Unsteady aerodynamic modeling based on fuzzy scalar radial basis function neural networks

Wang, Xu; Kou, Jiaqing; Zhang, Weiwei

doi:10.1177/0954410019836906

Cited by 27 publications

(9 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…9-11 With the increase of data volume and data complexity, machine learning methods are widely used in regression analysis, such as radial basis function neural networks (RBF), random forest regression, decision tree regression, and AdaBoost regression. 12-15 Because the performance degradation is mainly obtained by random distribution fitting based on engineering experience, those regression analysis models which take sample label error as loss function are not optimal for performance degradation prediction of the aeroengine. Therefore, this study proposes a regression analysis method of performance degradation based on the support vector regression (SVR) model, which uses two relaxation variables to control the sample isolation band and takes the band width and total loss as optimization target.…”

Section: Introductionmentioning

confidence: 99%

Performance degradation prediction of aeroengine based on attention model and support vector regression

Che

Wang

et al. 2021

Proceedings of the Institution of Mechanical Engineers, Part G:

View full text Add to dashboard Cite

Accurate performance degradation prediction of aeroengines can ensure the safety and reliability of the aircraft. Based on the mass long time series data of multiple state parameters, a novel performance degradation prediction method based on attention model (AM) and support vector regression (SVR) is proposed in this article. The AM uses the attention mechanism between encoder and decoder to realize weight distribution of different source samples, so as to realize time series prediction of state parameters. The SVR model is used to mine the mapping relationship between multiple state parameters and performance degradation. The performance degradation prediction results can be achieved by putting the time series prediction results of multiple state parameters into the SVR model. The turbofan engine degradation simulation dataset carried out using commercial modular aero-propulsion system simulation (C-MAPSS) is used to verify the effectiveness of the proposed method. The results demonstrate that it can get accurate time series prediction and performance degradation analysis results. Compared with other methods, the proposed attention model and support vector regression (AM-SVR) model has lower prediction error and higher stability when dealing with noised samples.

show abstract

Section: Introductionmentioning

confidence: 99%

Performance degradation prediction of aeroengine based on attention model and support vector regression

Che

Wang

et al. 2021

Proceedings of the Institution of Mechanical Engineers, Part G:

View full text Add to dashboard Cite

show abstract

“…Furthermore, Zhang et al [13] applied a multi-kernel RBF neural network for modeling unsteady aerodynamics considering different flow conditions. In addition, ROMs based on fuzzy logic [14][15][16] yield accurate and reliable results for nonlinear system identification purposes.…”

Section: Introductionmentioning

confidence: 99%

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

Zahn

Breitsamter

2021

CEAS Aeronaut J

View full text Add to dashboard Cite

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.

show abstract

“…17,18 However, this approach is timeconsuming and requires much computation cost even to perform only a single solution of wing transonic flutter. To overcome the disadvantage of the coupled CFD/CSD time marching method, ROM approaches, [19][20][21] based on CFD technique are developed to calculate the wing flutter speed with the assumption of dynamically linear aerodynamics. 16 Amongst these approaches, the system identification method 19,22 is a robust and effective technique to build the transonic aerodynamic ROM, which is used in the present study.…”

Section: Introductionmentioning

confidence: 99%

Transonic flutter characteristics of an airfoil with morphing devices

Guo

2020

Proceedings of the Institution of Mechanical Engineers, Part G:

View full text Add to dashboard Cite

An investigation into transonic flutter characteristic of an airfoil conceived with the morphing leading and trailing edges has been carried out. Computational fluid dynamics (CFD) is used to calculate the unsteady aerodynamic force in transonic flow. An aerodynamic reduced order model (ROM) based on autoregressive model with exogenous input (ARX) is used in the numerical simulation. The flutter solution is determined by eigenvalue analysis at specific Mach number. The approach is validated by comparing the transonic flutter characteristics of the Isogai wing with relevant literatures before applied to a morphing airfoil. The study reveals that by employing the morphing trailing edge, the shock wave forms and shifts to the trailing edge at a lower Mach number, and aerodynamic force stabilization happens earlier. Meanwhile, the minimum flutter speed increases and transonic dip occurs at a lower Mach number. It is also noted that leading edge morphing has negligible effect on the appearance of the shock wave and transonic flutter. The mechanism of improving the transonic flutter characteristics by morphing technology is discussed by correlating shock wave location on airfoil surface, unsteady aerodynamics with flutter solution.

show abstract

Unsteady aerodynamic modeling based on fuzzy scalar radial basis function neural networks

Cited by 27 publications

References 47 publications

Performance degradation prediction of aeroengine based on attention model and support vector regression

Performance degradation prediction of aeroengine based on attention model and support vector regression

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

Transonic flutter characteristics of an airfoil with morphing devices

Contact Info

Product

Resources

About