An efficient integrated aerothermoelasticity analysis system based on surrogate-based reduced order modeling for hypersonic vehicles

Aerothermoelasticity is one of the key technologies for hypersonic vehicles. Accurate and efficient computation of the aerothermodynamics is one of the primary challenges for hypersonic aerothermoelastic analysis. In order to improve the efficiency of computationally expensive models involving high fidelity models, surrogate-based reduced order models (ROMs) have been frequently employed to replace these high-fidelity aerothermodynamic simulations. Sampling-point approach and adding-point strategy are two key techniques that influenced the precision and efficiency of surrogated-based ROMs. Aiming at improving the accuracy of ROMs for aerothermodynamics, an enhanced algorithm of fast maximin Latin hypercube design (ESLE) is presented, which proves to be helpful to improve the precisions of ROMs by the same sampling points. In order to improve the efficiency of constructing ROMs for aerothermodynamics, an adding-point strategy based on fuzzy clustering (APSFC) is proposed. Furthermore, an integrated aerothermoelasticity analysis system is developed using aerothermodynamics ROMs, which consists of aerodynamic heating analysis, transient heat transfer analysis, thermal mode analysis and thermal flutter analysis. Test results for the three-dimensional aerothermodynamic over a hypersonic surface indicate that the proposed ESLE and APSFC algorithms are helpful to improve the efficiency and precision of the ROMs. The flutter velocity by ROMs in the integrated aero-thermal-structure analysis system will decline by nearly 10% compared with that by approximated engineering methods. In a word, the integrated aerothermoelasticity analysis system has a notable efficiency compared with the full CFD and acceptable higher precision than approximated engineering methods.