Risk Assessment of Scramjet Unstart Using Adjoint-Based Sampling Methods

Abstract

We demonstrate an adjoint based approach for accelerating Monte Carlo estimation of risk, and apply it to estimating the probability of unstart in a SCRamjet engine under uncertain conditions that are characterized by various Gaussian and non-Gaussian distributions. The adjoint equation is solved with respect to an objective function that is used to identify unstart and the adjoint solution is used to generate a linear approximation to the objective function. This linear surrogate is used to divide the uncertain input parameters into three different strata, corresponding to safe operation of the engine, uncertain operation and unstart. The probability of unstart within these strata is very different and as a result, stratified sampling significantly increases the efficiency of the risk assessment procedure by reducing the variance of the estimator. Using this technique, the Monte Carlo method was demonstrated to be accelerated by a factor of 5.4.

Dispersion Reduction for a Sounding Rocket Scramjet Flight Experiment

“…For a predetermined fuel mass-flow rate, if the air mass capture is greater than expected, the resulting equivalence ratio will be lower than expected and a lean blowout can occur [7]. Conversely, if the air mass capture is lower than expected, the resulting equivalence ratio will be higher than expected and engine unstart can occur [8]. It is also important that the freestream test conditions are within those able to be simulated by the ground test facilities, so that meaningful conclusions regarding the effects of the facility can be made from the resulting data sets.…”

Aerodynamic Risk Assessment using Parametric, Three-Dimensional Unstructured, High-Fidelity CFD and Adaptive Sampling

“…[1] Extra sampling points were allocated in the uncertain region between the safe operation and unstart of the engine. [2] When performing an optimization under uncertainty, introducing a surrogate model can greatly reduce the computational cost, but the results are less accurate than the original model. [3,4] Rather than using the approximations from surrogate models to directly replace real simulations, such approximations can be used to guide the allocation of real simulations.…”

Uncertainty quantification in LES of channel flow

“…Optimization‐based calibration methods have also been applied to RANS models . Other recent efforts have focused on using adjoint methods to assess model sensitivities . A review article by Najm provides a discussion on uncertainty quantification techniques, with emphasis on PCEs, in computational fluid dynamics.…”