Surrogate-Assisted Evolutionary Optimization of Large Problems

Chugh, Tinkle; Sun, Chaoli; Wang, Handing; Jin, Yaochu

doi:10.1007/978-3-030-18764-4_8

Cited by 24 publications

(16 citation statements)

References 54 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These algorithms are discussed in the following sections. Many optimization techniques based on bioinspired processes and Surrogate-Assisted natural methods have been formulated, and mathematical and statistical analysis of these algorithms has been performed [49][50][51][52]. The comparative performance study based on convergence trends was performed with different geometry parameterization techniques.…”

Section: Literature Surveymentioning

confidence: 99%

Aerodynamic Design Optimization of a Morphing Leading Edge and Trailing Edge Airfoil–Application on the UAS-S45

et al. 2021

View full text Add to dashboard Cite

This work presents an aerodynamic optimization method for a Droop Nose Leading Edge (DNLE) and Morphing Trailing Edge (MTE) of a UAS-S45 root airfoil by using Bezier-PARSEC parameterization. The method is performed using a hybrid optimization technique based on a Particle Swarm Optimization (PSO) algorithm combined with a Pattern Search algorithm. This is needed to provide an efficient exploitation of the potential configurations obtained by the PSO algorithm. The drag minimization and the endurance maximization were investigated for these configurations individually as two single-objective optimization functions. The aerodynamic calculations in the optimization framework were performed using the XFOIL solver with flow transition estimation criteria, and these results were next validated with a Computational Fluid Dynamics solver using the Transition Shear Stress Transport (SST) turbulence model. The optimization was conducted at different flight conditions. Both the DNLE and MTE optimized airfoils showed a significant improvement in the overall aerodynamic performance, and MTE airfoils increased the efficiency of CL3/2/CD by 10.25%, indicating better endurance performance. Therefore, both DNLE and MTE configurations show promising results in enhancing the aerodynamic efficiency of the UAS-S45 airfoil.

show abstract

Section: Literature Surveymentioning

confidence: 99%

Aerodynamic Design Optimization of a Morphing Leading Edge and Trailing Edge Airfoil–Application on the UAS-S45

et al. 2021

View full text Add to dashboard Cite

show abstract

“…First, RBFNN is a fast, computationally efficient, and easy-to-implement method for approximation tasks [15], [18], [67]. Second, RBFNNs have been widely used as surrogates in [12], [15], and [17], which are the compared algorithms in this article, and, therefore, using RBFNNs in BDDEA-LDG can help achieve fair comparisons. The settings of all RBFNNs in BDDEA-LDG are configured the same as those in DDEA-SE [15], so that their comparisons can be fair.…”

Section: A Experimental Setupmentioning

confidence: 99%

“…C min ≤ C ≤ C max (8) 0 ≤ θ z ≤ C ∀z ∈ Z (9) g min ≤ g z,i ≤ g max ∀z ∈ Z ∀i ∈ I (10) g z,1 + g z,1 = g z,5 + g z,6 ∀z ∈ Z (11) g z,3 + g z,4 = g z,7 + g z,8 ∀z ∈ Z (12) where OF is the objective function of three decision variables including cycle period (C), green splits (g), and offsets (θ ), Z is the intersection set (each Z has 8 g), I is the signal set of an intersection (containing green, yellow, and red signals), and C max and C min are maximum and minimum cycle length for a complete period, g max and g min are the maximum and minimum of a green splits, respectively. Equations (11) and (12) are for the ring-barrier diagram strategy such that the east-west and north-south movements will not contradict each other. The signal timing problem used in this article is a road with four intersections (Z = 4), both of which are T-junctions, as shown in Fig.…”

Section: Arterial Traffic Signal Timing Optimizationmentioning

confidence: 99%

Boosting Data-Driven Evolutionary Algorithm With Localized Data Generation

Zhan

Wang

et al. 2020

IEEE Trans. Evol. Computat.

119

View full text Add to dashboard Cite

By efficiently building and exploiting surrogates, data-driven evolutionary algorithms (DDEAs) can be very helpful in solving expensive and computationally intensive problems. However, they still often suffer from two difficulties. First, many existing methods for building a single ad hoc surrogate are suitable for some special problems but may not work well on some other problems. Second, the optimization accuracy of DDEAs deteriorates if available data are not enough for building accurate surrogates, which is common in expensive optimization problems. To this end, this article proposes a novel DDEA with two efficient components. First, a boosting strategy (BS) is proposed for self-aware model managements, which can iteratively build and combine surrogates to obtain suitable surrogate models for different problems. Second, a localized data generation (LDG) method is proposed to generate synthetic data to alleviate data shortage and increase data quantity, which is achieved by approximating fitness through data positions. By integrating the BS and the LDG, the BDDEA-LDG algorithm is able to improve model accuracy and data quantity at the same time automatically according to the problems at hand. Besides, a tradeoff is empirically considered to strike a better balance between the effectiveness of surrogates and the time cost for building them. The experimental results show that the proposed BDDEA-LDG algorithm can generally outperform both traditional methods without surrogates and other state-of-the-art DDEA son widely used benchmarks and an arterial traffic signal timing real-world optimization problem. Furthermore, the proposed BDDEA-LDG algorithm can use only about 2% computational budgets of traditional methods for producing competitive results.

show abstract

“…Surrogate-based methods allow us to reduce the computational cost of search algorithms, as opposed to parallelisation techniques that merely focus on reducing processing time. This kind of approach has been widely investigated in various problems employing evolutionary optimisation techniques [40,41,42,43],…”

Section: Introductionmentioning

confidence: 99%