclic loading range (-3mm to -1mm and 1mm to 3mm); The wear distribution coefficients are further obtained as follows:

[0014] Furthermore, the applicable conditions for the wear distribution method based on material hardness are as follows: (1) The drive foot and the slider are made of metal materials with different hardness. The wear process is mainly carried out by the material with lower hardness. (2) There is no obvious material transfer between the contact pairs or the amount of material transfer is negligible, and the contact surface is uniform, i.e., dry and without lubrication; The formula for the wear distribution between the slider and the drive foot is:

[0015]

[0016] In the formula, For the wear volume of the drive foot, The material hardness of the drive foot, The wear volume of the slider. The value is the material hardness of the slider.

[0017] The control method for the stick-slip piezoelectric actuator with a full life cycle provided in this application can effectively improve the linearity of the actuator's displacement output and its ability to maintain driving characteristics under external disturbances. It can effectively improve the driving stability of the stick-slip piezoelectric actuator throughout its entire operating range and life cycle, and provide technical support and practical guidance for the industrial application of stick-slip piezoelectric actuators. Attached Figure Description

[0018] Figure 1 A schematic diagram of a three-dimensional model of the stick-slip piezoelectric actuator provided in an embodiment of this application is shown; Figure 2 A schematic diagram illustrating the driving principle of the stick-slip piezoelectric actuator provided in an embodiment of this application is shown; Figure 3 A flowchart illustrating the control method for a stick-slip piezoelectric actuator throughout its entire lifecycle, as provided in an embodiment of this application, is shown. Figure 4 (a) shows a schematic diagram of the contact force-horizontal load dataset provided in an embodiment of this application; Figure 4 (b) shows a scatter plot of contact force-horizontal load-step displacement data provided in an embodiment of this application; Figure 4 (c) A schematic diagram of the driving feature surface of the piezoelectric actuator provided in an embodiment of this application is shown; Figure 5 This paper illustrates a control principle diagram based on a driving characteristic surface provided in an embodiment of this application. Figure 6 This invention provides a schematic diagram illustrating the principle of compensation for the contact force between the slider and the drive foot according to an embodiment of the present application. Figure 7 This application provides a schematic diagram illustrating the degradation relationship between contact force and displacement over time, and a schematic diagram illustrating the relationship between contact force change and displacement. Figure 8 A schematic diagram of the experimental system provided in an embodiment of this application is shown; Figure 9 A wiring diagram of the control system provided in an embodiment of this application is shown; Figure 10 This illustration shows a schematic diagram of the experimental results for wear distribution and contact force prediction provided in an embodiment of this application; Reference numerals: 1-Micro-displacement platform, 2-Adapter plate A, 3-Base, 4-Slider, 5-Piezoelectric stack, 6-Flexible mechanism, 7-Threaded hole, 8-Pressure sensor, 9-Adapter plate B, 10-Slide rail, 11-Locking knob, 12-Adjusting handle, 13-Prototype, 14-Power supply, 15-Power amplifier, 16-Personal computer, 17-Field programmable gate array, 18-Laser displacement sensor controller, 19-Pulley, 20-Laser sensor. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this technical solution clearer, the following detailed description, in conjunction with specific embodiments, further illustrates this technical solution. It should be understood that these descriptions are merely exemplary and not intended to limit the scope of this technical solution.

[0020] Example 1: First, let's introduce the application scenario of this application: the full life cycle driving of stick-slip piezoelectric actuators.

[0021] The structure and working principle of the stick-slip piezoelectric actuator will be introduced next: Please see as follows Figure 1 The diagram shows a three-dimensional model of a stick-slip piezoelectric actuator.

[0022] A stick-slip piezoelectric actuator is a device that can measure and actively adjust the contact force between the drive foot and the slider (4). It mainly consists of a base (3), a slider (4), a slide rail (10), an adapter plate A (2), an adapter plate B (9), a micro-displacement platform (1), a piezoelectric stack (5) (including vertical and horizontal piezoelectric stacks), a flexible mechanism (6), and a pressure sensor (8).

