11th AIAA/ISSMO Multidisciplinary Analysis and Optimization Conference · 2006 · 42 citations · 20 references
EngineeringComputer ScienceComputer ArchitectureDesign AutomationSystems ArchitectureAlgorithmic PerformanceArchitecture EvaluationParallel ArchitecturesSoftware ArchitectureSolution ArchitectureArchitectural ModularityArchitecture ModelsComputational ArchitectureModular ArchitectureModular DesignProblem Reformulation
A new MDO framework has been developed in Python to provide a platform for comparisons relating the performance of various MDO architectures. The framework eliminates the need for problem reformulation when solving MDO problems in multiple architectures. Once a problem has been described, the implementation in any architecture is automatic. In addition to providing a performance comparison of the more common architectures, the modular design of the framework allows rapid testing of newly developed architectures. Results generated from this study provide a strong foundation for identifying the performance trends of various architectures with a variety of problem classes. It also serves to verify the performance and potential of those tested.
20
Problem Formulation for Multidisciplinary Optimization
Evin Cramer, J. E. Dennis, Paul D. Frank et al. · SIAM Journal on Optimization · 1994 · 643 citations