An Adaptive Control Method for the Output Force of Piezoelectric Ceramics under a Temperature-Varying Environment

Abstract

The present invention belongs to the field of measuring the output characteristics of piezoelectric ceramics, and discloses a method for adaptively controlling the output force of piezoelectric ceramics in a temperature-changing environment. First, a test platform for measuring the output characteristics of a piezoelectric actuator in a temperature-changing environment is built, and experimental data of the input voltage and output force of the piezoelectric ceramic actuator are collected. Based on the test data, a dynamic hysteresis model of the output force-voltage of the piezoelectric actuator is constructed. On this basis, the hysteresis inverse model is solved to compensate for the output force lost due to the hysteresis effect. Finally, a variable-parameter algorithm for the output force-voltage model of piezoelectric ceramics in a temperature-changing environment is constructed, and appropriate model parameters are selected through temperature feedback to form a variable-parameter regulation method for the output force-voltage model in a temperature-changing environment. This method uses a variable-parameter algorithm to improve the practicability of the output force-voltage model, avoids the problems of reduced output performance of piezoelectric ceramics and unstable system control caused by temperature changes in an actual system, and takes into account factors such as dynamic hysteresis of the piezoelectric ceramic actuator.

Description

Technical Field

[0001] The present invention belongs to the field of measuring the output characteristics of piezoelectric ceramics, and relates to a method for adaptively controlling the output force of piezoelectric ceramics. Background Art

[0002] Aerodynamic model tests of aircraft can obtain their key aerodynamic characteristics, effectively reducing the development technical risks and costs. Due to the small interference with the flow field, the aircraft model is often fixed in the wind tunnel in a cantilever support manner to measure the aerodynamic force of the model. However, during the wind tunnel test, the model cantilever support has low stiffness and weak damping, and is prone to irregular vibrations under the disturbance of the unsteady flow field, resulting in inaccurate measurement data of the aerodynamic force at a fixed angle of attack. Therefore, it is necessary to actively suppress the vibration of the wind tunnel model to ensure the accuracy and smooth progress of the wind tunnel test data.

[0003] With the rapid development of piezoelectric materials, due to their characteristics such as fast response speed and compact structure, they can efficiently convert electrical energy into mechanical energy, and can achieve better vibration suppression effects as actuators for aircraft wind tunnel models. The vibration active control method based on piezoelectric ceramic actuators has been widely used. However, the output characteristics of piezoelectric ceramics are different at different temperatures, and there are obvious differences in the control effects. The output modeling method with fixed parameters cannot fully exert the vibration suppression ability of the actuator, and the output performance of the vibration control system of the wind tunnel model is limited. Therefore, it is necessary to consider the influence of variable temperature conditions on the vibration suppression output characteristics, and conduct modeling research on the output characteristics of the vibration suppression force, so as to achieve precise control of the vibration suppression force and efficient and rapid suppression of the vibration of the aircraft model.

[0004] The patent "Cooperative Control Method and System for Front and Rear Shock Absorbers of Longitudinal Vibration of High Aspect Ratio Model" by Liu Yu et al., with the patent number CN202210922741.X, introduces a cooperative control method for front and rear shock absorbers of longitudinal vibration of high aspect ratio models. By using the accelerometer signal of the centroid of the vibration suppression system as the feedback signal, and according to the principle of moment balance, the front shock absorber located between the balance and the strut and the rear shock absorber located on the middle bracket of the strut are respectively driven, which can effectively determine the output of the shock absorber according to the vibrations at different positions of the "model - balance - strut system", and achieve the cooperative control of the front and rear shock absorbers for the longitudinal vibration of the high aspect ratio model in the wind tunnel test. However, this method does not consider the relationship between the input and output changes of piezoelectric ceramics at different temperatures, and is not applicable to the active vibration suppression system of aircraft wind tunnel models that require temperature change and consideration of force - voltage dynamic hysteresis.

