Minimizing Off-axis Bending Effects on Flexible Surface Acoustic Wave Sensing - Powered by Integrated Machine Learning Algorithms

Aug 5, 2026

Upload by Maeve

Flexible surface acoustic wave (SAW) sensors are attractive for curved-surface conformability, wireless/passive operation, and digital output, yet off-axis bending can strongly distort sensing signals and restrict accurate monitoring on deformed geometries. The study uses AlScN/ultra-thin flexible glass SAW devices and targets temperature as the sensing parameter. Eight machine learning models learn correlations between response features and temperature, then the optimized approach achieves normalized RMSE under 1% and R2 above 0.997 under complex off-axis strain perturbations, verified on a jet-engine model surface.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages8

Reading Time20 min

AccessFree to download

Introduction

Flexible surface acoustic wave sensing advantages and applications

Off-axis bending interference problem

Method and modeling approach

Device selection and temperature as targeted parameter

Response characterization under off-axis deformation and temperature changes

Machine learning correlation and optimized prediction performance

Experimental and validation results

Robust temperature sensing under arbitrary off-axis bending on a curved surface

Minimizing Off-axis Bending Effects on Flexible Surface Acoustic Wave Sensing Powered by Integrated machine learning algorithms Zhangbin Ji1, Jian Zhou 1,*, Yihao Guo 1, Yanhong Xia 1, Dongfang Liang2, YongQing Fu3 Abstract—Flexible surface acoustic wave (SAW) sensors have gained significant attention due to their favorable attributes such as conformability to curved surfaces, wireless/passive functions, and digital outputs. However, bending, especially complex off-axis bending deformation, often causes severe interference to the targeted detection signals with flexible SAW sensors, limiting their accurate monitoring on the curved/deformed surfaces. To address such a critical issue, we selected AlScN/ultra-thin flexible glass-based SAW devices as an example, chose temperature as the targeted sensing parameter, and developed a model based on machine learning algorithms to minimize complex off-axis bending effects in temperature monitoring. Response characteristics of the flexible SAW devices to temperature variations and off-axis deformations were experimentally and theoretically investigated. Correlations between device’s responsive features and target parameter (temperature) were established using eight machine learning algorithms. The optimized model was established with a normalized root mean square error less than 1% and the determination coefficient R2 was larger than 0.997 for temperature predictions subject to complex off-axis strain perturbations. Finally, the flexible SAW sensor showed a highly consistent temperature sensing capability under arbitrary off-axis bending conditions on a curved surface of a jet engine model. Index Terms—Off-axis bending, Flexible SAW detection, Temperature, Machine learning. Anti-interference. I. INTRODUCTION Surface acoustic wave (SAW) sensors have received extensive interests owing to their wireless/passive nature, compact structures, and suitability for being used in harsh environments [1, 2] . In recent years, ultra-thin, bendable and flexible SAW sensors have received extensive interests [3-6] . Compared to those traditionally rigid SAW sensors, flexible SAW sensors are easily embedded onto curved or bent surfaces of object/equipment/device without adding extra volume or weight, thus avoiding or mitigating interference with the This work was supported by the National Science Foundation of China (No. 52075162), and the Science and Technology Innovation Program of Hunan Province (2023RC3099) . We also thank the Corning Corporation for the provision of flexible glass. (Corresponding author: Jian Zhou. E-mail: [j](jianzhou@hnu.edu.cn)[ianzhou@hnu.edu.cn](jianzhou@hnu.edu.cn);) Zhangbin Ji, Jian Zhou, Yihao Guo and Yanhong Xia are with the College of Mechanical and Vehicle Engineering, Hunan University, Changsha 410082, China (E-mail: [j](jizb@hnu.edu.cn)[izb@hnu.edu.cn](jizb@hnu.edu.cn); [j](jianzhou@hnu.edu.cn)[ianzhou@hnu.edu.cn](jianzhou@hnu.edu.cn); [yihaoguo@hnu.edu.cn](yihaoguo@hnu.edu.cn); [xiayh@hnu.edu.cn](xiayh@hnu.edu.cn)) original functionality or performance of the curved device. Due to these unique and appealing advantages, various SAW devices based on flexible substrates, such as metal foils, polymers, and flexible glass, have been developed [7-12] . These flexible SAW devices have suitable applications such as electronic skin [4], and can be used to detect and monitor targeted substances such as temperature [6, 10], strain [1, 13, 14], humidity [15, 16], ultraviolet (UV) radiation [17, 18], pH value [19], and Escherichia coli [20], on either bent or curved surfaces. However, practical applications of these flexible SAW sensors on curved surfaces often encounter some critical issues. When the flexible SAW sensors are deployed on surfaces with varied curvatures, different levels of bending strains will be generated. The sensing signals will exhibit significant variations, as both changes of the bending strains and the targeted signals will lead to freque

FAQs

Why does off-axis bending interfere with flexible SAW temperature sensing?

Off-axis bending changes the frequency shift not only through strain magnitude but also through the off-axis angle between acoustic wave propagation and strain direction, which distorts the targeted temperature signal.

What sensor platform and targeted parameter are used in the study?

The work selects AlScN/ultra-thin flexible glass-based SAW devices and uses temperature as the sensing parameter to reduce bending-induced interference.

How do the machine learning models improve temperature prediction accuracy?

Eight machine learning algorithms establish correlations between sensing response features and temperature, producing an optimized model with normalized RMSE below 1% and R2 greater than 0.997 under complex off-axis strain perturbations.

Aug 5, 2026

Upload by Maeve