Nonlinear identification via connected neural networks for unsteady aerodynamic analysis

Abstract In the present work, a nonlinear system identification strategy is proposed which is based on the series connection of a recurrent local linear neuro-fuzzy model (NFM) and a multilayer perceptron (MLP) neural network. The NFM with output feedback is initially used for multi-step ahead predictions, whereas the MLP neural network is a posteriori employed to perform a nonlinear quasi-static correction of the NFM's time-series response. The novel identification approach is utilized exemplarily as a reduced-order modeling (ROM) technique to lower the computational effort of unsteady aerodynamic simulations, although the approach is generally applicable to any nonlinear identification task. In order to demonstrate the method's fidelity for unsteady aerodynamic modeling, the NLR 7301 airfoil is investigated at transonic flow conditions, while the motion-induced aerodynamic forces are considered in particular. Therefore, the pitch and plunge degrees of freedom are simultaneously excited via forced motions to obtain the training data for model calibration, while the respective aerodynamic response is computed using a computational fluid dynamics (CFD) solver. The sequential nonlinear identification process as well as the generalization of the resulting model is presented. Besides, a Monte-Carlo-based training procedure, which is novel in the context of aerodynamic reduced-order modeling, is introduced to estimate statistical errors. It is shown that the essential linear and nonlinear system characteristics are accurately reproduced by the new approach compared to the full-order solution. Moreover, by examining the results in comparison to established ROM methods it is indicated that the connected neural network approach leads to an enhanced simulation and generalization performance.

[1]  S. Billings.  Nonlinear System Identification: NARMAX Methods in the Time, Frequency, and Spatio-Temporal Domains , 2013 .

[2]  Weiwei Zhang,et al.  Nonlinear Aerodynamic Reduced-Order Model for Limit-Cycle Oscillation and Flutter , 2016 .

[3]  Christian Breitsamter,et al.  Efficient unsteady aerodynamic loads prediction based on nonlinear system identification and proper orthogonal decomposition , 2016 .

[4]  W. Silva,et al.  Discrete-time linear and nonlinear aerodynamic impulse responses for efficient cfd analyses , 1997 .

[5]  Robert E. Bartels,et al.  Numerical investigation of transonic limit cycle oscillations of a two-dimensional supercritical wing☆ , 2003 .

[6]  Minglang Yin,et al.  Novel Wiener models with a time-delayed nonlinear block and their identification , 2016 .

[7]  Boris Laschka,et al.  Small Disturbance Navier-Stokes Computations for Low-Aspect-Ratio Wing Pitching Oscillations , 2010 .

[8]  Matthias Haupt,et al.  Efficient Surrogate Modelling of Nonlinear Aerodynamics in Aerostructural Coupling Schemes , 2014 .

[9]  Christian Breitsamter,et al.  Aeroelastic Prediction of Discrete Gust Loads Using Nonlinear and Time-Linearized CFD-Methods , 2015 .

[10]  O. Bendiksen.  Review of unsteady transonic aerodynamics: Theory and applications , 2011 .

[11]  A. Mannarino,et al.  Nonlinear aeroelastic reduced order modeling by recurrent neural networks , 2014 .

[12]  Boris Laschka,et al.  Small Disturbance Euler Equations: Efficient and Accurate Tool for Unsteady Load Prediction , 2000 .

[13]  Walter A. Silva,et al.  Development of Reduced-Order Models for Aeroelastic Analysis and Flutter Prediction Using the CFL3Dv6.0 Code , 2002 .

[14]  L. Sirovich.  Turbulence and the dynamics of coherent structures. I. Coherent structures , 1987 .

[15]  O. Nelles.  Nonlinear System Identification: From Classical Approaches to Neural Networks and Fuzzy Models , 2000 .

[16]  Domenico Quagliarella,et al.  Proper Orthogonal Decomposition, surrogate modelling and evolutionary optimization in aerodynamic design , 2013 .

[17]  Simon Haykin,et al.  Neural Networks: A Comprehensive Foundation , 1998 .

