Robust Multiobjective Optimization Through Collaborative Optimization and Linear Physical Programming
Charles D. McAllister, Timothy W. Simpson, Kemper E. Lewis, Achille Messac · 10th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference · 2004
Rennselaer Polytechnic Institute, Troy, NY, 12180 Multidisciplinary design optimization (MDO) is a concurrent engineering design tool for large-scale, complex systems design that can be affected through the optimal design of several smaller functional units or subsystems. Due to the multiobjective nature of many MDO problems, recent work has focused on formulating the MDO problem to help resolve tradeoffs between multiple, conflicting objectives. In this paper, we describe the novel integration of Linear Physical Programming within the Collaborative Optimization framework to enable designers to formulate multiple system-level objectives in terms of physically meaningful parameters. The proposed formulation extends our previous multiobjective formulation of Collaborative Optimization, which uses Goal Programming at the system level and subsystem level to enable multiple objectives to be considered at both levels during optimization. The proposed framework is demonstrated using two MDO applications: (1) the design of a Formula 1 racecar and (2) the configuration of an autonomous underwater vehicle. Results obtained from the proposed formulation are compared against a traditional formulation without Collaborative Optimization or Linear Physical Programming.