[0023] A threaded hole (7) is machined on one side of the flexible mechanism (6) to adjust the preload of the piezoelectric stack. The adjustment handle (12) of the micro-displacement stage is used to adjust the contact force between the drive foot and the slider (4) over a wide range. The locking knob (11) on one side of the micro-displacement stage is used to lock the micro-displacement stage, reducing the impact of the internal spring of the micro-displacement stage on the actuator. The pressure sensor (8) is used to measure the contact force between the drive foot and the slider (4) in real time. The slide rail (10) and the slider (4) adopt a cross roller slide rail (10), which can eliminate the influence of vertical load on the contact force between the drive foot and the slider (4).

[0024] Please see as follows Figure 2 The diagram shows the driving principle of a stick-slip piezoelectric actuator.

[0025] The actuator is configured with two piezoelectric stacks integrated into the flexible mechanism (6), wherein the vertical piezoelectric stack is used to actively adjust the contact force between the drive foot and the slider (4). During assembly, a voltage of 50V is first preloaded onto the vertical piezoelectric stack, and then the contact force between the drive foot and the slider (4) is adjusted to a suitable value (approximately 2.5N) via the handle of the micro-displacement platform (1). Based on the working principle of the parasitic inertial piezoelectric actuator and the structural characteristics of the target piezoelectric actuator, a drive signal is applied to the piezoelectric stack to drive the flexible mechanism (6) to move. Figure 2 As shown, each working cycle of the driver consists of three steps: Step 1: T = At that moment, the slider (4) is in the initial position (X0) and remains stationary. At this time, due to the preload effect, there is a certain positive pressure between the drive foot and the slider (4).

[0026] Step 2: <T< As the voltage applied to both ends of the horizontal piezoelectric stack slowly and continuously increases, the piezoelectric stack lengthens and pushes the flexible mechanism (6) to slowly extend. Static friction is generated between the drive foot and the slider (4), causing the slider (4) to produce a positive displacement. .

[0027] Step 3: <T< At that moment, the voltage drops sharply, the horizontal piezoelectric stack rapidly contracts and deforms, and the driving foot and slider (4) become subject to dynamic friction, causing the slider (4) to displace in the negative direction. The net displacement of slider (4) within one working cycle is Repeating the above steps can achieve large-stroke step motion output of slider (4).

[0028] The control method for the stick-slip piezoelectric actuator throughout its entire life cycle proposed in this application will be introduced next: Please see as follows Figure 3 The flowchart illustrates the control method for a stick-slip piezoelectric actuator throughout its entire lifecycle. (See attached flowchart.) Figure 3 As shown, the method includes: S101. Based on the current horizontal load data, obtain the reference value of the contact force data using the pre-calibrated driving characteristic surface equation, retrieve the contact force data of the current cycle from the database to obtain the contact force compensation value, and convert the contact force compensation value into the compensation voltage of the vertical piezoelectric stack.

[0029] In this step, the current horizontal load data is first acquired and substituted into the pre-calibrated driving characteristic surface equation (which characterizes the quadratic nonlinear mapping relationship between contact force, horizontal load, and output step displacement). The optimal target contact force under the current working condition is then analytically calculated as a reference value for the contact force data. Subsequently, the contact force data for the current cycle is retrieved from the database and compared with the reference value to obtain the contact force compensation value. Finally, based on the pre-set force-to-electric conversion coefficient of the vertical piezoelectric stack, the contact force compensation value is converted into a corresponding compensation voltage digital command (this digital command is converted into a low-voltage analog signal by the voltage output module and then transmitted to the power amplifier for proportional boost amplification). The generated compensation voltage is then applied to the two poles of the vertical piezoelectric stack. Driven by the compensation voltage, the vertical piezoelectric stack generates micro-displacement, thereby dynamically adjusting the clamping state between the slider and the contact foot, achieving precise compensation for contact force degradation caused by wear.

[0030] In practical implementation, the equations of the driving feature surface are pre-calibrated in the following manner: Data sets of contact force, horizontal load, and step displacement were obtained through orthogonal experiments. A scatter plot was plotted based on this dataset, and a quadratic surface was used to fit all sample points. When the coefficient of determination R² = 0.9647, the equation of the driving characteristic surface was obtained as follows:

[0031] In the formula, For step displacement, This is the resistance / horizontal load (i.e., the voltage data applied to the horizontal piezoelectric stack). This is contact force data.

