Improved collaboration pursuing method for multidisciplinary robust design optimization
- 102 Downloads
The collaboration pursuing method (CPM) is a computationally efficient approach for deterministic multidisciplinary design optimization (MDO). However, it has not been employed to handle problems with uncertainty. Moreover, obtaining actual uncertainty probability distributions is challenging. Compared to probability distributions, the interval information of uncertainties can be more easily obtained. Thus, multidisciplinary robust design optimization (MRDO) under interval uncertainty has been widely investigated. To overcome the inefficiency of existing methods for solving MRDO, this paper proposes an improved collaboration pursuing method (ICPM). In this method, the collaboration model (CM) is utilized to filter the samples that satisfy the coupled state equations in system analysis (SA) or multidisciplinary analysis (MDA). Then, a robustness discrepancy model (RDM) is developed to efficiently select candidate samples that meet the robustness requirements. Next, the mode pursuing sampling (MPS) method is utilized as a global optimizer to drive the optimization process and identify the robust optimum. Finally, a mathematical example and two engineering examples are utilized to evaluate the feasibility and effectiveness of the proposed method.
KeywordsMultidisciplinary robust design optimization (MRDO) Improved collaboration pursuing method (ICPM) Mode pursuing sampling (MPS) Interval uncertainty
This work was supported by the National Natural Science Foundation of China [grant numbers 51675196 and 51721092] and the Program for HUST Academic Frontier Youth Team.
- Parsons MG, Randall LS (2004) Formulation of multi-criterion design optimization problems for solution with scalar numerical optimization methods. J Ship Res 48(1):61–76Google Scholar
- Peri D, Campana EF (2003) Multidisciplinary design optimization of a naval surface combatant. J Ship Res 47(1):1–12Google Scholar
- Sellar RS (1997) Multidisciplinary design using artificial neural networks for discipline coordination and system optimization [D]. University of Notre Dame, IndianaGoogle Scholar
- Wujek BA, Renaud JE, Batill SM, Johoson EW, Brockman, J B (1996) Design flow management and multidisciplinary design optimization in application to aircraft concept sizing. AIAA-96-0713, 34th aerospace sciences meeting & exhibit, Reno, NYGoogle Scholar
- Zhou J, Li M, Xu M (2015) A new sequential multidisciplinary optimization method for bi-level decomposed systems. ASME 2015 International design engineering technical conferences and computers and information in engineering conference. American Society of Mechanical Engineers, V02BT03A031-V02BT03A031Google Scholar