Flight Turbulence Level Classificator using a Multilayer Perceptron Network Trained with Flight Test Data

Authors

  • Matheus Oliveira Universidade de São Paulo
  • Jorge Bidinotto Universidade de São Paulo

Keywords:

Artificial Neural Networks, Cross Validation, Flight Test, Flight Turbulence, Levenberg-Marquardt Backpropagation, Multilayer Perceptron, Turbulence Level Classification

Abstract

This study presents the development of an artificial neural network (ANN) that classifies from flight data, the flight turbulence level encountered by an aircraft. The input data is divided into three different groups that contributes for turbulence level classification: Flight Condition, Aerodynamic Configuration and Turbulence Measurement. There are two main methods applied at turbulence measurement, Power Espectrum Density of aircraft vertical acceleration signal and Discrete Gust calculated from inertial and anemometric aircraft data source. The ANN model developed is a Multilayer Perceptron which was trained with Levenberg-Marquardt Backpropagation algorithm, using flight test data of a specific aircraft prototype. The flight test data used at learning process consists of both recorded parameters and flight test crew subjective flight turbulence level classification. The most precise model developed (within the sixteen models proposed and analyzed) were trained also with Cross Validation method due to lack of samples that represents all possible characteristics of the flight turbulence phenomenon.

Downloads

Download data is not yet available.

Author Biographies

Matheus Oliveira, Universidade de São Paulo

Graduado em Engenharia de Controle e Automação pela Universidade Paulista. Atualmente é Engenheiro de Ensaios em Voo pela Embraer S.A. e mestrando no Departamento de Engenharia Aeronáutica da EESC/USP, onde realiza pesquisa em aplicações de sistemas inteligentes na área aeroespacial.

Jorge Bidinotto, Universidade de São Paulo

Possui graduac¸ ˜ao em Engenharia Mecˆanica com ˆenfase em Aeronaves pela EESC/USP, com mestrado e doutorado na mesma ´area, pela mesma instituic¸ ˜ao. Atualmente, ´e Professor Doutor do Departamento de Engenharia Aeron´autica da EESC/USP, atuando nas ´areas de Sistemas de Controle, Aviˆonica, Navega´ç´˜ao, Ensaios em Voo e Fatores Humanos em Aviação. Atuou durante 10 anos como Engenheiro de Ensaios em Voo pela Embraer S.A., totalizando aproximadamente 500 horas de voo de ensaios.

Published

2020-04-24

How to Cite

Oliveira, M., & Bidinotto, J. (2020). Flight Turbulence Level Classificator using a Multilayer Perceptron Network Trained with Flight Test Data. IEEE Latin America Transactions, 18(5), 954–961. Retrieved from https://latamt.ieeer9.org/index.php/transactions/article/view/2744