Modeling, Simulation, Fabrication, and Defect Analysis of Polyvinylidene Fluoride (PVDF) Acoustic Sensors
Abstract
Infrastructural damage to military vehicles due to corrosion remains costly and is a major concern for the Navy. To minimize the future risk of excessive damage to vehicles, it is imperative that efficient and reliable sensors are developed to track the progression of corrosion. One such sensor is a surface acoustic wave (SAW) sensor which is a class of micro-electrical mechanical sensors (MEMS). In this research, we developed a systematic approach to model, simulate, fabricate, and analyze a PVDF-based SAW sensor. The system consisted of three primary components 1) model and simulate the PVDF-based SAW sensor 2) fabricate the PVDF-based SAW sensor using the photolithographic processes and use optical microscopy to image the sensors 3) detect and classify the presence of defects within the PVDF-based SAW sensor. Following the system engineering approach, the proposed system was composed into two subsystems: SAW Sensor Design subsystem, and Defect Detection Model subsystem. All subsystems were developed, integrated, verified, and validated in this dissertation. To evaluate the proposed system, deep learning was employed to develop an autoencoder convolutional neural network to classify features with the optical image data. The efficacy of the autoencoder was validated using statistical analysis that evaluated classification accuracy.
Subject Area
Materials science|Computer science|Engineering|Acoustics
Recommended Citation
Omari Paul,
"Modeling, Simulation, Fabrication, and Defect Analysis of Polyvinylidene Fluoride (PVDF) Acoustic Sensors"
(2022).
ETD Collection for Tennessee State University.
Paper AAI29992086.
https://digitalscholarship.tnstate.edu/dissertations/AAI29992086