{"id":{"repo_id":"tamu","oai_identifier":"oai:oaktrust.library.tamu.edu:1969.1/1599914"},"canonical_url":"https://search.dev.ndltd.org/etd/tamu/oai:oaktrust.library.tamu.edu:1969.1/1599914","repository":{"repo_id":"tamu","name":"Texas A&M University","base_url":"https://oaktrust.library.tamu.edu/server/oai/request"},"display":{"title":"Predictive Analysis of Corrosion Resistance in Automotive Coatings Using Artificial Neural Networks","abstract":"This research integrates advanced methodologies to analyze the degradation and long-term performance of various coating systems. Employing Electrochemical Impedance Spectroscopy (EIS) and Artificial Neural Networks (ANN), we monitored barrier coatings over an extended period of 1000 days to form a baseline model. The primary aim was to analyze time series data to predict degradation stages and quantify the state of the coatings, thus providing a comprehensive understanding of their performance under various conditions. Low-frequency EIS data played a crucial role in this analysis, offering significant insights into the mechanistic interpretation of the coatings and capturing dynamic changes occurring over specific intervals. The ANN models developed in this study were employed in three different coating types, Physical Barrier Coating, Sacrificial Metal Coating and Hybrid Coating, then they are validated through scanning electron microscopy (SEM), which ensured the accuracy of predictions regarding the performance of three automotive coatings - physical barrier, sacrificial, and hybrid - under simulated corrosive conditions. Phase angle plots obtained from EIS measurements facilitated continuous monitoring of the coating states during steady-state conditions. These plots were also instrumental in training the ANN models for Time Series Prediction (TSP), allowing the ANN to learn features and dynamics directly from the data, bypassing the need for extensive parameter optimization based on domain experience. To simulate real-world conditions that automotive coatings face, such as those involving de-icing salts, an accelerated wet-dry cycling protocol was implemented. This protocol was designed based on snow-depth data collected from the Rust Belt states, spanning from 2015 to the winter of 2021, obtained from the Global Historical Climatology Network Daily (GHCNd) database. The cycling protocol involved 96 hours of wet conditions followed by 73 hours of dry conditions within a Cyclic Corrosion Test (CCT) chamber, using a 5% NaCl solution at 30 degrees C. Four distinct types of coating samples were prepared, each with varying corrosion control mechanisms and scratch depths, to evaluate their performance under these accelerated conditions. Real-time electrochemical assessments were conducted on these samples, and the data was analyzed over a 30-day period. The phase angle plot analysis revealed significant insights into the corrosion behavior of different coatings. Samples with intact coatings exhibited a gradual onset of corrosion, resembling steady-state conditions, indicating effective corrosion protection. Partially scratched samples showed a delayed activation of corrosion mechanisms, emphasizing the protective capabilities of sacrificial coatings. Conversely, samples with complete scratch depth displayed immediate activation of corrosion processes, highlighting the absence of a protective barrier and the vulnerability of the underlying substrate. This comprehensive approach, combining real-time EIS testing, ANN modeling, and accelerated corrosion protocols, provides a robust experimental-theoretical framework for predicting and understanding coating degradation and corrosion mechanisms. The integration of mechanistic and machine-learning concepts allows for the development of more accurate and reliable predictions of coating performance over extended periods. The findings from this study contribute significantly to the advancement of corrosion protection strategies in automotive and industrial applications. By understanding the intricate dynamics of coating degradation, more effective and durable protective coatings can be designed, ultimately enhancing the longevity and reliability of metallic substrates in harsh environmental conditions.","abstract_html":"This research integrates advanced methodologies to analyze the degradation and long-term performance of various coating systems. Employing Electrochemical Impedance Spectroscopy (EIS) and Artificial Neural Networks (ANN), we monitored barrier coatings over an extended period of 1000 days to form a baseline model. The primary aim was to analyze time series data to predict degradation stages and quantify the state of the