Inverse Nonlinearity Compensation of Hyperelastic Deformation in Dielectric Elastomer for Acoustic Actuation

Jin Woo Lee,Gwang Seok An,Jeong-Yun Sun,Kyogu Lee

2024-01-08

Abstract:This paper delves into the analysis of nonlinear deformation induced by dielectric actuation in pre-stressed ideal dielectric elastomers. It formulates a nonlinear ordinary differential equation governing this deformation based on the hyperelastic model under dielectric stress. Through numerical integration and neural network approximations, the relationship between voltage and stretch is established. Neural networks are employed to approximate solutions for voltage-to-stretch and stretch-to-voltage transformations obtained via an explicit Runge-Kutta method. The effectiveness of these approximations is demonstrated by leveraging them for compensating nonlinearity through the waveshaping of the input signal. The comparative analysis highlights the superior accuracy of the approximated solutions over baseline methods, resulting in minimized harmonic distortions when utilizing dielectric elastomers as acoustic actuators. This study underscores the efficacy of the proposed approach in mitigating nonlinearities and enhancing the performance of dielectric elastomers in acoustic actuation applications.

Audio and Speech Processing,Sound

What problem does this paper attempt to address?

### What problem does this paper attempt to solve?

This paper aims to solve the problem of harmonic distortion caused by nonlinear deformation during the acoustic driving process of dielectric elastomers (DEAs). Specifically, DEAs will undergo significant nonlinear elastic deformation under voltage driving, and this nonlinearity will lead to harmonic distortion of the output signal. Especially when applied as an acoustic actuator, these distortions are perceptible to the human ear.

To overcome this challenge, the authors propose inverse nonlinearity compensation (INC) through waveshaping technology. They establish a nonlinear ordinary differential equation (ODE) based on a hyperelastic model and use numerical integration and neural network approximation methods to solve the relationship between voltage and stretch. The ultimate goal is to find a mapping function \( g(x) \) so that the compensated input signal can linearly control the stretch of the DEA, thereby minimizing harmonic distortion.

### Main problem description

1. **Modeling system**: Construct an ODE model that describes the voltage - stretch relationship.
2. **Control system**: Find a signal mapping function \( g(x) \) such that the compensated DEA stretch \( \hat{\lambda} = f(\alpha \cdot g(x)) \) remains linearly related to the original signal \( x \).

### Mathematical expression

Let the stretch of the DEA be \( \lambda \), which is determined by pre - stretch and electromagnetic deformation and is affected by the amplified high - voltage signal \( V=\alpha \cdot x \). The research focuses on two aspects:

1. **Modeling**: Construct an ODE model that describes the voltage - stretch system and obtain the relationship \( f: \mathbb{R}\to\mathbb{R} \) between voltage and stretch by solving this ODE, that is:
\[
V \mapsto \lambda
\]

2. **Control**: Find a signal mapping function \( g: \mathbb{R}\to\mathbb{R} \) such that the compensated DEA stretch \( \hat{\lambda} = f(\alpha \cdot g(x)) \) remains linearly related to the original signal \( x \), that is:
\[
g(x) \mapsto \hat{x}
\]
and satisfies:
\[
f(\alpha \cdot g(x)) = a_0 + a_1 x
\]
where \( a_0, a_1\in\mathbb{R} \) and \( a_1\neq 0 \).

### Solution

To achieve the above goals, the authors propose two main methods:

1. **Numerical integration method**: Use the explicit Runge - Kutta method (RK45) to solve the ODE and obtain an approximate solution of the voltage - stretch relationship.
2. **Neural network approximation**: Use a multi - layer perceptron (MLP) to implicitly learn the mapping function of the voltage - stretch relationship and optimize the compensation function \( g(x) \) through end - to - end training.

### Result evaluation

By comparing the performance of different methods (such as power function fitting, piecewise quartic interpolation, etc.), the authors show the superiority of the proposed neural network approximation method in reducing harmonic distortion. Evaluation metrics include the \( \log\ell_1 \)-norm of stretch differences, the \( \ell_{STFT}^1 \)-norm of spectrogram differences, and the signal - to - distortion ratio (SDR). Experimental results show that the neural network method outperforms other baseline methods in all metrics. In particular, the SDR shows a higher signal - to - noise ratio, proving its effectiveness in acoustic applications.

In summary, this paper successfully solves the problem of harmonic distortion caused by nonlinear deformation of DEA in acoustic driving by combining numerical integration and neural network approximation, and significantly improves the DEA.