Material response model with artificial neural networks for ablation analysis of lightweight silicone-modified phenolic matrix nanocomposites

Abstract

Polymer matrix thermal protection materials are developing towards lightweight, better thermal insulation and oxidation resistance to satisfy the requirements of planetary entry vehicles, which have a strong gas-solid interaction in the aerodynamic heating environment. A material response model with an artificial neural network (ANN) is developed to study the thermochemical responses of the lightweight silicone-modified phenolic matrix nanocomposites. The ANN is designed to determine the real-time dimensionless char blowing rate and wall enthalpy for the material response model. A constant offset of the dimensionless char blowing rate is added to account for the contribution of multiple constituents of the material surface to the ablation wall. The material response model is validated against three plasma heating test cases in terms of surface temperature, in-depth temperature, and recession. A three-step ablation mechanism is revealed through molecular dynamics simulations, comprising thermal decomposition of the polymer, phase separation/rearrangement, and an enhancement in crystallinity. The effect of ambient pressure on the ablation response is further investigated and found that the lower pressures can lead to higher surface material consumption and higher recession rate due to the reduced blowdown flux of pyrolysis gases.