2005
DOI: 10.1109/tsmcc.2004.841906
|View full text |Cite
|
Sign up to set email alerts
|

Search Biases in Constrained Evolutionary Optimization

Abstract: Abstract-A common approach to constraint handling in evolutionary optimization is to apply a penalty function to bias the search towards a feasible solution. It has been proposed that the subjective setting of various penalty parameters can be avoided using a multi-objective formulation. This paper analyses and explains in depth why and when the multi-objective approach to constraint handling is expected to work or fail. Furthermore, an improved evolutionary algorithm based on evolution strategies and differen… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

3
233
0
1

Year Published

2010
2010
2024
2024

Publication Types

Select...
5
2
1

Relationship

0
8

Authors

Journals

citations
Cited by 414 publications
(237 citation statements)
references
References 19 publications
3
233
0
1
Order By: Relevance
“…Algorithm 3 with a population size, π = 50 and allowed a maximum of 250 iterations. Following [99,100], the probability factor, η used in SR was set to 0.45. Table 7 reports the mean,median and standard deviations (SD) of the number of function evaluations (NFE) over the 30 independent runs of DEBLP-SR to meet the convergence criteria (i.e.…”
Section: Discussionmentioning
confidence: 99%
See 1 more Smart Citation
“…Algorithm 3 with a population size, π = 50 and allowed a maximum of 250 iterations. Following [99,100], the probability factor, η used in SR was set to 0.45. Table 7 reports the mean,median and standard deviations (SD) of the number of function evaluations (NFE) over the 30 independent runs of DEBLP-SR to meet the convergence criteria (i.e.…”
Section: Discussionmentioning
confidence: 99%
“…Runarsson and Yao [99,100] proposed an alternative constraint handling method known as stochastic ranking (SR) 7 to aid in answering this question. To use SR, the first step is to obtain a measure of the constraint violation, v(x k ), of vector x k using Eqn.…”
Section: Stochastic Rankingmentioning
confidence: 99%
“…IVRA was used to reduce the search space and thereafter optimizers were used within this reduced space to find the optima. The optimization algorithms tested were Controlled Random Search (CRS) with local mutation (Kaelo and Ali 2006) and Improved Stochastic Ranking Evolution Strategy (ISRES) (Runarsson and Yao 2005).…”
Section: Methodsmentioning
confidence: 99%
“…In this paper, the upper level constraints in Eq 2 are handled using stochastic ranking [20]. Hence this version of DEBLP is termed DEBLP-SR.…”
Section: Differential Evolution For Bi-level Programmingmentioning
confidence: 99%
“…(See [20] for more details). In the first iteration (line 9) the best member of the population is the member that is assigned a rank of 1 (one) by the SR algorithm.…”
Section: Differential Evolution For Bi-level Programmingmentioning
confidence: 99%