Multidisciplinary design optimization of dynamic systems using surrogate modeling approach

Dynamical Systems;Surrogate Modeling;Multidisciplinary Design Optimization;Wind Turbine Design

Deshmukh, Anand ; Allison ; James T.

关键词: Dynamical Systems;    Surrogate Modeling;    Multidisciplinary Design Optimization;    Wind Turbine Design;   

Others  :  https://www.ideals.illinois.edu/bitstream/handle/2142/44209/Anand_Deshmukh.pdf?sequence=1&isAllowed=y

美国|英语

来源: The Illinois Digital Environment for Access to Learning and Scholarship

【 摘 要 】

This thesis aims to explore a new paradigm for efficient solution of multi-disciplinary design optimization (MDO) of dynamic systems. Many of the MDO problems in dynamic systems design often involve computationally expensive system simulations, severely limiting their design optimization. This work demonstrates a novel method which approximates the expensive system dynamics by cheap-to-evaluate surrogate models for system derivative functions. This is advantageous to do, since it preserves the inherent natureof dynamic system to certain accuracy and enables the effi cient solution of MDO problems at the same time. The proposed method is demonstrated on a real world example of wind turbine design and obtained results are veryencouraging.