[0005] Based on the problems existing in the above technologies, it is necessary to propose a method for adaptively controlling the output force of piezoelectric ceramics in a variable temperature environment. Summary of the Invention

[0006] In order to overcome the deficiencies of the prior art, the present invention proposes a method for adaptively controlling the output force of piezoelectric ceramics in a temperature-varying environment. Considering the influence of the temperature-varying working conditions on the output characteristics of piezoelectric ceramic actuators, precise control of the output force is achieved. The method first builds a test platform for the output characteristics of piezoelectric actuators in a temperature-varying environment, collects experimental data on the input voltage and output force of piezoelectric ceramic drivers, then constructs a dynamic hysteresis model of the output force-voltage of the piezoelectric actuator through the test data, solves the hysteresis inverse model on this basis to compensate for the output force lost due to the hysteresis effect, and finally constructs a variable-parameter algorithm for the output force-voltage model of piezoelectric ceramics in a temperature-varying environment. By temperature feedback, appropriate model parameters are selected to form a variable-parameter regulation method for the output force-voltage model in a temperature-varying environment, improving the driving accuracy and response stability of the piezoelectric actuator. The application of the variable-parameter algorithm improves the practicability of the output force-voltage model, avoids the problems of reduced output performance of piezoelectric ceramics and unstable system control caused by temperature changes in the actual system, and at the same time considers factors such as the dynamic hysteresis of piezoelectric ceramic actuators. Compared with traditional single-parameter modeling methods, it is more general and effective and can be applied to various systems containing piezoelectric ceramic actuators.

[0007] The technical solution of the present invention:

[0008] A method for adaptively controlling the output force of piezoelectric ceramics in a temperature-varying environment measures the output force of piezoelectric ceramics at different temperatures and input voltages through a test platform for the output characteristics of piezoelectric ceramic actuators, constructs an output force-voltage dynamic hysteresis model and a variable-parameter algorithm, and forms a variable-parameter regulation method for the model in a temperature-varying environment through temperature feedback; the method improves the practicability of the piezoelectric ceramic output force-voltage model by applying the variable-parameter algorithm, and avoids the problems of reduced output performance of piezoelectric ceramics and unstable system control caused by temperature changes in the actual system;

[0009] The specific steps are as follows:

[0010] The first step: Build a test platform for the output characteristics of piezoelectric ceramic actuators in a temperature-varying environment

[0011] Assemble a test platform for the output characteristics of piezoelectric ceramic force and position. Different temperature environments are realized through a high and low temperature test chamber; install the piezoelectric ceramic actuator on the piezoelectric ceramic output characteristic test device, and apply a pre-tightening force to the piezoelectric ceramic actuator; connect the power amplifier to the piezoelectric ceramic actuator; connect the piezoelectric ceramic actuator, thermocouple, and force sensor to the data acquisition module; collect the input voltage and output force data of the piezoelectric ceramic actuator at different temperatures;

[0012] The second step: Construct a dynamic hysteresis model of the output force-voltage of the piezoelectric ceramic actuator

[0013] Based on the improvement of the classical hysteresis model, the Prandtl-Ishlinskii (PI) model is selected as the basic framework. This model can better describe the hysteresis characteristics and has certain flexibility and scalability. Aiming at the limitations of the PI model in practical applications, a unilateral aubacklash operator with asymmetric characteristics is proposed. By improving the threshold distribution characteristics and adding an asymmetric correction coefficient, the strict symmetry of the backlash operator output is eliminated, and the corresponding API model is obtained;

[0014] Force-electricity hysteresis operator:

[0015] H a , r , b , c [ u ] ( t ) = max { au + r ,min { bu + c , H ( t − 1 ) } } (1)

[0016] Output force-voltage dynamic hysteresis API model:

[0017] F API [ u ] ( t ) = ∑ i = 1 n w i H a i , r i , b i , c i [ u ] ( t ) = w T H [ u ] ( t ) (2)

[0018] Modify the threshold parameter in the API model and associate the original hysteresis operator threshold parameter with the frequency:

[0019] (3)

