Invited Speaker

Prof. Liang Wang (Research Fellow)

Shanghai Institute of Ceramics, Chinese Academy of Sciences 

E-mail: L.Wang@mail.sic.ac.cn

Title: AI-assisted Visualization of the Service-induced Damage Process in Thermal Barrier Coatings

Profile:

Dr. Wang Liang, born in 1982 in Qichun, Hubei Province, China, is a Professor, Ph.D. Supervisor, and Postdoctoral Mentor. He specializes in the composition-structural design, performance optimization, and service performance prediction of high-temperature thermal protection coatings, employing macro-scale finite element simulations, artificial intelligence (traditional machine learning/deep learning), and experimental characterization. Recognized for his contributions, he has been honored as a Member of the Youth Innovation Promotion Association of the Chinese Academy of Sciences (CAS), an awardee of the inaugural "Outstanding Youth Support Program" at the Shanghai Institute of Ceramics (CAS), and a Distinguished Research Fellow of CAS. He has led or completed over 20 national and industry-funded projects, including the HY Action Plan, GF Basic Research Projects, GF Basic Strengthening Sub-Projects, the National Natural Science Foundation of China (NSFC) Major Research Plan Cultivation Project, NSFC-NSAF Joint Fund, two NSFC General Programs, NSFC Youth Fund, as well as technical collaborations with Sinopec Northwest Oilfield Branch, CSIC Longjiang Guanghan Gas Turbine Co., Ltd., and Tianmushan Laboratory. Dr. Wang has authored over 70 SCI-indexed papers in core journals, with more than 50 as first or corresponding author, including one highly cited paper exceeding 200 citations. He holds 9 Chinese invention patents as the first inventor and has received multiple awards, such as the Second Prize in the Basic Research Category of the Rare Earth Science and Technology Award (China Society of Rare Earths, ranked 2nd). He actively contributes to academia as a Member of ASM International, Youth Editorial Board Member of the Journal of Inorganic Materials, Editorial Board Member of Materials Protection, Committee Member of the Thermal Protection Materials Committee (China Society of Rare Earths), and Academic Committee Member of the Key Laboratory for Efficient Utilization of Non-metallic Mineral Resources in Southern Henan.

Abstract:

Thermal barrier coatings (TBCs) inevitably face failure during practical service, and predicting their failure modes and service lifespan remains a common challenge in the scientific community. To address this, we have developed an AI-driven, dual-modal ("ear-listening" and "eye-seeing") visualization framework for the full-process monitoring of TBCs’ service-induced damage. By dynamically analyzing in-situ acoustic emission (AE) signals from TBCs under mechanical loading and actual service conditions—including filtering, fast Fourier transform (FFT), wavelet (packet) analysis, K-Means unsupervised machine learning clustering, wavelet energy coefficient calculation, and neural network analysis (BP and RBF neural networks)—we designed a software system that establishes real-time correlations between internal crack propagation modes and AE signal features, enabling rapid "ear-listening" identification of coating failure modes. Additionally, leveraging batch and automated labeling of microstructural features in cross-sectional images of as-prepared and early-service TBCs, combined with deep learning and long short-term memory (LSTM) neural networks, we achieved dynamic prediction of microstructural evolution, particularly tracking the crack propagation patterns over extended service durations. This integrated approach realizes AI-powered dual-modal (auditory and visual) visualization of TBCs’ failure behavior, bridging the gap between real-time damage monitoring and predictive lifespan assessment.