Pareto Frontier Method for Multi-Disciplinary Optimization of Helicopter Rotors

Limitations of rotorcraft generally fall into the categories of speed, noise, vibration, and range. Rotor blade design plays a large role in how these limitations may be mitigated. Designers are always looking for new ways to design more optimum rotor systems. In this work, a methodology is presented that uses both low and high fidelity analysis tools to pick and examine a redesign of the HART model rotor. The design framework ModelCenter® is used to integrate the comprehensive rotorcraft analysis code RCAS with the aeroacoustics code PSU-WOPWOP. This low fidelity model assumes rigid blades, prescribed wake aerodynamics, and compact blade loading for aeroacoustics. Design variables define tip geometry and twist distribution with power required and noise in both hover and forward flight considered as objectives to improve. A design of computer experiments is performed. Surrogate models of the objectives are built and subsequently used to examine four million stochastically generated design variable combinations in what is called Monte Carlo simulation. The simulation results are filtered to identify the Pareto optimal designs within the group. One of these designs is selected for examination using higher fidelity CFD based tools. The hover analysis is performed with TURNS while analysis of forward flight is done using an elastic blade RCAS model coupled with the efficient rotorcraft CFD code GT-HYBRID. Noise prediction is made by PSU-WOPWOP using blade surface pressures from TURNS and GT-HYBRID. The optimum is found to exhibit better performance characteristics and reduced noise. An a posteriori examination of vibratory characteristics reveals the optimum produces more vibration in forward flight than the baseline, highlighting the need to consider vibration in during the first phase using low fidelity tools. This record was migrated from the OpenDepot repository service in June, 2017 before shutting down