Enhanced variable-fidelity surrogate-based optimization framework by Gaussian process regression and fuzzy clustering

Tian, Kuo; Li, Zengcong; Huang, Lei; Du, Kaifan; Jiang, Liangliang; Wang, Bo

doi:10.1016/j.cma.2020.113045

Cited by 33 publications

(9 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition to the foregoing algorithms, other widely used infill sampling criteria include the minimization of surrogate prediction [87,88], probability of improvement [89,90], mean square error [91,92], lower confidence bounding [87][88][89][90][91][92][93][94][95], and fuzzy clustering-based criterion [96,97]. Among the different infill sampling criteria, expected improvement, probability of improvement, mean square error, and lower confidence bounding are generally combined with Kriging or other metamodels with error estimation capability [98,99].…”

Section: Mode-pursing Sampling Methodsmentioning

confidence: 99%

“…Among the different infill sampling criteria, expected improvement, probability of improvement, mean square error, and lower confidence bounding are generally combined with Kriging or other metamodels with error estimation capability [98,99]. Moreover, as an ensemble, all criteria can be used to improve robustness; this is called a multi-infill strategy [97,100,101].…”

Section: Mode-pursing Sampling Methodsmentioning

confidence: 99%

See 1 more Smart Citation

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

Section: Mode-pursing Sampling Methodsmentioning

confidence: 99%

Section: Mode-pursing Sampling 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

“…Now we demonstrate the DGS-ES method on a real-world stiffened shell design problem. Due to its high strength and stiffness, the hierarchical stiffened shell has been widely used in aerospace engineering [31,32]. However, it is challenging to fully explore its optimal buckling load-carrying capacity.…”

Section: Tests On Rl Benchmark Problemsmentioning

confidence: 99%

Accelerating reinforcement learning with a Directional-Gaussian-Smoothing evolution strategy

Zhang

Tran

Zhang

2021

ERA

View full text Add to dashboard Cite

<p style='text-indent:20px;'>The objective of reinforcement learning (RL) is to find an optimal strategy for solving a dynamical control problem. Evolution strategy (ES) has been shown great promise in many challenging reinforcement learning (RL) tasks, where the underlying dynamical system is only accessible as a black box such that adjoint methods cannot be used. However, existing ES methods have two limitations that hinder its applicability in RL. First, most existing methods rely on Monte Carlo based gradient estimators to generate search directions. Due to low accuracy of Monte Carlo estimators, the RL training suffers from slow convergence and requires more iterations to reach the optimal solution. Second, the landscape of the reward function can be deceptive and may contain many local maxima, causing ES algorithms to prematurely converge and be unable to explore other parts of the parameter space with potentially greater rewards. In this work, we employ a Directional Gaussian Smoothing Evolutionary Strategy (DGS-ES) to accelerate RL training, which is well-suited to address these two challenges with its ability to (i) provide gradient estimates with high accuracy, and (ii) find nonlocal search direction which lays stress on large-scale variation of the reward function and disregards local fluctuation. Through several benchmark RL tasks demonstrated herein, we show that the DGS-ES method is highly scalable, possesses superior wall-clock time, and achieves competitive reward scores to other popular policy gradient and ES approaches.</p>

show abstract

“…Compared with K-means and other "hard clustering" algorithms [20][21][22], the Gaussian fuzzy clustering belongs to the "soft clustering" methods. e Gaussian fuzzy clustering is more flexible [23][24][25] and can determine the clustering members according to the probability of obtaining a better clustering effect. erefore, this paper proposes a guide selection method based on the Gaussian fuzzy clustering and constructs a new guide participation social force evacuation model based on fuzzy theory.…”

Section: Introductionmentioning

confidence: 99%

A Social Force Evacuation Model with Guides Based on Fuzzy Clustering and a Two-Layer Fuzzy Inference

Xiao

2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

Current emergency management research mainly specifies the positions of evacuation guides from a knowledge base of experience, disregarding the subjective perceived decision-making of pedestrians caught in an emergency situation. Therefore, in this paper, a fuzzy inference system for pedestrians to select guides is designed from the perspective of pedestrians, and a crowd evacuation model with guides under limited vision is constructed. First, selecting the indoor evacuation of people with limited vision as the context, the number and optimal initial positions of guides are determined by a Gaussian fuzzy clustering algorithm. Next, a two-layer fuzzy inference system based on a multifactor pedestrian selection guide is established. Then, from the comprehensive perspective of managers and pedestrians, an improved social force evacuation model with guides is constructed. A comparison of the evacuation times and evacuation processes of known methods with different scene population distributions is analyzed through simulations. The results show that the guide setting scheme of the improved model is more conducive to reducing evacuation times and balancing exit utilizations. The model can provide a basis for emergency management decision-making departments to formulate more flexible guidance strategies.

show abstract

Enhanced variable-fidelity surrogate-based optimization framework by Gaussian process regression and fuzzy clustering

Cited by 33 publications

References 38 publications

Metamodel-based multidisciplinary design optimization methods for aerospace system

Metamodel-based multidisciplinary design optimization methods for aerospace system

Accelerating reinforcement learning with a Directional-Gaussian-Smoothing evolution strategy

A Social Force Evacuation Model with Guides Based on Fuzzy Clustering and a Two-Layer Fuzzy Inference

Contact Info

Product

Resources

About