Engineering topics
D Douglas Boyd
Publications and source records attributed to D Douglas Boyd.
Comparison of Prediction Modeling Methodologies for Aeroacoustic Characterization of Hovering sUAS Rotors
This work compared artificial neural network and multivariate orthogonal function modeling methodologies for the prediction and characterization of isolated hovering sUAS rotor aerodynamics and aeroacoustics. Design of Experiments was used to create input feature spaces over 9 input features: the number of rotor blades, rotor size, rotor speed, the amount of blade twist, blade taper ratio, tip chord length, collective pitch, airfoil camber, and airfoil thickness. CAMRAD~II and AARON were executed at the points defined by the input feature space to predict aerodynamic and aeroacoustic quantities. These predicted aerodynamic and aeroacoustic data were then used to generate artificial neural networks and polynomial response surface models. The two prediction model methodologies were evaluated over test data previously unseen by the models, which showed good prediction capabilities for both model types, with slightly lower prediction error for the artificial neural networks. A characterization study was performed, which showed that input features correspondent to the spanwise sectional blade lift and drag were the most significant factors to the aerodynamic thrust and power, respectively. It was also shown that the aeroacoustic quantities were highly dependent on variations in rotor speed and size, which affect the Doppler factor for tonal noise and the spanwise Reynolds number for broadband noise.
Effect of Rotor Blade Elasticity on UAM Quadrotor Acoustics
The 6-passenger quadrotor concept vehicle designed under the NASA Revolutionary Vertical Lift Technology (RVLT) Project is considered for acoustic analysis. The tip speed is 550 ft/sec, with three blades per rotor (550/3). Originally, the blades were rigid, uniform spanwise, and with flap and pitch degrees of freedom. The blade model has since been updated: a lag hinge was added, and nonuniformities and elastic properties were introduced. Four blade models are considered: 1) original model, rigid uniform flap-pitch; 2) rigid uniform flap-lag-pitch; 3) rigid nonuniform flap-lag-pitch; and 4) elastic nonuniform flap-lag-torsion. Predictions are made for three flight conditions (approach, flyover, and takeoff) using the four blade models. The RVLT Toolchain is exercised using CAMRAD II and pyaaron/AARON/ANOPP2. Quadrotor trim and performance, 0.75R vertical blade loading for all four rotors, and noise sources are analyzed. Also, the contributions of the front and rear rotor pairs to noise are studied. In approach and flyover, a 2 dBA loading noise difference (delta) is predicted between blade models 1 and 4 (delta for takeoff is smaller, 1 dBA). Most of this noise delta is due to the lag hinge and nonuniformities, which is consistent with the results of a 2022 study that had considered only the approach condition; the current results extend this conclusion to flyover and takeoff also. The insensitivity of quadrotor noise to blade elasticity is currently attributed to the high blade torsional stiffness (frequency 6.41 per rev) and the small blade radius (9 ft) of the 550/3 design. Suggestions for potential follow-on work are given.
A Synthesis Method to Auralize Rotor Noise During Transitional Flight
Explore the source record for details and available documents.