A Markov chain Monte Carlo analysis of carbon-phenolic ablation through a lumped physical model

Abstract

Payloads and other rocket components use Thermal Protection Systems (TPS) to avoid the effects of high heat fluxes generated by aerodynamic heating or propellant burning. The most used type of TPS is the ablative thermal shield, employing polymeric composite ablators. Ablative and thermophysical properties of such materials must be known for TPS design. These properties are often difficult to access experimentally, and vary according to the manufacturing process, ratio between reinforcement and matrix and the fiber direction in the case of composites, and other process variables. Inverse solution methods were initially applied in computational models to solve the heat transfer problems in space vehicles. This work proposes an inverse methodology, using the Markov chain Monte Carlo (MCMC) method, to determine the ablative properties of carbon-phenolic materials manufactured by biased tape wrapping, starting from arc-jet ablation results of samples of this material. A simplified but robust lumped physical model is used to represent the direct problem. The obtained values of these properties were used in the direct problem in a more refined mesh, and the simulation was compared to the experiments. Good agreement between numerical and experimental data was obtained, allowing the use of this methodology in estimation of ablation properties.