[0032] Here, in order to accurately fit the driving characteristic surface and calibrate the driving characteristic surface equation, this embodiment designs an orthogonal experiment to collect various parameters. The initial working conditions are set as follows: driving frequency 1Hz, driving voltage 100V, sawtooth wave symmetry 100%; the contact force is adjusted by the micro-displacement platform (1) at intervals of 0.5N, so that the contact force range covers 1N~4N; a horizontal load of 0~600g is applied at intervals of 100g, and the output displacement of the slider (4) is measured by a laser displacement sensor. For each set of experimental parameters, displacement data of 50 driving cycles are collected, and the average value of the step displacement is calculated as the effective experimental data, as shown below. Figure 4 (a) shows a schematic diagram of the contact force-horizontal load dataset.

[0033] Based on the above contact force-horizontal load-step displacement experimental dataset, the following plot is drawn: Figure 4 (b) is a scatter plot of contact force-horizontal load-step displacement data; a quadratic surface is used to fit all sample points, and the coefficient of determination R is... 2 When the coefficient of performance is 0.9647, the fit is good, and it can accurately characterize the variation law of the driving characteristic surface. See here for reference... Figure 4 (c) shows a schematic diagram of the driving feature surface of the piezoelectric actuator.

[0034] For further details, please refer to... Figure 5 The control principle diagram based on the driving characteristic surface shown is as follows: Figure 6 The diagram shows the principle of compensating for the contact force between the slider and the drive foot. Figure 5In the three-dimensional coordinate system, the X-axis represents the operating resistance, the Y-axis represents the contact force between the driving mechanisms, and the Z-axis represents the step displacement of the actuator. The constructed surface is the three-dimensional driving characteristic surface of the piezoelectric actuator. The surface height represents the actual output step capability of the actuator, and the highest point of the surface is the optimal operating point under the current operating condition. Different symbols are used to distinguish the entire control process in the figure: red dots represent the initial optimal operating point, blue dots represent the operating point under disturbance, black dots represent the optimal operating point after compensation and adjustment, blue solid arrows represent the disturbance trajectory, blue dashed arrows represent the hypothetical disturbance trajectory, brown solid arrows represent the feedback control path, gray solid arrows represent the feedforward control path, and hollow dots represent the hypothetical operating point after decoupling. Its core control principle is based on orthogonal decoupling of disturbances and dynamic optimization in a dual-loop space using the three-dimensional driving characteristic surface. The complete control process is as follows: During the disturbance occurrence phase, the driver faces a complex coupled disturbance during actual operation: it includes dynamic disturbances of resistance caused by sudden changes in operating condition resistance (X-axis direction) and contact force attenuation and degradation caused by long-term wear of the drive pair (Y-axis direction). Under the combined effect of the two types of coupled disturbances, the driver's operating point shifts along the blue disturbance trajectory from the initial optimal operating point (red dot, peak of the surface) to the non-ideal operating point after disturbance (blue dot, low slope of the surface), and the output step displacement is significantly attenuated. At this time, it is difficult for a single control loop to simultaneously take into account both disturbance response speed and control accuracy.

[0035] Disturbance decoupling and branch-loop control stage. This application uses contact force compensation to orthogonally decompose the coupled disturbance trajectory vector, transforming the complex multi-physics coupling problem into a single-variable adjustment process in two independent coordinate axes: resistance and contact force. This results in two independent sets of hypothetical disturbance trajectories and hypothetical operating points, which are then targeted by dual loops of feedforward control and feedback control: Feedback control (brown arrow path): Based on real-time collected contact force and step displacement data, closed-loop adjustment is performed to address real-time dynamic disturbances in the X-axis direction caused by sudden changes in resistance, eliminating output offset caused by resistance fluctuations; Feedforward control (gray arrow path): Based on a pre-calibrated and stored historical wear degradation database, the amount of contact force attenuation caused by drive pair wear is predicted, and the drive voltage compensation is output in advance to overcome contact force degradation caused by long-term wear in the Y-axis direction, achieving contact force pre-compensation.

