Design space optimization using a numerical design continuation method

2002 ◽  
Vol 53 (8) ◽  
pp. 1979-2002 ◽  
Author(s):  
Il Yong Kim ◽  
Byung Man Kwak
Keyword(s):  
Design Space ◽  
Continuation Method ◽  
Numerical Design

Parametric exploration on the airfoil design space by numerical design of experiment methodology and multiple regression model

2019 ◽  
Vol 234 (1) ◽  
pp. 3-18 ◽  
Author(s):  
Xingxing Li ◽  
Ke Yang
Keyword(s):  
Regression Model ◽  
Wind Turbine ◽  
Multiple Regression ◽  
Design Space ◽  
Leading Edge ◽  
Multiple Regression Model ◽  
Highly Nonlinear ◽  
Numerical Design ◽  
Wind Turbine Airfoil ◽  
The Cost

Robust airfoil design is crucial to efficient, stable, and safe operation for modern wind turbines. However, even for deterministic wind turbine airfoil design, the problem is complex regarding to aerodynamic, acoustic, and structural requirements of wind turbine blades. Therefore, this study aims to assess the design variable impact, identify significant variables, and obtain the correlation with the airfoil responses, to reduce the cost of the airfoil robust optimization. In this paper, the optimal hypercube design method was applied to an airfoil designed by the National Advisory Committee for Aeronautics, NACA 63-421, which is commonly employed in the outboard modern wind turbine blade, to perform the numerical design of experiments. Then, a parametric exploration on the characteristics of airfoil design space by the multiple regression model and statistical analysis method were conducted. It was identified that in regular design space, the variations of aerodynamic and structural parameters are dominated by the airfoil camber and radius of leading edge. Meanwhile, the chord-wise position of the maximum thickness also has strong impacts on the airfoil performance. In further, the overall design spaces are explored to be highly nonlinear in aerodynamic and acoustic responses because of the nonlinear effects of the airfoil chord-wise position of the maximum camber and radius of leading edge. Strong but undesirable correlations were demonstrated between the maximum lift-to-drag ratio and the total sound pressure level. These findings could serve as a valuable guidance for wind turbine airfoil robust design to screen the stochastic design variables, simplify the design space, and reduce the cost.

Bayesian Optimization of Equilibrium States in Elastomeric Beams

2019 ◽  
Author(s):  
David Yoo ◽  
Carson Wiley ◽  
Andrew Gillman ◽  
Vincent Chen ◽  
Abigail Juhl ◽  
...  
Keyword(s):  
Vibration Isolation ◽  
Design Space ◽  
Continuation Method ◽  
Computational Cost ◽  
Computational Design ◽  
Equilibrium States ◽  
Bayesian Optimization ◽  
Single Element ◽  
Element Analysis ◽  
Trade Off

Abstract Architected elastomers have demonstrated great potential for energy absorption, multi-resonant vibration isolation, and multi-bandgap acoustic control, due to the reversibility and programmability of their mechanical instabilities. However, computational design tools are needed to explore the large parameter space that regulates buckling-based mechanical instability behavior. In this study, we develop a machine-learning-based design optimization framework to control equilibrium states of a bistable elastomeric beam, while also regulating the required energy to transition between these configurations. Leveraging symmetry to reduce the design space, the research is performed on a single element, an inclined, slender beam, through a Fourier series parameterization. To evaluate the force-displacement response of the bistable beam, a nonlinear finite element analysis (FEA) with an arc-length continuation method is employed in a commercial FEA tool. Due to the highly non-convex bistability response of the beam in this proposed design space and the computational cost of the FEA analysis, a Bayesian optimization is implemented to promote a better trade-off between the number of function evaluations and rate of convergence. Bayesian optimization depends on several optimization parameters, which are systematically tuned in this study to characterize their role on the optimization process. With the proposed method, the equilibrium displacement and the ratio of output to input energy between stable states can be optimized within a few tens of iterations. A multi-objective optimization is also carried out to study the trade-off between equilibrium position and the energetics to transition between the bistabilities.