[0020] Make the calculation output of the hysteresis model change with the change of the input voltage frequency, and obtain the output force-voltage dynamic hysteresis model of the piezoelectric ceramic actuator; where, H a , r , b , c [ u ] ( t ) is the output of the force-electricity hysteresis operator, is the output of the hysteresis operator at the previous moment, is the input voltage, 、 are the asymmetric correction coefficients of the hysteresis operator, is the threshold of the force-electricity hysteresis operator, is a numerical constant, F API [ u ] ( t ) is the output force of the piezoelectric ceramic under the force-electricity hysteresis, is the number of force-electricity hysteresis operators, is the weight of the force-electricity hysteresis operator, is the weight matrix of the force-electricity hysteresis operator, H [ u ] ( t ) is the matrix of the force-electricity hysteresis operator, is a numerical constant, is the voltage derivative, is the initial value of the numerical constant;

[0021] The third step: Solve the compensation output force of the hysteresis inverse model of the piezoelectric ceramic actuator

[0022] The numerical method is used to solve the hysteresis inverse model. First, the output force data at different input voltages and input frequencies are generated by using the built test platform for the force-position output characteristics of piezoelectric ceramics to form a data set. A neural network is constructed, with the output force and frequency as the input layer neurons and the input voltage as the output layer neuron. The neural network is trained with a large amount of data to enable it to learn the mapping law from the output force and frequency to the input voltage, thereby obtaining the inverse model. When a desired output force is given, it is input into the inverse model to obtain the corresponding compensation voltage, and then this compensation voltage is applied to the piezoelectric ceramic actuator as the actual input voltage, so as to compensate for the output force lost due to hysteresis and make the output force closer to the desired output force;

[0023] Step 4: Construct the variable parameter algorithm for the hysteresis model under the variable temperature environment

[0024] Based on the above API dynamic hysteresis model, a temperature-related term is introduced. Let the environmental temperature be T, and the temperature change function h(T) with respect to time is constructed. The asymmetric correction coefficient of the original hysteresis operator in the API model is associated with the temperature change function h(T) with respect to time to reflect the effect law of temperature on the model parameters and make the model characteristics change with temperature. Then, the appropriate model parameters are selected through temperature feedback to improve the universality of the model;

[0025] (4)

[0026] Among them, 、 are the temperature correction coefficients, is the temperature change amount, 、 are the offset amounts of the temperature correction coefficients;

[0027] Embed the optimized API dynamic hysteresis model into the control system of the piezoelectric ceramic actuator, and automatically match the appropriate parameters according to the real-time temperature and the expected output force to solve the required input voltage, achieving accurate and intelligent control.

[0028] The beneficial effects of the present invention:

[0029] The proposed adaptive control method for the output force of piezoelectric ceramics under the temperature-changing environment in this method, and the constructed API dynamic hysteresis model can accurately solve the input voltage of the piezoelectric ceramic actuator in different temperature environments, solving the problems that the dynamic hysteresis error of the piezoelectric ceramic actuator is too large and the output performance of the piezoelectric ceramic actuator is limited due to large temperature changes in the actual system, making the control result more accurate and with higher precision; The model establishment and training are convenient and fast, with strong adaptability, and can be applied to all systems containing piezoelectric ceramic actuators. Description of the Drawings

[0030] Figure 1 ​​​Schematic diagram of the connection of the output characteristic test platform for the variable-temperature environment piezoelectric actuator of the present invention;

[0031] Figure 2 Flowchart of the output force adaptive control method for the piezoelectric ceramic actuator of the present invention;

[0032] Figure 3 Output force data curve of the piezoelectric ceramic actuator at different actuation frequencies when the ambient temperature is 20°C. Among them, the X-axis is the input voltage and the Y-axis is the output force.

[0033] Figure 4 Output force-voltage dynamic hysteresis model prediction result curve of the piezoelectric ceramic actuator at different actuation frequencies. Among them, the X-axis is the input voltage and the Y-axis is the output force.

[0034] Figure 5 Output force data curve of the piezoelectric ceramic actuator at different temperatures in the embodiment. Among them, the X-axis is the input voltage and the Y-axis is the output force.

[0035] Figure 6 Relationship curve between the partial asymmetry correction coefficient of the dynamic hysteresis model and temperature. Among them, the X-axis is the ambient temperature and the Y-axis is the value of the symmetry correction coefficient.