[0036] In the collaborative optimization and compensation closed-loop stage, the predictive pre-compensation of feedforward control and the real-time fine-tuning of feedback control are vector superimposed in the three-dimensional driving feature surface space. Together, they drive the system operating point from the disturbed non-ideal state (blue dot) to return along the control path to the optimal operating point (black dot) of the surface peak that can output the maximum step displacement under the current load conditions. This ultimately achieves efficient decoupling and high-precision active compensation control of contact force under actual working conditions.

[0037] S102. Use a laser displacement sensor to collect the step displacement of the slider and use a pressure sensor to collect the actual contact force data between the slider and the drive foot.

[0038] S103. Based on the step displacement and voltage data of the vertical voltage stack, the total wear volume is obtained using the Arcard adhesive wear model.

[0039] The Archard adhesive wear model is as follows:

[0040] In the formula, Q represents the total wear volume, and k is the wear distribution coefficient, which is related to the material and contact state of the contact pair. Where L is the normal load (voltage data of the vertical voltage stack), H is the sliding distance, and H is the material hardness.

[0041] In practical implementation, the wear distribution coefficient is pre-calibrated in the following way: The contact force compensation function of the vertical piezoelectric stack is not enabled, and the contact force between the slider and the drive foot is set to 2.5N through the micro-displacement platform; Before and after the drive cycle, the contact force data between the slider and the drive foot is collected by the pressure sensor within the displacement coordinate range of -3mm to 3mm, so as to obtain the amount of contact force degradation. Based on the displacement coordinate range of the working range being -1mm to 1mm, and considering that only the drive foot experiences wear outside the working range, the changes in contact force caused by slider wear were obtained. Changes in contact force caused by wear of the drive feet as follows:

[0042]

[0043] In the formula, This represents the average contact force degradation within the displacement coordinate range of -1mm to 1mm. The average contact force degradation within the combined non-cyclic loading range (-3mm to -1mm and 1mm to 3mm); The wear distribution coefficients are further obtained as follows:

[0044] The working range and non-working range are fixed displacement coordinate intervals divided based on the wear state of the slider surface.

[0045] Here, the working range is a pre-defined displacement interval where the drive foot and the slider undergo long-term cyclic friction, causing cumulative wear on the slider. The non-working range is a brand-new, unworn displacement interval where the slider does not participate in cyclic friction, does not experience cumulative wear, and maintains its initial shape. The working and non-working ranges are defined by fixed wear areas, and their coordinate intervals do not change as the experimental procedure progresses.

[0046] Here, due to inherent errors in the theoretical model, this embodiment calibrates the wear distribution coefficient experimentally to improve the accuracy of wear distribution. The contact force compensation function of the vertical piezoelectric stack is not enabled in the experiment. The contact force between the drive foot and the slider (4) is set to approximately 2.5N using a micro-displacement platform (1). The total wear is indirectly characterized by the contact force data. The specific loading process is detailed in Table 1. The results are as follows: Figure 7 The diagrams illustrating the degradation relationship between contact force and displacement over time and the relationship between contact force change and displacement are used to explain the process. Data acquisition and processing were carried out in three stages: before cyclic loading, contact force data was acquired at a frequency of 10Hz within the displacement coordinate range of -3mm to 3mm (Data 1); during the cyclic loading stage, the actuator operated 1968 times at a frequency of 50Hz within the displacement coordinate range of -1mm to 1mm; after cyclic loading, contact force data was acquired again at a frequency of 10Hz within the displacement coordinate range of -3mm to 3mm (Data 2). After denoising and interpolation processing of Data 1 and Data 2, fitting was performed. The fitting results and the original data are shown below. Figure 7 As shown in the upper part, the contact force decreases during the loading process.

[0047] Combination Figure 7 The different markings (dark red for working range -1mm~1mm, light red for non-working range -3mm~-1mm and 1mm~3mm) show the following analysis of the causes of wear and contact force changes: Within the working range, both the drive foot and the slider (4) experience wear due to stick-slip drive, and the change in contact force is caused by the combined wear of the two; outside the working range, only the drive foot experiences wear, and the change in contact force is entirely attributed to the wear of the drive foot.