A Reference Error Formulation for Multi-Fidelity Design Optimization

2017 ◽  
Author(s):  
Ahmed H. Bayoumy ◽  
Michael Kokkolaras
Keyword(s):  
Design Optimization ◽  
Computational Models ◽  
Design Space ◽  
Computational Cost ◽  
Trust Region ◽  
Model Management ◽  
Flexible Beam ◽  
Management Framework ◽  
Numerical Design ◽  
Region Model

We consider the problem of selecting among different computational models of various fidelity for evaluating the objective and constraint functions in numerical design optimization. Typically, higher-fidelity models are associated with higher computational cost. Therefore, it is desirable to employ them only when necessary. We introduce a reference error formulation that aims at determining whether lower-fidelity models (that are computationally cheaper) can be used in certain areas of the design space as the latter is being explored during the optimization process. The proposed approach is implemented using an existing trust region model management framework. We demonstrate the link between feasibility and fidelity and the key features of the proposed approach using the design example of a cantilever flexible beam subject to high accelerations.

Numerical Design of Matching Structures for One-Dimensional Finite Superlattices

2013 ◽  
Vol E96.C (11) ◽  
pp. 1440-1443 ◽  
Author(s):  
Hirofumi SANADA ◽  
Megumi TAKEZAWA ◽  
Hiroki MATSUZAKI
Keyword(s):  
One Dimensional ◽  
Numerical Design

The Environmental Design Space: Modeling and Performance Updates

2021 ◽  
Author(s):  
Luis Salas Nunez ◽  
Jimmy C. Tai ◽  
Dimitri N. Mavris
Keyword(s):  
Design Space ◽  
Environmental Design ◽  
Space Modeling ◽  
And Performance

The impact of design space constraints on the noise and emissions from derivative engines for civil supersonic aircraft

2021 ◽  
Author(s):  
Laurens Voet ◽  
Prakash Prashanth ◽  
Raymond Speth ◽  
Jayant Sabnis ◽  
Choon Tan ◽  
...  
Keyword(s):  
Design Space ◽  
Supersonic Aircraft ◽  
The Impact

Proposal of an instrument to evaluate interaction spaces in a VLE

2019 ◽  
pp. 1-13
Author(s):  
Rubí Estela Morales-Salas ◽  
Daniel Montes-Ponce
Keyword(s):  
Learning Environment ◽  
Design Space ◽  
Collaborative Work ◽  
Virtual Learning ◽  
Interaction Space ◽  
Learning Objects ◽  
Traditional Education ◽  
Virtual Learning Environment ◽  
Discussion Forums ◽  
Exhibition Space

A virtual learning environment is conceived as an interaction space that ease the realization of mediated activities by technology, in this case the internet; besides using multimedia materials, learning objects, social networks, among others; which have changed imminently the traditional education. In this article an instrument is proposed in a checklist format, to evaluate any platform that has interaction spaces such as a Virtual Learning Environment, in this case responding to four spaces or general indicators: information Space, Mediation / Interaction Space, Instructional Design Space and Exhibition Space. Criteria are used according to the interactions and activities carried out by the consultant and virtual student. These, in turn, come up from the analysis and interaction of the advisers achieved in the discussion forums and portfolio activities through collaborative work. It was situated as a qualitative research, with a descriptive nature since it is not limited to data collection only, but also it refers and analyzes the interaction of the advisers achieved in the discussion forums and portfolio activities through the collaborative work of the workshop course "Virtual Learning Environments" developed in a virtual learning environment.

Optimization Study to Develop Once-a-day Controlled Release Formulation of Metoclopramide with Predictable Design Space

1970 ◽  
Vol 1 (1) ◽  
pp. 71-76
Author(s):  
Rajesh Dubey ◽  
Udaya K. Chowdary ◽  
Venkateswarlu V.
Keyword(s):  
Controlled Release ◽  
Drug Release ◽  
Design Space ◽  
Release Kinetics ◽  
Release Profile ◽  
Release Formulation ◽  
Linear Effect ◽  
Controlled Release Formulation ◽  
The Difference ◽  
Face Centered

A controlled release formulation of metoclopramide was developed using a combination of hypromellose (HPMC) and hydrogenated castor oil (HCO). Developed formulations released the drug over 20 hr with release kinetics following Higuchi model. Compared to HCO, HPMC showed significantly higher influence in controlling the drug release at initial as well as later phase. The difference in the influence can be explained by the different swelling and erosion behaviour of the polymers. Effect of the polymers on release was optimized using a face-centered central composite design to generate a predictable design space. Statistical analysis of the drug release at various levels indicated a linear effect of the polymers’ levels on the drug release. The release profile of formulations containing the polymer levels at extremes of their ranges in design space was found to be similar to the predicted release profile

Sign in / Sign up

Export Citation Format

Share Document