Georgia Institute of Technology
Real-Time Corrosion Monitoring and AI-Driven EIS Spectra Prediction for Aluminium Beverage Cans
Abstract
dc:description.abstractCoatings applied to the interior surfaces of aluminum beverage can-lids play a critical role in preventing corrosion and maintaining the integrity of the product throughout its shelf life. However, these coatings are susceptible to long-term degradation due to mechanical stresses during fabrication and the corrosive nature of various beverages. Traditional multi-month pack tests are laborious and time-consuming, prompting the need for faster, yet accurate, alternatives. This thesis presents an innovative method for evaluating can lid coating degradation by incorporating in-situ Electrochemical Impedance Spectroscopy (EIS) monitoring under real-world conditions using actual beverages. A custom-engineered testing chamber was designed to house the complex 3D geometry of the lids and tested in various condition sincluding a variety of different beverages and simulants. The developed methodology enables accelerated yet insightful coating performance assessments and has the potential to serve as an efficient surrogate to conventional long-duration tests. In parallel, this research explores the integration of machine learning into electrochemical analysis to enhance corrosion prediction capabilities. A Long Short-Term Memory (LSTM) neural network was developed to predict EIS spectra, enabling real-time and data-driven assessment of coating degradation. The model was trained using impedance data collected from the custom test setup and utilized a three-dimensional input structure comprising frequency, real impedance (Z′), and imaginary impedance (Z″). The LSTM model effectively captured dynamic changes across Nyquist, Bode magnitude, and phase angle plots, outperforming conventional analytical methods. This study demonstrates the potential of combining advanced electrochemical techniques with machine learning to revolutionize corrosion monitoring. The approach not only improves the evaluation of coated aluminum lids but also offers scalability to broader applications involving aluminum and its alloys in diverse industrial environments.
Degree
thesis:*- Level thesis:degree_level
- Masters
- Department dc:contributor.department
- Materials Science and Engineering
- Grantor dc:publisher
- Georgia Institute of Technology
- Year dc:date.issued
- 2025
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Jegan, Haeshini
- Advisor dc:contributor.advisor
-
- Liu, Nian
Subjects
dc:subject × 7Rights
- Language dc:language.iso
- en_US
Identifiers
dc:identifier.*- Handle dc:identifier.uri
- https://hdl.handle.net/1853/81571