[0048] Therefore, it can be seen that if a stick-slip piezoelectric actuator operates in a long-term cycle within a specific working range, its slider (4) will only experience wear within that working range, while the surface of the slider (4) outside the working range will not experience wear. However, while the slider (4) is only experiencing wear within the working range, the drive foot is also experiencing wear. If it moves to the non-working range, the contact force data within that range will change. In order to accurately quantify the adhesive wear process between the drive foot and the slider (4), this embodiment introduces the Archard adhesive wear model to improve the quantitative calculation and characterization capability of the wear behavior of the stick-slip piezoelectric actuator.

[0049]

[0050] S104. The total wear volume is proportionally allocated to the slider and drive foot using a wear distribution method based on material hardness. Then, the obtained slider wear data, drive foot wear data, and actual contact force data are stored in the contact force database.

[0051] The applicable conditions for the wear distribution method based on material hardness are as follows: (1) The drive foot and the slider are made of metal materials with different hardness. The wear process is mainly carried out by the material with lower hardness. (2) There is no obvious material transfer between the contact pairs or the amount of material transfer is negligible, and the contact surface is uniform, i.e., dry and without lubrication; The formula for the wear distribution between the slider and the drive foot is:

[0052]

[0053] In the formula, For the wear volume of the drive foot, The material hardness of the drive foot, The wear volume of the slider. The value is the material hardness of the slider.

[0054] S105. Based on the wear data of the slider and drive foot within the working range in the database, predict the contact force data outside the working range and store it in the database.

[0055] In this step, when the drive foot is limited to cyclic movement within the working range, controllable wear of the slider and drive foot can be generated; when the range of motion of the drive foot is extended to the non-working range, the change in contact force caused solely by wear of the drive foot can be tested and predicted.

[0056] When the drive foot and slider operate cyclically within a specified working range for an extended period, the slider will experience cumulative wear in that area, while the drive foot will also experience uniform wear throughout its entire stroke. When the drive foot leaves the worn working area of ​​the slider and enters a completely new, unworn area of ​​the slider, the contact force will change only due to the wear of the drive foot itself. Based on this characteristic, the contact force of the actuator throughout its entire stroke can be accurately predicted through artificially controlled cyclic wear manufacturing.

[0057] In practical implementation, the contact force data outside the working range is predicted using the following methods: Control the drive foot to complete one single movement within the full stroke of the slider, and record the initial contact force reference data for the entire stroke; Conduct controlled wear simulation: Complete a set number of cyclic loading cycles within a specified working range. After completion, record the contact force data of the last cycle within the working range. Since the contact force degradation during the cyclic phase is caused by both slider wear and drive foot wear, the total contact force degradation value is first calculated based on the difference between the initial contact force and the contact force of the last cycle within the working range. Then, the total contact force degradation value is decomposed using a pre-calibrated wear distribution coefficient to obtain the average contact force degradation value of the slider during the cyclic phase and the average contact force degradation value of the drive foot throughout the entire stroke. Since the degradation of contact force outside the working range is caused only by the wear of the drive foot itself, a spatial equidistant offset method is adopted. The average value of the degradation of contact force of the drive foot within the full stroke is subtracted from the initial contact force reference data of the full stroke to obtain the contact force data of the non-working range.

[0058] Example 2: To verify the effectiveness of the full life-cycle control method of the stick-slip piezoelectric actuator proposed in this application, a prototype stick-slip piezoelectric actuator (13) was fabricated in this embodiment, and a testing mechanism was established to obtain the following results: Figure 8 The experimental system shown is a schematic diagram. The flexible mechanism (6) is fabricated by wire electrical discharge machining (WEDM), the piezoelectric stack is Coremorrow Pst150, the slide rail (10) and slider (4) are Zhaosong VRT3080A, the pressure sensor (8) is DAYSENSOP DYLY-107, and the pulley (19) and other structural components are all custom-made parts.

