Expected improvement for expensive optimization: a review

Zhan, Dawei; Xing, Huanlai

doi:10.1007/s10898-020-00923-x

Cited by 105 publications

(39 citation statements)

References 142 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Further details of the variants of efficient global optimization are discussed in Refs. [75][76][77][78].…”

Section: Efficient Global Optimization Methodsmentioning

confidence: 99%

Metamodel-based multidisciplinary design optimization methods for aerospace system

Shi

et al. 2021

Astrodyn

View full text Add to dashboard Cite

The design of complex aerospace systems is a multidisciplinary design optimization (MDO) problem involving the interaction of multiple disciplines. However, because of the necessity of evaluating expensive black-box simulations, the enormous computational cost of solving MDO problems in aerospace systems has also become a problem in practice. To resolve this, metamodel-based design optimization techniques have been applied to MDO. With these methods, system models can be rapidly predicted using approximate metamodels to improve the optimization efficiency. This paper presents an overall survey of metamodel-based MDO for aerospace systems. From the perspective of aerospace system design, this paper introduces the fundamental methodology and technology of metamodel-based MDO, including aerospace system MDO problem formulation, metamodeling techniques, state-of-the-art metamodel-based multidisciplinary optimization strategies, and expensive black-box constraint-handling mechanisms. Moreover, various aerospace system examples are presented to illustrate the application of metamodel-based MDOs to practical engineering. The conclusions derived from this work are summarized in the final section of the paper. The survey results are expected to serve as guide and reference for designers involved in metamodel-based MDO in the field of aerospace engineering.

show abstract

“…Further details of the variants of efficient global optimization are discussed in Refs. [75][76][77][78].…”

Section: Efficient Global Optimization Methodsmentioning

confidence: 99%

Metamodel-based multidisciplinary design optimization methods for aerospace system

Shi

et al. 2021

Astrodyn

View full text Add to dashboard Cite

show abstract

“…To improve the accuracy, we need to add some new sample points generated by the infill criteria to the initial Kriging model. The infill criterion adopted in this paper is the mean square error (MSE) infillsampling criterion, which can effectively improve the global precision of the Kriging model [21], [22]. The main idea of the MSE criterion is that the point with the maximum mean squared error in the design space is infilled into the Kriging model.…”

Section: Infill Criteriamentioning

confidence: 99%

“…where w is the inertial weight used to improve the calculation speed and the quality, w∈[w min , w max ] (refer to (21)). c 1 is the cognition learning rate, and c 2 is the social learning rate.…”

Section: Yesmentioning

confidence: 99%

Kriging Aerodynamic Modeling and Multi-Objective Control Allocation for Flying Wing UAVs With Morphing Trailing-Edge

Zhao

Yang

2021

IEEE Access

View full text Add to dashboard Cite

Flexible skin and continuous deformable control surface are the basic of adaptive wing technology for future aircraft. This paper presents a continuous morphing trailing-edge and its control allocation method for the flying wing Unmanned Aerial Vehicle (UAV). Firstly, we apply the Kriging method to establish the aerodynamic model of the morphing trailing-edge, with the initial sample points generated by the non-uniform optimal Latin Hypercube Sampling (LHS). Then, based on the Kriging model, the multi-objective control allocation problem is converted into a standard optimization form. To solve such problem, we design a Comprehensive Multi-Objective Particle Swarm Optimization (C-MOPSO) algorithm and an improved Hierarchical MOPSO (H-MOPSO) algorithm, in which multiple optimization objectives are prioritized and hierarchically optimized using the PSO algorithm. As for performance analysis, an attitude angle tracking flight control system is established to validate the effectiveness of our proposed control allocation methods. Simulation results show that both the C-MOPSO and H-MOPSO methods have similar performance in attitude angle tracking, while H-MOPSO achieves better multi-objective allocation performance.INDEX TERMS Flying wing UAV, morphing trailing-edge, Kriging aerodynamic modeling, multiobjective control allocation, Particle Swarm Optimization.

show abstract

“…Herein we present a mixed variable multi-objective Bayesian optimisation algorithm and provide three test cases with comparison to both random sampling and a mixed variable version of NSGA-II provided in jMetalPy [26]. The algorithm looks to extend Bayesian multi-objective methodologies to the mixed variable domain, which to the best of the authors' knowledge has limited prior work [27]. Utilising the recently proposed Expected Improvement Matrix [28] and an adapted distance metric, the algorithm provides an efficient approach to optimising expensive to evaluate mixed variable multiobjective optimisation problems without the need for reparameterization of the discrete variables.…”

Section: Introductionmentioning

confidence: 99%

MVMOO: Mixed variable multi-objective optimisation

2021

View full text Add to dashboard Cite

In many real-world problems there is often the requirement to optimise multiple conflicting objectives in an efficient manner. In such problems there can be the requirement to optimise a mixture of continuous and discrete variables. Herein, we propose a new multi-objective algorithm capable of optimising both continuous and discrete bounded variables in an efficient manner. The algorithm utilises Gaussian processes as surrogates in combination with a novel distance metric based upon Gower similarity. The MVMOO algorithm was compared to an existing mixed variable implementation of NSGA-II and random sampling for three test problems. MVMOO shows competitive performance on all proposed problems with efficient data acquisition and approximation of the Pareto fronts for the selected test problems.

show abstract

Expected improvement for expensive optimization: a review

Cited by 105 publications

References 142 publications

Metamodel-based multidisciplinary design optimization methods for aerospace system

Metamodel-based multidisciplinary design optimization methods for aerospace system

Kriging Aerodynamic Modeling and Multi-Objective Control Allocation for Flying Wing UAVs With Morphing Trailing-Edge

MVMOO: Mixed variable multi-objective optimisation

Contact Info

Product

Resources

About