coatings, thus providing a comprehensive understanding of their performance under various conditions. Low-frequency EIS data played a crucial role in this analysis, offering significant insights into the mechanistic interpretation of the coatings and capturing dynamic changes occurring over specific intervals. The ANN models developed in this study were employed in three different coating types, Physical Barrier Coating, Sacrificial Metal Coating and Hybrid Coating, then they are validated through scanning electron microscopy (SEM), which ensured the accuracy of predictions regarding the performance of three automotive coatings - physical barrier, sacrificial, and hybrid - under simulated corrosive conditions. Phase angle plots obtained from EIS measurements facilitated continuous monitoring of the coating states during steady-state conditions. These plots were also instrumental in training the ANN models for Time Series Prediction (TSP), allowing the ANN to learn features and dynamics directly from the data, bypassing the need for extensive parameter optimization based on domain experience. To simulate real-world conditions that automotive coatings face, such as those involving de-icing salts, an accelerated wet-dry cycling protocol was implemented. This protocol was designed based on snow-depth data collected from the Rust Belt states, spanning from 2015 to the winter of 2021, obtained from the Global Historical Climatology Network Daily (GHCNd) database. The cycling protocol involved 96 hours of wet conditions followed by 73 hours of dry conditions within a Cyclic Corrosion Test (CCT) chamber, using a 5% NaCl solution at 30 degrees C. Four distinct types of coating samples were prepared, each with varying corrosion control mechanisms and scratch depths, to evaluate their performance under these accelerated conditions. Real-time electrochemical assessments were conducted on these samples, and the data was analyzed over a 30-day period. The phase angle plot analysis revealed significant insights into the corrosion behavior of different coatings. Samples with intact coatings exhibited a gradual onset of corrosion, resembling steady-state conditions, indicating effective corrosion protection. Partially scratched samples showed a delayed activation of corrosion mechanisms, emphasizing the protective capabilities of sacrificial coatings. Conversely, samples with complete scratch depth displayed immediate activation of corrosion processes, highlighting the absence of a protective barrier and the vulnerability of the underlying substrate. This comprehensive approach, combining real-time EIS testing, ANN modeling, and accelerated corrosion protocols, provides a robust experimental-theoretical framework for predicting and understanding coating degradation and corrosion mechanisms. The integration of mechanistic and machine-learning concepts allows for the development of more accurate and reliable predictions of coating performance over extended periods. The findings from this study contribute significantly to the advancement of corrosion protection strategies in automotive and industrial applications. By understanding the intricate dynamics of coating degradation, more effective and durable protective coatings can be designed, ultimately enhancing the longevity and reliability of metallic substrates in harsh environmental conditions.","abstract_has_math":false,"creators":["Ponce Valderrama, Victor Hugo 1989-"],"institution":"Texas A&M University","degree_name":"Doctor of Philosophy","degree_level":"Doctoral","degree_discipline":"Materials Science and Engineering","degree_department":null,"school":null,"contributors":[],"advisors":["Castaneda-Lopez, Homero"],"committee_chairs":[],"committee_members":["Mansoor, Bilal","Liang, Hong","Banerjee, Sarbajit"],"year":2025,"date_issued":"2025-12","date_published":"2025-12","updated_at":"2026-08-21T16:48:40Z","subjects":["Engineering, Materials Science"],"languages":["English"],"rights":[],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"https://hdl.handle.net/1969.1/1599914","outbound_label":"Handle","outbound_source":"dc:identifier.uri"},"source_record":{"url":"https://oaktrust.library.tamu.edu/server/oai/request?verb=GetRecord&metadataPrefix=dim&identifier=oai%3Aoaktrust.library.tamu.edu%3A1969.1%2F1599914","prefix":"dim"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Castaneda-Lopez, Homero"]},{"key":"dc:contributor.committeemember","label":"Committee Member","values":["Mansoor, Bilal","Liang, Hong","Banerjee, Sarbajit"]},{"key":"dc:creator","label":"Author","values":["Ponce Valderrama, Victor Hugo 1989-"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2026-03-05T21:01:52Z"]},{"key":"dc:date.issued","label":"Date","values":["2025-12"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Materials Science and Engineering"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Doctoral"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Doctor of Philosophy"]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["Texas A&M University"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Engineering, Materials Science"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language.iso","label":"Language (ISO)","values":["English"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://hdl.handle.net/1969.1/1599914"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["This research integrates advanced methodologies to analyze the degradation and long-term performance of various coating systems. Employing Electrochemical Impedance Spectroscopy (EIS) and Artificial Neural Networks (ANN), we monitored barrier coatings over an extended period of 1000 days to form a baseline model. The primary aim was to analyze time series data to predict degradation stages and quantify the state of the coatings, thus providing a comprehensive understanding of their performance under various conditions. Low-frequency EIS data played a crucial role in this analysis, offering significant insights into the mechanistic interpretation of the coatings and capturing dynamic changes occurring over specific intervals. The ANN models developed in this study were employed in three different coating types, Physical Barrier Coating, Sacrificial Metal Coating and Hybrid Coating, then they are validated through scanning electron microscopy (SEM), which ensured the accuracy of predictions regarding the performance of three automotive coatings - physical barrier, sacrificial, and hybrid - under simulated corrosive conditions. Phase angle plots obtained from EIS measurements facilitated continuous monitoring of the coating states during steady-state conditions. These plots were also instrumental in training the ANN models for Time Series Prediction (TSP), allowing the ANN to learn features and dynamics directly from the data, bypassing the need for extensive parameter optimization based on domain experience. To simulate real-world conditions that automotive coatings face, such as those involving de-icing salts, an accelerated wet-dry cycling protocol was implemented. This protocol was designed based on snow-depth data collected from the Rust Belt states, spanning from 2015 to the winter of 2021, obtained from the Global Historical Climatology Network Daily (GHCNd) database. The cycling protocol involved 96 hours of wet conditions followed by 73 hours of dry conditions within a Cyclic Corrosion Test (CCT) chamber, using a 5% NaCl solution at 30 degrees C. Four distinct types of coating samples were prepared, each with varying corrosion control mechanisms and scratch depths, to evaluate their performance under these accelerated conditions. Real-time electrochemical assessments were conducted on these samples, and the data was analyzed over a 30-day period. The phase angle plot analysis revealed significant insights into the corrosion behavior of different coatings. Samples with intact coatings exhibited a gradual onset of corrosion, resembling steady-state conditions, indicating effective corrosion protection. Partially scratched samples showed a delayed activation of corrosion mechanisms, emphasizing the protective capabilities of sacrificial coatings. Conversely, samples with complete scratch depth displayed immediate activation of corrosion processes, highlighting the absence of a protective barrier and the vulnerability of the underlying substrate. This comprehensive approach, combining real-time EIS testing, ANN modeling, and accelerated corrosion protocols, provides a robust experimental-theoretical framework for predicting and understanding coating degradation and corrosion mechanisms. The integration of mechanistic and machine-learning concepts allows for the development of more accurate and reliable predictions of coating performance over extended periods. The findings from this study contribute significantly to the advancement of corrosion protection strategies in automotive and industrial applications. By understanding the intricate dynamics of coating degradation, more effective and durable protective coatings can be designed, ultimately enhancing the longevity and reliability of metallic substrates in harsh environmental conditions."]