[0059] For further details, please refer to... Figure 9 The diagram shows the wiring schematic of the control system. This diagram illustrates the overall structure of the control system, specifically its components as follows: 1. Hardware components of the control mechanism Controller: cRIO-9035 CompactRIO controller of Kintex-7 70T field programmable gate array (17), with a clock frequency of 40MHz, to meet the data acquisition, calculation and output requirements under high frequency drive; Signal modules: voltage input module (cRIO9205, NI) and voltage output module (cRIO9263, NI), which respectively realize signal acquisition and output; Sensing devices: laser sensor (20) (LK-H020, KEYENCE), laser displacement sensor controller (18) (LK-HD500, KEYENCE), used to collect the step displacement of slider (4); Power components: power amplifier (15) (PD200X4, PiezoDrive), power supply (14) (DP832, RIGOL), to realize power amplification of drive signals and power supply of mechanism; Host computer: Personal computer (16) used to run control programs and interact with data.

[0060] 2. Software Components of the Experimental Mechanism Host computer software: Runs LabVIEW programs and is equipped with a user interface, which can realize functions such as drive mode setting, parameter configuration, and real-time data display; Communication method: The personal computer (16) establishes TCP / IP communication with the cRIO-9035 controller through the network cable to complete the issuance of control commands and the uploading of experimental data; Controller processing: The cRIO-9035 has built-in control logic, which works with the signal module to complete the real-time processing of contact force / displacement signal acquisition, compensation calculation, and drive signal output.

[0061] 3. Experimental study on the self-updating effect of the database: Through the established experimental mechanism, long-term wear experiments of the actuator and comparative experiments on contact force data prediction were conducted, yielding the following results: Figure 10 The diagram shows the experimental results of wear distribution and contact force prediction.

[0062] Based on the initial contact force data (green scatter points represent the global initial contact force data Data3, and dark green curves represent its fitted curves), cyclic loading, contact force prediction, and global sampling verification were carried out in three groups. The specific process is as follows: The first group completed 1000 cycles of loading at a frequency of 10Hz within the range of -3mm to -2mm. Based on the contact force data after loading within this range, the contact force in the non-working range was predicted (the light blue curve is the predicted data). The contact force data Data4 (light blue scatter) of the entire working stroke was sampled, and the dark green curve is its fitted curve. The second group completed 3450 cycles of loading at a frequency of 50Hz within the range of 1mm to 2mm. Based on the data in this range, the contact force outside the working range was predicted (the light purple curve is the predicted data). The contact force data Data5 (light purple scatter points) of the entire working stroke was sampled and obtained. The dark purple curve is its fitting curve. The third group completed 2930 cyclic loading cycles at a frequency of 25Hz within the range of -2mm to 0mm. Based on the data within this range, the global contact force was predicted (the light red curve represents the predicted data). The global contact force data Data6 (light red scatter points) was obtained by sampling, and the dark red curve is its fitted curve. The feasibility and accuracy of the self-updating method were verified through multiple comparisons.

[0063] Figure 10Dark red and light red are used to distinguish the different causes of contact force changes. The light red area corresponds to the non-cyclic load range, and the contact force changes in this range are caused entirely by the wear of the drive foot. The dark red area corresponds to the cyclic load range, and the contact force changes in this range are caused by the combined wear of the drive foot and the surface of the slider (4).

[0064] To evaluate the effectiveness of contact force prediction, this embodiment uses mean absolute error (MAE), mean square error (MSE), root mean square error (RMSE), and mean absolute percentage error (MAPE) as evaluation indicators. Contact force prediction tests show that the mean percentage error is less than 1%.

[0065] Experimental results show that the wear allocation mechanism based on the wear model and contact force database self-updating proposed in this application can effectively predict contact force data in the non-cyclic load range, significantly improve the wear allocation accuracy and contact force feedforward control performance, and verify the effectiveness and practicality of the method.

[0066] The above content is only a preferred embodiment of the present invention. For those skilled in the art, many changes can be made in the specific implementation and application scope based on the ideas of the present invention. As long as these changes do not depart from the concept of the present invention, they all fall within the protection scope of the present invention.

