MODELING OF STORE SEPARATION BEHAVIOR BASED ON A NEURAL NETWORK AND UNSTEADY FLOW SOLUTIONS

Date

2015-09-10

Author

Erinç, Erdoğan
Tuncer, İsmail Hakkı

Metadata

Show full item record

Item Usage Stats

147
views

0
downloads

In this study a neural network based method is developed for the prediction of separation characteristics of external store weapons carried under aircraft wings. The method is based on an artificial neural network trained by high fidelity unsteady flow solutions. The unsteady flow solutions as the store separates from the carriage and the resulting six degrees of freedom motion of the store are computed conditions by a commercial flow solver for various flight conditions. The trajectory of the store and the unsteady aerodynamic loads acting on it are used to train a neural network. A simulation is used for evaluating the 6 degrees of freedom motion of a store based on the aerodynamic loads predicted by the neural network as a function of the flight conditions and the instantaneous position of the store. The efficiency and the accuracy of the method developed are determined by comparing the predicted and computed trajectories of the store for flight conditions not used in the training set. It is shown that the method developed can successfully predict the store separation characteristics.

URI

http://aiac.ae.metu.edu.tr
https://hdl.handle.net/11511/81124
https://www.researchgate.net/publication/333972151_MODELING_OF_STORE_SEPARATION_BEHAVIOR_BASED_ON_A_NEURAL_NETWORK_AND_UNSTEADY_FLOW_SOLUTIONS

Conference Name

Ankara International Aerospace Conference, 10 - 12 Eylül 2015

Collections

Department of Aerospace Engineering, Conference / Seminar

Suggestions

OpenMETU
Core

Aerodynamic Design Analysis Of Missiles With Strake Configurations At Supersonic Mach Numbers Usta, Engin; Kıvanç, Arslan; Tuncer, İsmail Hakkı (2015-09-10) In this study a neural network based method is developed for the prediction of separation characteristics of external store weapons carried under aircraft wings. The method is based on an artificial neural network trained by high fidelity unsteady flow solutions. The unsteady flow solutions as the store separates from the carriage and the resulting six degrees of freedom motion of the store are computed conditions by a commercial flow solver for various flight conditions. The trajectory of the store and the...
ESTIMATION OF AERODYNAMIC LOADS OF A PROPELLER THROUGH IMPROVED BLADE ELEMENT AND MOMENTUM THEORY AND PROPELLER DESIGN OPTIMIZATION KAYA, DERYA; Kutay, Ali Türker; Department of Aerospace Engineering (2021-6) This study focuses on accurate prediction of total forces and moments acting on a propeller in all flight conditions through Blade Element and Momentum Theory (BEMT) and design optimization of a UAV propeller. Under various flight conditions such as hover, vertical climb, and forward flight, propeller generates different aerodynamic loads in different free-stream velocities, propeller disk angles of attack, and propeller’s angular speeds. For this reason, it is important to have a mathematical model that p...
Investigation of rotor wake interactions in helicopters using 3d unsteady free vortex wake methodology Yemenici, Öznur; Uzol, Oğuz; Department of Aerospace Engineering (2010) This thesis focuses on developing and examining the capabilities of a new in-house aerodynamic analysis tool, AeroSIM+, and investigating rotor-rotor aerodynamic interactions for two helicopters, one behind the other in forward flight. AeroSIM+ is a 3-D unsteady vortex panel method potential flow solver based on a free vortex wake methodology. Validation of the results with the experimental data is performed using the Caradonna-Tung hovering rotor test case. AeroSIM+ code is improved for forward flight cond...
Nonlinear aeroservoelastic modelling and analysis of aircraft with control surface freeplay Yurtsever, Utku; Şahin, Melin; Kayran, Altan; Department of Aerospace Engineering (2022-11-24) In this study, nonlinear aeroservoelastic analysis of aircraft with control surface freeplay is performed using the fictitious mass approach. For the demonstration of the nonlinear aeroservoelastic analysis methodology, an available very light aircraft (VLA) design configuration is selected. The aeroelastic model of the aircraft is obtained by combining the structural and aerodynamic models of the aircraft, which are prepared by the finite element modelling and analysis tool MSC.Patran/MSC.Nastran and the a...
Optimization of compliant parts of a hybrid trailing edge control surface of a morphing unmanned aerial vehicle Arslan, Pınar; Gürses, Ercan; Department of Aerospace Engineering (2017) In this thesis, optimization studies are conducted for compliant parts of a hybrid trailing edge control surface of an unmanned aerial vehicle (UAV). The geometry of the control surface was taken from a previous study conducted in [1], and then regenerated parametrically through Design Modeler tool of ANSYS Workbench v15.0. The finite element model of the control surface is created by using ANSYS Workbench v15.0 Static Structural module. The optimization study of the compliant part is conducted by using Ada...

Citation Formats

E. Erinç and İ. H. Tuncer, “MODELING OF STORE SEPARATION BEHAVIOR BASED ON A NEURAL NETWORK AND UNSTEADY FLOW SOLUTIONS,” presented at the Ankara International Aerospace Conference, 10 - 12 Eylül 2015, 2015, Accessed: 00, 2021. [Online]. Available: http://aiac.ae.metu.edu.tr.