A dynamic data-driven response prediction method for thermal protection tiles and experimental validation

Abstract

Intelligent thermal protection systems with self-perception and decision-making functions are preferable for future spacecraft. Online thermal response prediction is a key step to intelligent thermal protection systems. In this paper, the concept of Dynamic Data-Driven Application System between simulations and experiments is employed to achieve online thermal response prediction. In this framework, measured temperature data are injected into an online heat transfer model, and the real-time temperature at the bottom of the thermal protection system is calculated. A demonstration test system is built and experiments were carried out to validate the method. The effects of dynamic data and the number of sensors on the temperature prediction accuracy are analyzed. Dynamic data-driven temperature predictions under transient heat load are also carried out. The dynamic data-driven model can reduce the temperature prediction error from 25.9% to less than 10% based on the experimental results, and increasing the number of sensors can also improve the prediction accuracy. The model also has good adaptability under accidental overload and cyclic load. The Dynamic Data-Driven Application System framework presented in this work can be used for online safety assessment of thermal protection systems and active flight control.