[0036] Figure 7 Output force-voltage dynamic hysteresis model prediction result curve of the piezoelectric ceramic actuator at different temperatures. Among them, the X-axis is the input voltage and the Y-axis is the output force.

[0037] In the figure: 1 - power amplifier, 2 - high and low temperature test chamber, 3 - piezoelectric ceramic output characteristic test device, 4 - piezoelectric ceramic actuator, 5 - thermocouple, 6 - force sensor, 7 - data acquisition module, 8 - real-time controller, 9 - upper computer. Specific implementation mode

[0038] The implementation process of the present invention will be described in detail below in combination with the technical solution and the drawings.

[0039] Figure 1 Schematic diagram of the output force data acquisition of the piezoelectric ceramic actuator. The used piezoelectric ceramic output characteristic test device 3 is internally provided with a thermocouple 5 and a force sensor 6, which can obtain the ambient temperature and the output force of the piezoelectric ceramic actuator 4 in real time. The overall test system can ensure force-electricity synchronous measurement and acquisition, and has the characteristics of compact structure and easy use.

[0040] Figure 2It is a flow chart of the adaptive control method for the output force of piezoelectric ceramics in a temperature-changing environment. The entire prediction and modeling method is divided into the following four parts: assembling a test system for the output characteristics of piezoelectric ceramics, constructing a dynamic hysteresis model of the output force-voltage of a piezoelectric ceramic actuator, solving the hysteresis inverse model of the piezoelectric ceramic actuator to compensate for the output force, and constructing a variable-parameter algorithm for the hysteresis model in a variable-temperature environment and conducting experimental verification.

[0041] The first step: Build a test platform for the output characteristics of a piezoelectric actuator in a temperature-changing environment

[0042] As Figure 1 shown, assemble a test platform for the output characteristics of the force and position of piezoelectric ceramics. Different temperature environments are realized through a high and low temperature test chamber 2; the piezoelectric ceramic output characteristic test device 3 is an integrally processed configuration, and the bottom is connected to a force sensor 6 through a bolt. A through groove is opened in the center of the piezoelectric ceramic output characteristic test device 3 for installing a piezoelectric ceramic actuator 4. An initial pre-tightening force is applied to the piezoelectric ceramic actuator 4 through the bolt at the top of the piezoelectric ceramic output characteristic test device 3; a power amplifier 1 is connected to the piezoelectric ceramic actuator 4; the piezoelectric ceramic actuator 4, a thermocouple 5, and a force sensor 6 are connected to a data acquisition module; collect the input voltage and output force data of the piezoelectric ceramic actuator 4 at different temperatures;

[0043] Conduct a data acquisition experiment. Control the test system through LabVIEW software to output a sinusoidal waveform voltage input signal with a variable frequency; the upper computer 9 transmits the control signal to the real-time controller 8, and then the real-time controller 8 transmits it to the power amplifier 1. The power amplifier 1 converts the control signal into a voltage and inputs it to the piezoelectric ceramic actuator 4. Place the piezoelectric ceramic output characteristic test device 3 in the high and low temperature test chamber 2 and conduct multiple groups of experiments at different temperatures. The data acquisition module 7 reads and stores the output signal of the pressure sensor in real time, thereby completing the acquisition of output force data;

[0044] In this embodiment, the test system is controlled by LabVIEW software to input a sinusoidal waveform voltage with a gradually increasing amplitude. The formula is as follows:

[0045] (5)

[0046] In the formula, the parameters are taken as 、 .

