Skip to content

Applying a Multi-Layer Perceptron Deep Neural Network to predict Lift and Drag performance of airfoils

Notifications You must be signed in to change notification settings

pptstall18/Airfoil-Optimization-using-DNN

Repository files navigation

This repository contains hyperparameter optimization of a DNN by performing grid search on model training parameters consisting of: Activation functions, batch size, learning rate, number of neurons per layer, number of layers, optimizer choice. All denoted by hp_xxxx

main.py is the main script used for training the model naca4_clcd_turb_st_3para.pkl is the pickle file containing data on airfoils used for model training/testing

final.pdf is the report analyzing DNN performance in predicting wing performance once model hyperparameters are optimized

About

Applying a Multi-Layer Perceptron Deep Neural Network to predict Lift and Drag performance of airfoils

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages