Application of discrete adjoint method to sensitivity and uncertainty analysis in steady-state two-phase flow simulations

TL;DR

This paper develops a discrete adjoint sensitivity analysis framework for steady-state two-phase flow simulations, demonstrating its efficiency and accuracy in uncertainty propagation compared to traditional methods.

Contribution

It introduces a novel discrete adjoint method and implicit solver for two-phase flow, enabling efficient sensitivity and uncertainty analysis.

Findings

Adjoint sensitivities match analytical results in faucet flow.

The method accurately propagates uncertainty in the BFBT benchmark.

Efficiency gains over Monte Carlo methods are demonstrated.

Abstract

Verification, validation and uncertainty quantification (VVUQ) have become a common practice in thermal-hydraulics analysis. An important step in the uncertainty analysis is the sensitivity analysis of various uncertain input parameters. The common approach for computing the sensitivities, e.g. variance-based and regression-based methods, requires solving the governing equation multiple times, which is expensive in terms of computational cost. An alternative approach to compute the sensitivities is the adjoint method. The cost of solving an adjoint equation is comparable to the cost of solving the governing equation. Once the adjoint solution is available, the sensitivities to any number of parameters can be obtained with little cost. However, successful application of adjoint sensitivity analysis to two-phase flow simulations is rare. In this work, an adjoint sensitivity analysis framework is developed based on the discrete adjoint method and a new implicit forward solver. The framework is tested with the faucet flow problem and the BFBT benchmark. Adjoint sensitivities are shown to match analytical sensitivities very well in the faucet flow problem. The adjoint method is used to propagate uncertainty in input parameters to the void fraction in the BFBT benchmark test. The uncertainty propagation with the adjoint method is verified with the Monte Carlo method and is shown to be efficient.