},{"key":"dc:format.mimetype","label":"Dc Format Mimetype","values":["application/pdf"]},{"key":"dc:title","label":"Title","values":["Predictive Analysis of Corrosion Resistance in Automotive Coatings Using Artificial Neural Networks"]}]}],"canonical_facts":{"dc:contributor.advisor":["Castaneda-Lopez, Homero"],"dc:contributor.committeemember":["Mansoor, Bilal","Liang, Hong","Banerjee, Sarbajit"],"dc:creator":["Ponce Valderrama, Victor Hugo 1989-"],"dc:date.accessioned":["2026-03-05T21:01:52Z"],"dc:date.issued":["2025-12"],"dc:description.abstract":["This research integrates advanced methodologies to analyze the degradation and long-term performance of various coating systems. Employing Electrochemical Impedance Spectroscopy (EIS) and Artificial Neural Networks (ANN), we monitored barrier coatings over an extended period of 1000 days to form a baseline model. The primary aim was to analyze time series data to predict degradation stages and quantify the state of the coatings, thus providing a comprehensive understanding of their performance under various conditions. Low-frequency EIS data played a crucial role in this analysis, offering significant insights into the mechanistic interpretation of the coatings and capturing dynamic changes occurring over specific intervals. The ANN models developed in this study were employed in three different coating types, Physical Barrier Coating, Sacrificial Metal Coating and Hybrid Coating, then they are validated through scanning electron microscopy (SEM), which ensured the accuracy of predictions regarding the performance of three automotive coatings - physical barrier, sacrificial, and hybrid - under simulated corrosive conditions. Phase angle plots obtained from EIS measurements facilitated continuous monitoring of the coating states during steady-state conditions. These plots were also instrumental in training the ANN models for Time Series Prediction (TSP), allowing the ANN to learn features and dynamics directly from the data, bypassing the need for extensive parameter optimization based on domain experience. To simulate real-world conditions that automotive coatings face, such as those involving de-icing salts, an accelerated wet-dry cycling protocol was implemented. This protocol was designed based on snow-depth data collected from the Rust Belt states, spanning from 2015 to the winter of 2021, obtained from the Global Historical Climatology Network Daily (GHCNd) database. The cycling protocol involved 96 hours of wet conditions followed by 73 hours of dry conditions within a Cyclic Corrosion Test (CCT) chamber, using a 5% NaCl solution at 30 degrees C. Four distinct types of coating samples were prepared, each with varying corrosion control mechanisms and scratch depths, to evaluate their performance under these accelerated conditions. Real-time electrochemical assessments were conducted on these samples, and the data was analyzed over a 30-day period. The phase angle plot analysis revealed significant insights into the corrosion behavior of different coatings. Samples with intact coatings exhibited a gradual onset of corrosion, resembling steady-state conditions, indicating effective corrosion protection. Partially scratched samples showed a delayed activation of corrosion mechanisms, emphasizing the protective capabilities of sacrificial coatings. Conversely, samples with complete scratch depth displayed immediate activation of corrosion processes, highlighting the absence of a protective barrier and the vulnerability of the underlying substrate. This comprehensive approach, combining real-time EIS testing, ANN modeling, and accelerated corrosion protocols, provides a robust experimental-theoretical framework for predicting and understanding coating degradation and corrosion mechanisms. The integration of mechanistic and machine-learning concepts allows for the development of more accurate and reliable predictions of coating performance over extended periods. The findings from this study contribute significantly to the advancement of corrosion protection strategies in automotive and industrial applications. By understanding the intricate dynamics of coating degradation, more effective and durable protective coatings can be designed, ultimately enhancing the longevity and reliability of metallic substrates in harsh environmental conditions."],"dc:format.mimetype":["application/pdf"],"dc:identifier.uri":["https://hdl.handle.net/1969.1/1599914"],"dc:language.iso":["English"],"dc:subject":["Engineering, Materials Science"],"dc:title":["Predictive Analysis of Corrosion Resistance in Automotive Coatings Using Artificial Neural Networks"],"dc:type":["Thesis"],"thesis:degree_discipline":["Materials Science and Engineering"],"thesis:degree_level":["Doctoral"],"thesis:degree_name":["Doctor of Philosophy"],"thesis:institution_name":["Texas A&M University"]},"updated_at":"2026-08-21T16:48:40Z"}