[18]  Kumpati S. Narendra,et al.  Identification and control of dynamical systems using neural networks , 1990, IEEE Trans. Neural Networks.

[19]  Weiwei Zhang,et al.  Efficient Method for Limit Cycle Flutter Analysis Based on Nonlinear Aerodynamic Reduced-Order Models , 2012 .

[20]  P. Beran,et al.  Reduced-order modeling: new approaches for computational physics , 2004 .

[21]  Daniella E. Raveh,et al.  Identification of computational-fluid-dynamics based unsteady aerodynamic models for aeroelastic analysis , 2004 .

[22]  Daniella E. Raveh,et al.  Reduced-Order Models for Nonlinear Unsteady Aerodynamics , 2001 .

[23]  David J. Lucia,et al.  Reduced order modeling of a two-dimensional flow with moving shocks , 2003 .

[24]  Jonathan E. Cooper,et al.  Introduction to Aircraft Aeroelasticity and Loads , 2007 .

[25]  Max F. Platzer,et al.  Transonic flutter computations for the NLR 7301 supercritical airfoil , 2001 .

[26]  Christian Breitsamter,et al.  Efficient Computation of Unsteady Aerodynamic Loads Using Computational-Fluid-Dynamics Linearized Methods , 2013 .

[28]  R. Dwight,et al.  Energy budget analysis of aeroelastic limit-cycle oscillations , 2017 .

[29]  Weiwei Zhang,et al.  Layered reduced-order models for nonlinear aerodynamics and aeroelasticity , 2017 .

[30]  Holger Mai,et al.  Experiments on Heave / Pitch Limit-Cycle Oscillations of a Supercritical Airfoil , 2004 .

[31]  Wolfgang Polifke,et al.  Black-box system identification for reduced order model construction , 2014 .

[32]  Lennart Ljung,et al.  System Identification: Theory for the User , 1987 .

[33]  Edoardo Patelli,et al.  Robust artificial neural network for reliability and sensitivity analyses of complex non-linear systems , 2017, Neural Networks.

[34]  Bryan Glaz,et al.  Reduced-Order Nonlinear Unsteady Aerodynamic Modeling Using a Surrogate-Based Recurrence Framework , 2010 .

[35]  William E. Faller,et al.  Neural network prediction and control of three-dimensional unsteady separated flowfields , 1995 .

[36]  Daniel D. McCracken,et al.  The Monte Carlo Method , 1955 .

[37]  Earl H. Dowell,et al.  Modeling of Fluid-Structure Interaction , 2001 .

[38]  Christian Breitsamter,et al.  Neurofuzzy-Model-Based Unsteady Aerodynamic Computations Across Varying Freestream Conditions , 2016 .

[39]  M. Winter,et al.  Reduced-Order Modeling of Unsteady Aerodynamic Loads using Radial Basis Function Neural Networks , 2014 .

[40]  Jer-Nan Juang,et al.  An eigensystem realization algorithm for modal parameter identification and model reduction. [control systems design for large space structures] , 1985 .

[41]  D. Krige.  A statistical approach to some basic mine valuation problems on the Witwatersrand, by D.G. Krige, published in the Journal, December 1951 : introduction by the author , 1951 .

[42]  Jeffrey P. Thomas,et al.  Proper Orthogonal Decomposition Technique for Transonic Unsteady Aerodynamic Flows , 2000 .

[43]  D. Marquardt.  An Algorithm for Least-Squares Estimation of Nonlinear Parameters , 1963 .

[44]  Weiwei Zhang,et al.  Reduced-Order-Model-Based Flutter Analysis at High Angle of Attack , 2007 .

[45]  Christian Breitsamter,et al.  Coupling of Recurrent and Static Neural Network Approaches for Improved Multi-step Ahead Time Series Prediction , 2018 .

[46]  D. Broomhead,et al.  Radial Basis Functions, Multi-Variable Functional Interpolation and Adaptive Networks , 1988 .