Abstract
Gear design requires the designer to compromise many design variables in order to determine the best performance of a gear set. Unfortunately the designer has a multiplicity of goals including keeping both bending and pitting stresses under an allowable value, minimizing scoring, achieving minimum efficiency and trying to minimize noise. This latter response variable is rarely considered in the initial gear design. In this work, noise is considered to be one of the more important design considerations.
One approach to multi-variable gear design that has been tried is design optimization. Usually optimization techniques are limited in the number of variables that can be handled and with so many response variables, it is difficult to come up with an objective function that reflects the considerations of a real gear designer.
In this paper we present a simulation-based approach to gear design that allows the designer to essentially “run all of the cases”. The simulation accounts for the true load distribution of the gears when computing response variables. Also, such factors as manufacturing tolerances may be included in the simulation so that truly robust designs may be obtained. Rather than using an objective function approach, designs are analyzed with a dominance filter that assesses each response variable in a manner that results in the “best” design. After these “best” designs are found, an interactive viewer allows the selection of those designs that best meet the designer’s goals with regard to all design variables.
Several examples are presented in this paper. In each case, over 65,000 designs are evaluated and the dominance filter results in from 200 to 900 successful designs, depending on the tolerances that are applied. After sorting with the viewer the designer usually ends up with from 5 to 20 designs whose features may vary significantly, but have similar performances.
Comparison between classical ‘P&O’ algorithm and FLC of MPPT for GPV under partial shading
