Neural Network Based Representation of UH-60A Pilot and Hub Accelerations

Abstract

Neural network relationships between the full-scale, experimental hub accelerations and the componding pilot floor vertical vibration are studied. The present physics-based, quantitative effort represents an initial systematic study on the UH-60A Black Hawk hub accelerations. The NASNArmy UH-6OA Airloads Program flight test database was used. A "maneuvereffect-factor 0,'' derived using the roll-angle and the pitch-rate, was used. Three neural network based representation-cases were considered. The pilot floor vertical vibration was considered in the first case and the hub accelerations were separately considered in the second case. The third case considered both the hub accelerations and the pilot floor vertical vibration. Neither the advance ratio nor the gross weight alone could be used to predict the pilot floor vertical vibration. However, the advance ratio and the gross weight together could be used to predict the pilot floor vertical vibration over the entire flight envelope. The hub accelerations data were modeled and found to be of very acceptable quality. The hub accelerations alone could not be used to predict the pilot floor vertical vibration. Thus, the hub accelerations alone do not drive the pilot floor vertical vibration. However, the hub accelerations, along with either the advance ratio or the gross weight or both, could be used to satisfactorily predict the pilot floor vertical vibration. The hub accelerations are clearly a factor in determining the pilot floor vertical vibration. potation MEFManeuver effect factor, Equation The present study is the first systematic effort that considers hub accelerations in a quantitative manner, and attempts to identify numerical relationships between the hub accelerations and t?e fuelage vibrations. Also, this study was undertaken to obtain a better understanding of the basic dynamics underlying the main rotordependent fuselage vibration and the associated hub accelerations. The present study builds up on previous neural network studies that were conducted at NASA Ames in the areas of rotorcraft performance, acoustics, and dynamics (Refs. 2 to 8).Flight test data from the NASNArmy UH-60A Airloads Program (Refs. 9 and 10) were used in this study.Reference 4 studied the neural network based modeling of the UH-60A peak, 4P pilot floor vertical vibration (PW) for real-time applications. In Ref.

Adaptive Neuro-Fuzzy Modeling of UH-60A Pilot Vibration

“…1. The inputs for the ANFIS are the same parameters that were used earlier 4 for training the neural networks and details are given later in the Results section. The single output of ANFIS is the predicted peak, 4P pilot floor vertical vibration, PVV, of UH-60A helicopter.…”

“…It has been shown that neural network based representations can be obtained for the UH-60A pilot vibration and the hub accelerations 4,5,6 . Similarly, tilt-rotor performance and noise were considered using only neural networks, without any fuzzy logic considerations 7,8 .…”

“…Previous studies 4,5 on the representation of UH-60A pilot vibration have used only neural networks, i.e., fuzzy logic considerations were not included. It has been shown that neural network based representations can be obtained for the UH-60A pilot vibration and the hub accelerations 4,5,6 .…”

“…Similarly, tilt-rotor performance and noise were considered using only neural networks, without any fuzzy logic considerations 7,8 . On-line or real-time implementation considerations were not addressed in 4,5 because the backpropagation training algorithm is very slow and cannot be used on-line.…”

“…A 200 point database that approximates the entire UH-60A flight envelope, and which was used in 4,5,6 is used here also for training and testing purposes. The NASA/Army Airloads Program flight test database 9 was the source of the above 200 point database.…”