Piezoelectric micromanipulator dataset for hysteresis identification

This data in brief presents displacement measurements acquired from a piezoelectric cantilevered actuator when subjected to harmonic excitations. The micro displacements are measured with optical sensors. The dataset has been used recently for the purpose of nonlinear black-box modelling, where the hysteretic behaviour of such devices has been modelled [1,2]. We hope to enable reproducibility by sharing the data used in [1,2], which are previous works by the authors, allowing the comparison of new methods on a common basis. Additionally, researchers interested in piezoelectric actuators for high precision tasks may also benefit on working with the present dataset.


a b s t r a c t
This data in brief presents displacement measurements acquired from a piezoelectric cantilevered actuator when subjected to harmonic excitations. The micro displacements are measured with optical sensors. The dataset has been used recently for the purpose of nonlinear black-box modelling, where the hysteretic behaviour of such devices has been modelled [1,2]. We hope to enable reproducibility by sharing the data used in [1,2], which are previous works by the authors, allowing the comparison of new methods on a common basis. Additionally, researchers interested in piezoelectric actuators for high precision tasks may also benefit on working with the present dataset. © 2020 The Author(s). Published by Elsevier Inc. This is an open access article under the CC BY license (http://creativecommons. org/licenses/by/4.0/).

Data description
The dataset is composed of two input/output data pairs. The system is excited with a sine voltage input of (a) 150 V and 1 Hz; and (b) 5 V and 300 Hz. The dataset (b) has been employed for identification in Refs. [1,2]. The goal of measuring both datasets is to evaluate the modelling activity when working under different frequencies and amplitudes. Dataset (a) has not been explored thus far in any publications.
The datasets (a) and (b) are provided in two comma separated values (CSV) files. In these files, which can be visualized in any text editor, each line refers to a sampling time instant. Both CSV files are zipped in a single file, which is provided as a supplement to this article. In this zip file, there is also a MATLAB code to plot the data. It is possible to find below a detailed description of both datasets, according to their filename: (a) h50us.csv: this dataset contains 200,001 measurements sampled in time every 50 microseconds. The second, third, and fourth columns are respectively the vector with time, output displacement, and input voltage. The measurements are drifted, so a pre-processing is needed (in the code provided this is already arranged). (b) hysteresis_v_150_1hz.csv: the second dataset has the same file structure with respect to the columns as (a) but measured every 20 milliseconds with 50,001 samples. The sinusoidal signal starts at approximately 3.25 seconds, so the first samples of the dataset should be discarded (in the code provided this is already arranged).
Specifications Table   Subject Control and Systems Engineering Specific subject area System identification and hysteresis modelling Type of data Deflection and displacement measurements How data were acquired Optical sensors Data format Raw Parameters for data collection The piezoelectric actuator was excited using a sine driving voltage. The amplitudes and frequencies tested were 5V and 150 V and 300Hz and 1 Hz, respectively. Description of data collection The entire test bench constructed to collect data is composed of (a) the piezoelectric actuator, (b) an optical sensor (LK2420 from Keyence company) which is employed to measure the deflection (displacement) of the above actuator and has been tuned to have 10nm resolution and in excess of 5kHz bandwidth, (c) a computer which is used to generate the sine driving voltage and to acquire the measurement from the optical sensor, (d) a dSPACE (type DS1104) acquisition board that serves as digital-to-analogic and as analogicto-digital converters between the computer and the rest of the physical setup, with

Value of the Data
The dataset provided is important for nonlinear modelling of hysteretic systems Researchers in the system identification community at large may benefit for testing nonlinear modelling techniques. Researchers and engineers working with piezoelectric actuators for high precision positioning applications may also be interested with and benefit from the data. The present dataset enable comparison among methods for modelling a phenomenon that is frequently found in positioning applications, but not only. Table 1 and depicted graphically in Fig. 1. They are sampled at different rates as the dynamics due to the excitation is faster in case (a). It is possible to see that the voltage amplitudes and frequencies are different for each file, as Table 1 shows. Nonetheless, the amplitudes for the deflections are in the same order of magnitude, as the input/output gain for each frequency is different for the system. For the sinusoidal-like type of input, it is possible to plot the graph with input versus output through time, where the hysteretic behaviour can be clearly observed.

Experimental design, materials, and methods
A schematic of the experimental benchmark setup is described in Fig. 2. The components used in this setup are described in detail in Table 2. The piezoelectric micromanipulator is manufactured with  Fig. 3. The dataset is provided with a MATLAB code (read_plot_data.m) that reads the data into memory and plots the graphs given in Fig. 1.   Fig. 2. Schematic of the measurement setup and the data flow among its components.

Table 2
Hardware used for instrumentation of the test bench and respective purpose description.

Hardware
Manufacturer Purpose DS1104 dSPACE -converts the sine driving voltage generated from MATLAB-Simulink in the computer into analogic voltage outside the computer, -converts the measured displacement from the sensor into numeric measurement inside MATLAB-Simulink. LK2420 Keyence Optical sensor that measures the deflection (displacement) of the piezoelectric actuator. A400DI FLC Amplifies the voltage from the acquisition board and computer before driving the piezoelectric actuator. Computer (with MATLAB-Simulink) Any MATLAB-Simulink has been used to program the voltage to be amplified and sent to the actuator and used to save or display the measurement from the sensor.