A methodology for rapid vehicle scaling and configuration space exploration

The Configuration-space Exploration and Scaling Methodology (CESM) entails the representation of component or sub-system geometries as matrices of points in 3D space. These typically large matrices are reduced using minimal convex sets or convex hulls. This reduction leads to significant gains in collision detection speed at minimal approximation expense. (The Gilbert-Johnson-Keerthi algorithm [79] is used for collision detection purposes in this methodology.) Once the components are laid out, their collective convex hull (from here on out referred to as the super-hull) is used to approximate the inner mold line of the minimum enclosing envelope of the vehicle concept. A sectional slicing algorithm is used to extract the sectional dimensions of this envelope. An offset is added to these dimensions in order to come up with the sectional fuselage dimensions. Once the lift and control surfaces are added, vehicle level objective functions can be evaluated and compared to other designs. The size of the design space coupled with the fact that some key constraints such as the number of collisions are discontinuous, dictate that a domain-spanning optimization routine be used. Also, as this is a conceptual design tool, the goal is to provide the designer with a diverse baseline geometry space from which to chose. For these reasons, a domain-spanning algorithm with counter-measures against speciation and genetic drift is the recommended optimization approach. The Non-dominated Sorting Genetic Algorithm (NSGA-II) [60] is shown to work well for the proof of concept study.;There are two major reasons why the need to evaluate higher fidelity, custom geometric scaling laws became a part of this body of work. First of all, historical-data based regressions become implicitly unreliable when the vehicle concept in question is designed around a disruptive technology. Second, it was shown that simpler approaches such as photographic scaling can result in highly suboptimal concepts even for very small scaling factors. Yet good scaling information is critical to the success of any conceptual design process. In the CESM methodology, it is assumed that the new technology has matured enough to permit the prediction of the scaling behavior of the various subsystems in response to requirement changes. Updated subsystem geometry data is generated by applying the new requirement settings to the affected subsystems. All collisions are then eliminated using the NSGA-II algorithm. This is done while minimizing the adverse impact on the vehicle packing density. Once all collisions are eliminated, the vehicle geometry is reconstructed and system level data such as fuselage volume can be harvested. This process is repeated for all requirement settings. Dimensional analysis and regression can be carried out using this data and all other pertinent metrics in the manner described by Mendez [124] and Segel [173]. The dominant parameters for each response show up as in the dimensionally consistent groups that form the independent variables. More importantly the impact of changes in any of these variables on system level dependent variables can be easily and rapidly evaluated. In this way, the conceptual design process can be accelerated without sacrificing analysis accuracy. Scaling laws for take-off gross weight and fuselage volume as functions of fuel cell specific power and power density for a notional General Aviation vehicle are derived for the proof of concept.;CESM enables the designer to maintain design freedom by portably carrying multiple designs deeper into the design process. Also since CESM is a bottom-up approach, all proposed baseline concepts are implicitly volumetrically feasible. System level geometry parameters become fall-outs as opposed to inputs. This is a critical attribute as, without the benefit of experience, a designer would be hard pressed to set the appropriate ranges for such parameters for a vehicle built around a disruptive technology. Furthermore, scaling laws generated from custom data for each concept are subject to less design noise than say, regression based approaches. Through these laws, key physics-based characteristics of vehicle subsystems such as energy density can be mapped onto key system level metrics such as fuselage volume or take-off gross weight. These laws can then substitute some historical-data based analyses thereby improving the fidelity of the analyses and reducing design time. (Abstract shortened by UMI.)...