Filters








4 Hits in 16.6 sec

Surrogate-Based Aerodynamic Design Optimization: Use of Surrogates in Aerodynamic Design Optimization

M. Ahmed, N. Qin
2009 International Conference on Aerospace Sciences and Aviation Technology  
One of the latest developments in this field is the introduction of surrogates or metamodels.  ...  The role of optimization in various practical applications has been increasing over the years. In the field of aerodynamics, design optimization is rapidly evolving in many ways.  ...  They used gradient-based optimizer to search the minimum weight of a plate with variable thickness under bending and buckling loads.  ... 
doi:10.21608/asat.2009.23442 fatcat:xcqggw22ifflhodwqj4wxaxts4

Robust optimization of mechanical systems affected by large system and component variability = Robustheitsbasierte Optimierung für mechanische Systeme mit großer Streuung der relevanten System- und Elementgrößen

Pablo Enrique Leslabay
2010
The last step is to integrate the knowledge attained through parameter analysis and robustness assessment into an optimization loop, in order to improve the stochastic performance of the different output  ...  Abstract The following work deals with methods based on computer-simulation aimed both to determine the robustness and/or reliability of complex mechanical systems affected by large tolerances in the individual  ...  Within the scope of Collaborative Research Center 499 "Design, Production and Quality Assurance of Molded Micro Components made of Metallic and Ceramic Materials"  ... 
doi:10.5445/ir/1000017557 fatcat:ydmooy3iajcqbjvhmhhebr4ydy

Bayesian Network Classifiers for Set-Based Collaborative Design

David Shahan, Carolyn C. Seepersad
2010 Volume 1: 36th Design Automation Conference, Parts A and B   unpublished
Because metamodels interpolate the output of simulations based upon a discrete set of design points, they can be independently developed and shared for the purposes of concurrent distributed design.  ...  Implementing such a design could be costly because the number of variables in the optimization could be high.  ...  sigmas %fully connected BN I = 1; %repeat I times M = 100; %the number of data points per each of I iterations end end %plot the decision surface %surface(xh1(:,1),xh1(:,2),dph1,'FaceAlpha',0.2,'EdgeColor  ... 
doi:10.1115/detc2010-28724 fatcat:47mpo7765jbefdzo5tbubgkobm

Seventh International Workshop on Simulation, 21-25 May, 2013, Department of Statistical Sciences, Unit of Rimini, University of Bologna, Italy. Book of Abstracts

Mariagiulia Matteucci
2013
Seventh International Workshop on Simulation, 21-25 May, 2013, Department of Statistical Sciences, Unit of Rimini, University of Bologna, Italy. Book of Abstracts  ...  We discuss some properties of this test, while the goal of our simulations studies is to evaluate its power. In this simulation studies d = 10000 and k varies from 10 to 100.  ...  We present an evaluation of the approach through limited simulations, and the analysis of two clinical data sets, one on warfarin initiation, the other on the time evolution of a delirium index.  ... 
doi:10.6092/unibo/amsacta/3677 fatcat:3c3ua63y5jgl5drpjxd3qwo5yi