Sensitivity of Piezoelectric Stack Actuators

Xishan Jiang, Jing Zheng, Ning Wang, Jie PanSensors 2023 · DOI

TL;DR

This paper introduces a novel definition of sensitivity for piezoelectric stack actuators, focusing on the spectral output force in response to a single-frequency voltage input. By modeling the actuator as a two-part system (nonlinear hysteresis and linear mechanical), they derive analytical expressions for its steady-state response, revealing that the output force contains harmonics of the input frequency. This work provides a framework for better understanding and designing piezoelectric actuators for applications requiring precise force generation.

What Problem Does This Paper Address?

Piezoelectric stack actuators are widely used for precise motion control due to their ability to convert electrical energy into mechanical energy. However, a significant challenge in their application is the inherent nonlinear hysteresis effect. This hysteresis means the actuator's output (displacement or force) depends not only on the current input but also on its past states, making precise control difficult. While various models exist to describe this hysteresis, a clear and practical metric to evaluate the actuator's performance, specifically its ability to generate a desired force output at specific frequencies, has been lacking. This paper aims to address this by defining and analyzing the "sensitivity" of piezoelectric actuators, focusing on how effectively they translate a single-frequency electrical input into a predictable force output spectrum. This is particularly relevant for applications like active acoustic control where specific frequency responses are critical.

Methods

The authors propose a two-subsystem model for the piezoelectric stack actuator:

  1. Nonlinear Hysteresis Subsystem: This part models the core hysteresis behavior. They employ a modified Bouc-Wen model, a well-established phenomenological model for hysteresis, to describe the relationship between input voltage and the internal "hysteresis displacement" ($h$).
  2. Linear Mechanical Subsystem: This part represents the physical dynamics of the actuator, including a proof mass, stiffness, and damping. This subsystem is treated as a linear oscillator.

The core of their method involves:

  • Defining Sensitivity: Sensitivity is defined as the ratio of the output force spectrum to the input voltage amplitude at a specific frequency.
  • Analytical Modeling: They derive analytical solutions for the nonlinear differential equations governing the system. This involves approximating the hysteresis behavior and using Fourier series expansions to analyze the resulting force spectrum. They specifically focus on the steady-state response, showing that a single-frequency input voltage generates an output force spectrum containing the driving frequency and its odd harmonics.
  • State-Space Representation: The coupled nonlinear differential equations are converted into a state-space model for numerical analysis.
  • Numerical Simulation: The analytical model is validated using numerical simulations in MATLAB.
  • Experimental Validation: A laboratory experiment is conducted using a piezoelectric stack actuator with a proof mass to measure its vibration response. The experimental results are then compared with the analytical and numerical models.

Datasets

Not discussed in the paper. The study relies on simulated data generated from their model and experimental data collected from a custom-built setup.

Key Insights and Ideas

  • Hysteresis as a Harmonic Generator: The most striking insight is that the nonlinear hysteresis mechanism inherently transforms a single-frequency input voltage into a driving force that contains not only the fundamental frequency but also its odd harmonics. This is a direct consequence of the non-linear nature of the hysteresis loop.
  • Two-Stage System for Sensitivity: Decomposing the actuator into a nonlinear hysteresis stage followed by a linear mechanical stage provides a powerful framework for understanding and predicting its behavior. The sensitivity is then determined by how the linear stage responds to the harmonic-rich force generated by the nonlinear stage.
  • Resonance for Maximum Sensitivity: The analysis reveals that maximum sensitivity can be achieved by tuning the input voltage frequency close to the natural frequency of the linear mechanical subsystem. This is a crucial finding for designing actuators for specific frequency responses.
  • Analytical Tractability of Nonlinearity: While nonlinear, the authors demonstrate that the steady-state response of the hysteresis can be approximated analytically, allowing for the derivation of closed-form expressions for the force spectrum. This is a significant step beyond purely numerical approaches for certain aspects of actuator analysis.

Key Results

  • Harmonic Content: The piezoelectric driving force spectrum, when driven by a single-frequency voltage, was found to be dominated by the fundamental frequency and its odd harmonics (e.g., $F_{PZT}(\omega_0)$ and $F_{PZT}(3\omega_0)$). Higher-order harmonics were significantly smaller.
  • Analytical Model Validation: The derived analytical expressions for the steady-state force spectrum and displacement showed good agreement with numerical simulations and experimental measurements. For instance, the error between experimental and analytical/numerical results was reported to be less than 1 dB at 30 Hz.
  • Sensitivity Dependence: The sensitivity was shown to be dependent on both the driving frequency and the amplitude of the input voltage. The increase in force components with voltage amplitude was not linear, as expected from a nonlinear system.
  • Resonance Observation: Experimental results clearly showed resonance peaks in the sensitivity spectrum when the driving frequency approached the natural frequency of the actuator system.

Limitations and Future Work

  • Proof Mass Requirement: The current model and experimental setup utilize a proof mass. The authors suggest that future work should consider configurations that avoid the need for a heavy proof mass, potentially simplifying the system.
  • Spillover Effect: The paper acknowledges the need to discuss and suppress the "spillover effect," which refers to unwanted frequency components generated by the nonlinear mapping from the input to the output.
  • Other Configurations: The authors mention that other actuator configurations should be explored to generalize their findings.
  • Heavy Proof Mass: The experimental validation used a relatively heavy proof mass (1000 g), which might influence the system dynamics. Future work could explore the impact of lighter proof masses or different mounting strategies.

Summary updated

BibTeX
@article{jiang2023sensitivity,
  author = {Xishan Jiang and Jing Zheng and Ning Wang and Jie Pan},
  title = {{Sensitivity of Piezoelectric Stack Actuators}},
  journal = {Sensors},
  year = {2023},
  doi = {10.3390/s23239542},
}
Figure