Research On Probabilistic Analysis Method Of Thermal Protection System For Hypersonic Vehicle

Thermal protection system(TPS)is one of the most important systems of hypersonic vehicle,which can guarantee the internal temperature condition and the system's structural integrity in high temperature environment.Due to the uncertain factors in the manufacturing process,TPS components differ from each other universally in terms of size,material properties and so on.This difference has an impact on the reliability and overall performance of TPS.In order to determine the influence of parameter uncertainty on the reliability and overall performance of TPS,this paper makes the probabilistic analysis of the rigid ceramic tile TPS,which is widely used in hypersonic vehicle.Firstly,the parametric model of rigid ceramic tile TPS is established.Based on the model,the sensitivity analysis of rigid ceramic tile TPS under short-time high heat transfer condition is carried out.Thus the influence of parameters on thermal protection properties is ensured,so as to grasp the key point,reduce costs and improve efficiency in the design and manufacture of TPS.Secondly,this paper combined the finite element method and Monte Carlo simulation and established the probabilistic thermal analysis system of TPS.In consideration of the uncertainty of material parameter and the thickness of each thermal protection layer,the probability distribution of system temperature field is obtained by using the response surface method and the other probability analysis methods.The evaluation of the thermal reliability for TPS is also done in this paper.On this basis,a new probabilistic design method for rigid ceramic tile TPS is proposed.Numerical examples show that the new probabilistic design method can determine the minimum thickness of TPS on the premise of ensuring the reliability,so it is helpful to reduce the conservativeness of the design.For the design of TPS,it is of guiding significance to reasonably determine the thickness of the ceramic tile,which can effectively reduce the weight under the premise of ensuring the system performance.Finally,this paper proposed the precise time integration(PTI)method and Krylov subspace method which integrated with the dimensional expansion method.These two methods are used to solve transient heat transfer problems instead of conventional direct integration method.By using the dimensional expansion method,the transient heat transfer equation is transformed into its equivalent homogeneous form,thus avoiding the matrix inversion.On this basis,the PTI algorithm and the Krylov subspace method are used to solve the homogeneous equations,which save the computing time and improve the calculation accuracy.It provides the possibility for the calculation of a large number of probability samples.A finite element model of rigid ceramic tile TPS is used to validate the the feasibility of the proposed algorithms.The calculation results are compared with the results of commercial software ANSYS and the traditional direct integration method.It has been demonstrated that the proposed dimensional expanding precise time integration(DE-PTI)method is more accurate and insensitive to integration step length,and the dimensional expanding Krylov subspace method is more suitable for large-scale finite element models.