Statistical Design of Thermal Protection System Using Physics-Informed Neural Network
ASCE-ASME Journal of Risk and Uncertainty in Engineering Systems, Part B: Mechanical Engineering · 01.12.2026 · 12 (4)
DOI: 10.1115/1.4072775 ↗Kurz gefasst
Fachartikel in der Zeitschrift ASCE-ASME Journal of Risk and Uncertainty in Engineering Systems, Part B: Mechanical Engineering, online erschienen am 11.11.2026. Verfasst von Karthik Reddy Lyathakula und 4 weiteren Autor:innen. Die Arbeit nennt 40 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Thermal protection systems (TPS) of space vehicles are designed computationally rather than experimentally. They are validated using ground experiments, but all aspects of the flight cannot be replicated on ground. This ground-to-flight mapping introduces uncertainties that need to be accounted for while designing any thermal protection system. Thus, precise computational models along with uncertainty quantification in the models are required to design the TPS. The focus of this study is to estimate the thermal material parameters of TPS based on the target reliability requirements using statistical methods. To perform uncertainty quantification (UQ) of a system, a simulated model of the system needs to be solved many times on statistical samples, increasing the computational time and cost. In this study, a physics-informed neural network (PINN) model is used instead of traditional physics-based numerical solutions. The accuracy of PINN is comparable to that of the numerical solution. To find the parameter distribution, sampling of the parameter space is performed using sequential Monte Carlo (SMC) method. Combining the use of both PINN predictive model and SMC sampling, the framework can approximate the parameter distributions that satisfy the TPS design reliability constraints. The framework achieved a remarkable increase in the speed of performing the reliability analysis; specifically, the time required to process 10,000 samples was reduced 175 times using the proposed PINN-SMC approach. This reliability analysis can be used for design optimization of the TPS based on risk analysis along with other systems of vehicle.
Zitieren
APA
Lyathakula, K. R., Aseem, M., Goswami, B. R. D., Cesmeci, S., & Burragallae, D. (2026). Statistical Design of Thermal Protection System Using Physics-Informed Neural Network. ASCE-ASME Journal of Risk and Uncertainty in Engineering Systems, Part B: Mechanical Engineering, 12(4). https://doi.org/10.1115/1.4072775
Vancouver
Lyathakula KR, Aseem M, Goswami BRD, Cesmeci S, Burragallae D. Statistical Design of Thermal Protection System Using Physics-Informed Neural Network. ASCE-ASME Journal of Risk and Uncertainty in Engineering Systems, Part B: Mechanical Engineering. 2026;12(4). doi:10.1115/1.4072775
BibTeX
@article{Lyathakula2026,
author = {Lyathakula, Karthik Reddy and Aseem, Muhammad and Goswami, Basab Ranjan Das and Cesmeci, Sevki and Burragallae, Dhanraj},
title = {Statistical Design of Thermal Protection System Using Physics-Informed Neural Network},
journal = {ASCE-ASME Journal of Risk and Uncertainty in Engineering Systems, Part B: Mechanical Engineering},
year = {2026},
volume = {12},
number = {4},
doi = {10.1115/1.4072775}
}
EUVIMED