Researchers applied PDS to two versions of the rotor design software, one with 10 design variables, and one with 56 design variables. In both cases, it converged to a "best-known" optimum on the first try. The simulation uses vibration suppression as a figure of merit for a given design, and as is often the case in computational engineering, the analysis is better understood than is the relevance of the particular figure of merit. The analysis involves coupling thermodynamics, structures, and aerodynamics, but the versions run for this project assume a constant rotor wake. Boeing is working to parallelize the full physics simulation including wake calculations, and PDS' successes on the constant wake problems qualifies CRPC researchers for a try at the full problem.
Boeing uses the 10-variable problem to more or less eliminate ineffective methods, but the 56-variable problem is considered more realistic. PDS gave more than a 50% reduction in vibration over the baseline design Boeing provided to start the optimization. Currently, Boeing is assessing the optimum design suggested by PDS. The question is whether it is a completely new design or a refinement of a previously known design.