[0047] The second step: Construct a dynamic hysteresis model of the output force-voltage of a piezoelectric ceramic actuator

[0048] Based on the improvement of the classical hysteresis model, the Prandtl-Ishlinskii (PI) model is selected as the basic framework. This model can better describe the hysteresis characteristics and has certain flexibility and scalability. Aiming at the limitations of the PI model in practical applications, a unilateral aubacklash operator with asymmetric characteristics is proposed. By improving the threshold distribution characteristics and adding an asymmetric correction coefficient, the strict symmetry of the backlash operator output is eliminated, and the corresponding API model is obtained;

[0049] Force-electricity hysteresis operator:

[0050] H a , r , b , c [ u ] ( t ) = max { au + r ,min { bu + c , H ( t − 1 ) } } (1)

[0051] Output force-voltage dynamic hysteresis API model:

[0052] F API [ u ] ( t ) = ∑ i = 1 n w i H a i , r i , b i , c i [ u ] ( t ) = w T H [ u ] ( t ) (2)

[0053] Modify the threshold parameter in the API model and associate the original hysteresis operator threshold parameter with the frequency:

[0054] (3)

[0055] Make the calculation output of the hysteresis model change with the change of the input voltage frequency, and obtain the output force-voltage dynamic hysteresis model of the piezoelectric ceramic actuator. Among them, H a , r , b , c [ u ] ( t ) is the output of the force-electricity hysteresis operator, is the output of the hysteresis operator at the previous moment, is the input voltage, 、 are the asymmetric correction coefficients of the hysteresis operator, is the threshold of the force-electricity hysteresis operator, is a numerical constant, F API [ u ] ( t ) is the output force of the piezoelectric ceramic under the force-electricity hysteresis, is the number of force-electricity hysteresis operators, is the weight of the force-electricity hysteresis operator, is the weight matrix of the force-electricity hysteresis operator, H [ u ] ( t ) is the force-electricity hysteresis operator matrix, is a numerical constant, is the voltage derivative, is the initial value of the numerical constant;

[0056] The third step: Solve the compensation output force of the hysteresis inverse model of the piezoelectric ceramic actuator

[0057] The numerical method is used to solve the hysteresis inverse model. First, the output force data at different input voltages and input frequencies are generated by the piezoelectric ceramic force-position output characteristic test platform built, forming a data set. A neural network of the input voltage and output force of the piezoelectric ceramic actuator is constructed through MATLAB software. The output force and actuation frequency are used as the input layer neurons, and the input voltage is used as the output layer neuron. The initial weights and thresholds of each node are set as random numbers between 1 and -1. The neural network is trained with a large amount of data to enable it to learn the mapping law from the output force and frequency to the input voltage, so as to obtain the inverse model. When a desired output force is given, it is input into the inverse model to obtain the corresponding compensation voltage, and then this compensation voltage is applied to the piezoelectric ceramic actuator as the actual input voltage, so as to compensate for the output force lost due to hysteresis and make the output force closer to the desired output force;

[0058] Step 4: Construct the variable parameter algorithm of the hysteresis model under the variable temperature environment

[0059] Based on the above API dynamic hysteresis model, a temperature-related term is introduced. Let the ambient temperature be T, and a temperature variation function h(T) with respect to time is constructed. The asymmetric correction coefficient of the original hysteresis operator in the API model is associated with the temperature variation function h(T) with respect to time to reflect the influence of temperature on the model parameters and make the model characteristics change with temperature. Then, the ambient temperature data is fed back by the thermocouple to select the appropriate model parameters to improve the universality of the model;

[0060] (4)

[0061] The optimized API dynamic hysteresis model is embedded in the control system of the piezoelectric ceramic actuator. Preset experiments are carried out on the piezoelectric ceramic force-position output device. According to the real-time temperature and the expected output force, appropriate parameters are automatically matched to solve the required input voltage. At the same time, the output force of the piezoelectric ceramic actuator is measured in real time and compared with the expected output value to verify the effect of the established adaptive control method for the output force of the piezoelectric ceramic actuator.

[0062] The proposed adaptive control method for the output force of the piezoelectric ceramic actuator under the temperature-varying environment in this method. The established API dynamic hysteresis model of the piezoelectric ceramic actuator takes into account the influence of the dynamic hysteresis of the piezoelectric ceramic and the external environmental temperature, and can accurately solve the input voltage of the piezoelectric ceramic actuator under different temperature environments. It solves the problems that the dynamic hysteresis error of the piezoelectric ceramic actuator is too large and the output performance of the piezoelectric ceramic actuator is limited due to large temperature changes in the actual system, making the control result more accurate and with higher precision; The model establishment and training are convenient and fast, with strong adaptability, and can be applied to all systems containing piezoelectric ceramic actuators.

