Airfoil buffet aerodynamics at plunge and pitch excitation based on long short-term memory neural network prediction

R Zahn,C Breitsamter

doi:10.1007/s13272-021-00550-6

Abstract

In the present study, a nonlinear system identification approach based on a long short-term memory (LSTM) neural network is applied for the prediction of transonic buffet aerodynamics. The identification approach is applied as a reduced-order modeling (ROM) technique for an efficient computation of time-varying integral quantities such as aerodynamic force and moment coefficients. Therefore, the nonlinear identification procedure as well as the generalization of the ROM are presented. The training data set for the LSTM–ROM is provided by performing forced-motion unsteady Reynolds-averaged Navier–Stokes simulations. Subsequent to the training process, the ROM is applied for the computation of the aerodynamic integral quantities associated with transonic buffet. The performance of the trained ROM is demonstrated by computing the aerodynamic loads of the NACA0012 airfoil investigated at transonic freestream conditions. In contrast to previous studies considering only a pitching excitation, both the pitch and plunge degrees of freedom of the airfoil are individually and simultaneously excited by means of an user-defined training signal. Therefore, strong nonlinear effects are considered for the training of the ROM. By comparing the results with a full-order computational fluid dynamics solution, a good prediction capability of the presented ROM method is indicated. However, compared to the results of previous studies including only the airfoil pitching excitation, a slightly reduced prediction performance is shown.

Highlights

Introduction and motivationUnsteady dynamic aeroelastic and aerodynamic phenomena, such as flutter and buffet, are of paramount importance concerning safety and efficiency requirements of passenger aircraft
recurrent neural networks (RNNs) are defined as a class of neural networks, characterized by internal self-connections, which allow for sequential information processing
To investigate the prediction capability of the long short-term memory (LSTM)–reduced-order model (ROM), the training procedure introduced in Sect. 2.3 is applied for the computation of the lift ( CL ) and pitching moment coefficient ( CMy ) of the NACA0012 airfoil

Summary

Introduction

Introduction and motivationUnsteady dynamic aeroelastic and aerodynamic phenomena, such as flutter and buffet, are of paramount importance concerning safety and efficiency requirements of passenger aircraft. Referred to as shock buffet, is characterized by shock–boundary layer interaction, resulting in self-sustained cycles of shock movement and partial flow separation. A comprehensive overview of computational and experimental studies of transonic buffet is given by Giannelis et al [1]. Due to the self-sustained flow characteristics, the resulting aerodynamic forces and moments are characterized by unsteady variations. The applied computational method must account for viscous effects to represent boundary layer interaction and flow separation. To resolve these relevant mechanisms, the effort of computational methods is still time- and cost-consuming, resulting in a high demand for fast and accurate surrogate methods to decrease computational time and costs. Considering aerospace applications in particular, a short overview of linear and nonlinear ROM approaches is given in

Methods

Results

Conclusion