VFS Website
  • VIEW CART
  • CUSTOMER SUPPORT
  • MY STORE ACCOUNT
  • CONTACT US
  • STORE HOME
  • 5Prime
  • Forum Proceedings
  • Workshops
  • Technical Meetings
  • Vertiflite
  • Books, CDs & Gifts


Unable to log in or get member pricing? Having trouble changing your password?

Please review our Frequently Asked Questions for complete information on these and other common situations.
 

Vertical Flight Library & Store

CHECKOUT

0 Item(s) In Cart Total: $0.00


Neural Network and Machine Learning Allocation of Redundant Controls for Power Optimization on a Compound Helicopter

Jean-Paul Reddinger, Farhan Gandhi

May 8, 2017

  • Your Path :
  • Home
  • > Neural Network and Machine Learning Allocation of Redundant Controls for Power Optimization on a Compound Helicopter

On Sale: Neural Network and Machine Learning Allocation of Redundant Controls for Power Optimization on a Compound Helicopter

  • Presented at Forum 73
  • 13 pages
  • SKU # : 73-2017-0333
  • Your Price : $30.00
  • Join or log in to receive the member price of $5.00!


VFS member?
Don't add this to your cart just yet!
Be sure to log in first to receive the member price of $5.00!

 

Add to Wish List

Reward Value:
(60) Member Points

Neural Network and Machine Learning Allocation of Redundant Controls for Power Optimization on a Compound Helicopter

Authors / Details: Jean-Paul Reddinger, Farhan Gandhi

Abstract
For a compound helicopter with control of main rotor speed, auxiliary propeller thrust, and stabilator pitch; a predictive neural network is trained to estimate power as a function of the redundant control settings for a range of flight speeds using a comprehensive database of 2,335 prior Rotorcraft Comprehensive Analysis System (RCAS) simulations. This neural network can be used as a surrogate model for a gradient based optimization to find the redundant control settings that produce a trim state with a minimized power requirement. The study highlights the importance of modeling relevant constraints (main rotor flapping and control limits), as well as having training data in the same locality that the predictions are being made. For producing accurate results with sparse initial datasets (20 trim states), a process of machine learning is demonstrated, wherein the neural network is iteratively updated using the resulting power from each prior optimization. When used appropriately, both the machine learning and fixed models demonstrate ability to allocate redundant controls such that the power requirement is within 1% - 3% of the true minimum.

Popular Products

  • Master Card
  • Visa
  • American Express
  • Customer Support
  • Contact Us
  • Privacy and Security Policies
  • Refund Policies

Copyright © 2022 The Vertical Flight Society. All rights reserved.