Adjoint-Based Aeroelastic Design Optimization Using a Harmonic Balance Method

Abstract

Turbomachinery blades characterized by highly-loaded, slender profiles and operating under unsteady flow may suffer from aeroelastic shortcomings, like forced response and flutter. One of the ways to mitigate these aeroelastic effects is to redesign the blade profiles, so as to increase aero-damping and decrease aero-forcing. Design optimization based on high-fidelity aeroelastic analysis methods is a formidable task due to the inherent computational cost. This work presents an adjoint-based aeroelastic shape-optimization framework based on reduced order methods for flow analysis and forced response computation. The flow analysis is carried out through a multi-frequency fully-turbulent harmonic balance method, while the forced response is computed by means of the energy method. The capability of the design framework is demonstrated by optimizing two candidate cascades, namely, i) a transonic compressor cascade and, ii) a supersonic impulse turbine rotor operating with toluene as working fluid, initially designed by means of the method of waves. The outcomes of the optimization show significant improvements in terms of forced-response in both cases as a consequence of aero-damping enhancement.

Non-iterative generation of an optimal mesh for a blade passage using deep reinforcement learning

“…The ranges of each element in BSP are set wide enough to include blades in the previous literature with different shapes and applications [35][36][37][38] as follows:…”

Transient forward harmonic adjoint sensitivity analysis

“…Adjoint sensitivity analysis is the method of choice for a setting with many design parameters [5,10]. Transient adjoint sensitivity analysis [10,11] as well as sensitivity analysis based on HB solvers [12,13] have been published and used before. In more recent publications, transient adjoint sensitivity analyses have been applied for a larger variety of nonlinear problems such as electromagnetics, electroquasistatics or nonlinear damped mechanical systems [14][15][16].…”

Shape and Layout Optimization of Bare-Tube Heat Exchangers Using the Adjoint Method and CAD-based Parametrization

“…In this context, the aim of the research presented in this manuscript is to apply an adjoint-based CFD optimization framework to optimize the shape and layout of bare-tube heat exchangers simultaneously. Such an optimization framework was previously developed for turbomachinery problems [12][13] and was recently extended to design tube bundles [14]. This work extends this framework further by including the tube longitudinal pitch among the optimization variables.…”