Claims

1. A method for adaptively controlling the output force of piezoelectric ceramics under temperature-varying conditions, characterized in that: Here are the steps: Step 1: Build a test platform for the output characteristics of piezoelectric ceramic actuators under temperature variation The piezoelectric ceramic actuator is installed on the piezoelectric ceramic output characteristic test device, and a preload force is applied to the piezoelectric ceramic actuator; the power amplifier is connected to the piezoelectric ceramic actuator; the piezoelectric ceramic actuator, the thermocouple, and the force sensor are all connected to the data acquisition module; the input voltage and output force data of the piezoelectric ceramic actuator at different temperatures are collected; Step 2: Construct the piezoelectric ceramic actuator output force-voltage dynamic hysteresis API model The PI model is selected as the basic framework, and a unilateral aubacklash operator with asymmetric characteristics is proposed. By improving the threshold distribution characteristics and adding an asymmetric correction coefficient, the strict symmetric characteristics of the backlash operator output are eliminated, and the output force-voltage dynamic hysteresis API model is obtained. Mechano-electric hysteresis operator: （1） Output force-voltage dynamic hysteresis API model: （2） Modify the threshold parameters in the output force-voltage dynamic hysteresis API model and associate the original hysteresis operator threshold parameters with frequency: （3） The output of the hysteresis model changes with the input voltage frequency, and the output force-voltage dynamic hysteresis API model of the piezoelectric ceramic actuator is obtained; in, is the output of the force-electric hysteresis operator, is the hysteresis operator output at the previous moment, is the input voltage, , is the asymmetric correction coefficient of the hysteresis operator, is the threshold of the mechano-electric hysteresis operator, is a numeric constant, is the output force of the piezoelectric ceramic under force-electric hysteresis, is the number of force-electric hysteresis operators, is the weight of the electro-mechanical hysteresis operator, is the weight matrix of the electromechanical hysteresis operator, is the electromechanical hysteresis operator matrix, is a numeric constant, is the voltage derivative, is the initial value of the numerical constant; Step 3: Solve the hysteresis inverse model of the piezoelectric ceramic actuator to compensate the output force The hysteresis inverse model is solved by a numerical method. First, the output characteristic test platform is used to generate output force data generated under different input voltages and input frequencies to form a data set. A neural network is constructed, with the output force and input frequency as input layer neurons and the input voltage as output layer neurons. The neural network is trained with the data in the data set to learn the mapping law from output force, input frequency to input voltage, and obtain the hysteresis inverse model. When the desired output force is given, it is input into the hysteresis inverse model to obtain the corresponding compensation voltage, and then the compensation voltage is applied to the piezoelectric ceramic actuator as the actual input voltage, thereby compensating for the output force lost due to hysteresis and making the output force closer to the desired output force. Step 4: Construct variable parameter algorithm for hysteresis model under variable temperature environment On the basis of the output force-voltage dynamic hysteresis API model, the temperature-related term is introduced, the ambient temperature is set as T, and the temperature-time variation function h(T) is constructed. The original asymmetric correction coefficient of the hysteresis operator in the output force-voltage dynamic hysteresis API model is associated with the temperature-time variation function h(T) to reflect the effect of the ambient temperature on the output force-voltage dynamic hysteresis API model parameters, so that the output force-voltage dynamic hysteresis API model characteristics change with temperature, and then the adaptive output force-voltage dynamic hysteresis API model parameters are selected through temperature feedback to improve the universality of the output force-voltage dynamic hysteresis API model. （4） in, , is the temperature correction coefficient, is the temperature change, , is the offset of the temperature correction coefficient; The optimized output force-voltage dynamic hysteresis API model is embedded in the control system of the piezoelectric ceramic actuator, and the required input voltage is automatically solved by automatically matching appropriate parameters according to the real-time temperature and expected output force, thus achieving precise and intelligent control.

Images