Efficient PCA-driven EAs and metamodel-assisted EAs, with applications in turbomachinery

Kyriacou, Stylianos; Asouti, Varvara; Giannakoglou, Kyriakos C.

doi:10.1080/0305215x.2013.812726

Cited by 23 publications

(8 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There have been other methods that use principal components for highdimensional surrogate-based optimization. For example, Kyriacou et al (2014) used principal components analysis (PCA) in a metamodel-assisted evolutionary algorithm to guide the application of evolution operators and the training of the metamodel. Moreover, Chen et al (2015) used Karhunen-Loève expansion, which is similar to PCA, to reduce design-space dimensionality in metamodel-based shape optimization.…”

Section: Introductionmentioning

confidence: 99%

Efficient global optimization for high-dimensional constrained problems by using the Kriging models combined with the partial least squares method

Bouhlel¹,

Bartoli²,

Regis³

et al. 2018

Engineering Optimization

View full text Add to dashboard Cite

In many engineering optimization problems, the number of function evaluations is often very limited because of the computational cost to run one high-fidelity numerical simulation. Using a classic optimization algorithm, such as a derivative-based algorithm or an evolutionary algorithm, directly on a computational model is not suitable in this case. A common approach to addressing this challenge is to use black-box surrogate modeling techniques. The most popular surrogate-based optimization algorithm is the Efficient Global Optimization (EGO) algorithm, which is an iterative sampling algorithm that adds one (or many) point(s) per iteration. This algorithm is often based on an infill sampling criterion, called expected improvement, which represents a trade-off between promising and uncertain areas. Many studies have shown the efficiency of EGO, particularly when the number of input variables is relatively low. However, its performance

show abstract

Section: Introductionmentioning

confidence: 99%

Efficient global optimization for high-dimensional constrained problems by using the Kriging models combined with the partial least squares method

Bouhlel¹,

Bartoli²,

Regis³

et al. 2018

Engineering Optimization

View full text Add to dashboard Cite

show abstract

“…In some aerodynamic designs, the application of PCA is mainly on the data processing, e.g., classifier. Some examples can be seen in reduced-order models for simulations balancing computation cost and flowfield accuracy in numerous fields [22,[53][54][55][56].…”

Section: Principal Component Analysis As Data Manipulatormentioning

confidence: 99%

The Objective Space and the Formulation of Design Requirement in Natural Laminar Flow Optimization

Wang

Sun

2020

Applied Sciences

View full text Add to dashboard Cite

Design requirement is as important in aerodynamic design as in other industries because it sets up the objective for the samples in design space to approach. Natural Laminar Flow (NLF) optimization belongs to the type of aerodynamic design problems featured by the combination of distinct aerodynamic performance, where the design requirement is often formulated in form of summation of laminar-related performance and pressure drag performance with different weight assignment according to different perspectives. However, the formulations are rather experience-oriented and are decided non-quantitatively. Inspired by data manipulation approaches in design space (spanned by design variables of geometrical representation parameters) in many aerodynamic designs, this paper proposes new formulations of design requirement in NLF optimization via consideration of objective space (projection of design space through aerodynamics) and shows the impact of the corresponding formulation of design requirement to the result of NLF optimization in cases of transonic airfoil and aero engine compressor blade design from two perspectives: Pareto front convergence and improving effect of accessory performance. The paper uses Principal Component Analysis (PCA) to obtain the eigenvectors of objective space to extract the intrinsic information about specific problem. The method is realized in two cases with satisfactory result.

show abstract

“…In the same context, Kyriacou et al. 26 showed that for certain applications, omitting the PCs with the largest variances leads to improved modelling.…”

Section: Principal Component Analysismentioning

confidence: 99%

“…Jolliffe, 24 however, states and proves that this strategy does not apply to all cases and results strongly depend on the specific application. In the same context, Kyriacou et al 26 showed that for certain applications, omitting the PCs with the largest variances leads to improved modelling.…”

Section: Principal Component Analysismentioning

confidence: 99%

Neural network based dynamic model and gust identification system for the Jetstream G-NFLA

Antonakis

Lone

Cooke

2016

Proceedings of the Institution of Mechanical Engineers, Part G:

View full text Add to dashboard Cite

Artificial neural networks are an established technique for constructing non-linear models of multi-input-multi-output systems based on sets of observations. In terms of aerospace vehicle modelling, however, these are currently restricted to either unmanned applications or simulations, despite the fact that large amounts of flight data are typically recorded and kept for reasons of safety and maintenance. In this paper, a methodology for constructing practical models of aerospace vehicles based on available flight data recordings from the vehicles' operational use is proposed and applied on the Jetstream G-NFLA aircraft. This includes a data analysis procedure to assess the suitability of the available flight databases and a neural network based approach for modelling. In this context, a database of recorded landings of the Jetstream G-NFLA, normally kept as part of a routine maintenance procedure, is used to form training datasets for two separate applications. A neural network based longitudinal dynamic model and gust identification system are constructed and tested against real flight data. Results indicate that in both cases, the resulting models' predictions achieve a level of accuracy that allows them to be used as a basis for practical real-world applications.

show abstract

Efficient PCA-driven EAs and metamodel-assisted EAs, with applications in turbomachinery

Cited by 23 publications

References 19 publications

Efficient global optimization for high-dimensional constrained problems by using the Kriging models combined with the partial least squares method

Efficient global optimization for high-dimensional constrained problems by using the Kriging models combined with the partial least squares method

The Objective Space and the Formulation of Design Requirement in Natural Laminar Flow Optimization

Neural network based dynamic model and gust identification system for the Jetstream G-NFLA

Contact Info

Product

Resources

About