Claims

1. A control method for a stick-slip piezoelectric actuator with a complete life cycle, characterized in that, The method includes: Based on the current horizontal load data, the reference value of the contact force data is obtained by using the pre-calibrated driving characteristic surface equation. The contact force data of the current cycle is obtained from the database to obtain the contact force compensation value, and the contact force compensation value is converted into the compensation voltage of the vertical piezoelectric stack. A laser displacement sensor is used to collect the step displacement of the slider, and a pressure sensor is used to collect the actual contact force data between the slider and the drive foot. Based on step displacement and voltage data of vertical voltage stack, the total wear volume is obtained using the Archard adhesive wear model; A wear volume allocation method based on material hardness is used to proportionally distribute the total wear volume to the slider and the drive foot. Then, the obtained slider wear data, drive foot wear data, and actual contact force data are stored in the contact force database. Based on the wear data of the slider and drive foot within the working range in the database, the contact force data in the non-working range is predicted and stored in the database; wherein, the working range and the non-working range are fixed displacement coordinate intervals divided based on the wear state of the slider surface.

2. The method as described in claim 1, characterized in that, Contact force data outside the working range is predicted using the following method: Control the drive foot to complete one single movement within the full stroke of the slider, and record the initial contact force reference data for the entire stroke; Conduct controlled wear simulation: Complete a set number of cyclic loading cycles within a specified working range. After completion, record the contact force data of the last cycle within the working range. Since the contact force degradation during the cyclic phase is caused by both slider wear and drive foot wear, the total contact force degradation value is first calculated based on the difference between the initial contact force and the contact force of the last cycle within the working range. Then, the total contact force degradation value is decomposed using a pre-calibrated wear distribution coefficient to obtain the average contact force degradation value of the slider during the cyclic phase and the average contact force degradation value of the drive foot throughout the entire stroke. Since the degradation of contact force outside the working range is caused only by the wear of the drive foot itself, a spatial equidistant offset method is adopted. The average value of the degradation of contact force of the drive foot within the full stroke is subtracted from the initial contact force reference data of the full stroke to obtain the contact force data of the non-working range.

3. The method as described in claim 1, characterized in that, The equations of the driving feature surface are pre-calibrated using the following method: Data sets of contact force, horizontal load, and step displacement were obtained through orthogonal experiments. A scatter plot was plotted based on this dataset, and a quadratic surface was used to fit all sample points. When the coefficient of determination R² = 0.9647, the equation of the driving characteristic surface was obtained as follows: ； In the formula, For step displacement, This is the resistance / horizontal load (i.e., the voltage data applied to the horizontal piezoelectric stack). This is contact force data.

4. The method as described in claim 1, characterized in that, The Archard adhesive wear model is as follows: ； In the formula, Q represents the total wear volume, and k is the wear distribution coefficient, which is related to the material and contact state of the contact pair. Where L is the normal load (voltage data of the vertical voltage stack), H is the sliding distance, and H is the material hardness.

5. The method as described in claim 2, characterized in that, The wear distribution coefficient can be pre-calibrated using the following method: The contact force compensation function of the vertical piezoelectric stack is not enabled, and the contact force between the slider and the drive foot is set to 2.5N through the micro-displacement platform; Before and after the drive cycle, the contact force data between the slider and the drive foot is collected by the pressure sensor within the displacement coordinate range of -3mm to 3mm, so as to obtain the amount of contact force degradation. Based on the displacement coordinate range of the working range being -1mm to 1mm, and considering that only the drive foot experiences wear outside the working range, the changes in contact force caused by slider wear were obtained. Changes in contact force caused by wear of the drive feet as follows: ； ； In the formula, This represents the average contact force degradation within the displacement coordinate range of -1mm to 1mm. The average contact force degradation within the combined non-cyclic loading range (-3mm to -1mm and 1mm to 3mm); The wear distribution coefficients are further obtained as follows: 。 6. The method as described in claim 1, characterized in that, The applicable conditions for the wear distribution method based on material hardness are as follows: (1) The drive foot and the slider are made of metal materials with different hardness. The wear process is mainly carried out by the material with lower hardness. (2) There is no obvious material transfer between the contact pairs or the amount of material transfer is negligible, and the contact surface is uniform, i.e., dry and without lubrication; The formula for the wear distribution between the slider and the drive foot is: ； ； In the formula, For the wear volume of the drive foot, The material hardness of the drive foot, The wear volume of the slider. The value is the material hardness of the slider.

Images