An improved epsilon constraint-handling method in MOEA/D for CMOPs with large infeasible regions

被引：230

作者：

Fan, Zhun ^{[1
]}

Li, Wenji ^{[1
]}

Cai, Xinye ^{[2
]}

Huang, Han ^{[3
]}

Fang, Yi ^{[1
]}

You, Yugen ^{[1
]}

Mo, Jiajie ^{[2
]}

Wei, Caimin ^{[4
]}

Goodman, Erik ^{[5
]}

机构：

[1] Shantou Univ, Dept Elect Engn, Shantou 515063, Guangdong, Peoples R China

[2] Nanjing Univ Aeronaut & Astronaut, Coll Comp Sci & Technol, Nanjing 210016, Jiangsu, Peoples R China

[3] South China Univ Technol, Sch Software Engn, Guangzhou 515063, Guangdong, Peoples R China

[4] Shantou Univ, Dept Math, Shantou 515063, Guangdong, Peoples R China

[5] Michigan State Univ, BEACON Ctr Study Evolut Act, E Lansing, MI 48824 USA

来源：

SOFT COMPUTING | 2019年 / 23卷 / 23期

基金：

中国博士后科学基金; 中国国家自然科学基金;

关键词：

Constrained multi-objective evolutionary algorithms; Epsilon constraint handling; Constrained multi-objective optimization; Robot gripper optimization; REJECTIVE MULTIPLE TEST; EVOLUTIONARY ALGORITHMS; OPTIMIZATION; DECOMPOSITION; SELECTION; TESTS;

D O I：

10.1007/s00500-019-03794-x

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

This paper proposes an improved epsilon constraint-handling mechanism and combines it with a decomposition-based multi-objective evolutionary algorithm (MOEA/D) to solve constrained multi-objective optimization problems (CMOPs). The proposed constrained multi-objective evolutionary algorithm (CMOEA) is named MOEA/D-IEpsilon. It adjusts the epsilon level dynamically according to the ratio of feasible to total solutions in the current population. In order to evaluate the performance of MOEA/D-IEpsilon, a new set of CMOPs with two and three objectives is designed, having large infeasible regions (relative to the feasible regions), and they are called LIR-CMOPs. Then, the 14 benchmarks, including LIR-CMOP1-14, are used to test MOEA/D-IEpsilon and four other decomposition-based CMOEAs, including MOEA/D-Epsilon, MOEA/D-SR, MOEA/D-CDP and CMOEA/D. The experimental results indicate that MOEA/D-IEpsilon is significantly better than the other four CMOEAs on all of the test instances, which shows that MOEA/D-IEpsilon is more suitable for solving CMOPs with large infeasible regions. Furthermore, a real-world problem, namely the robot gripper optimization problem, is used to test the five CMOEAs. The experimental results demonstrate that MOEA/D-IEpsilon also outperforms the other four CMOEAs on this problem.

引用

页码：12491 / 12510

页数：20

共 42 条

[11] Exploration and Exploitation in Evolutionary Algorithms: A Survey [J].

Crepinsek, Matej ;

Liu, Shih-Hsi ;

Mernik, Marjan .

ACM COMPUTING SURVEYS, 2013, 45 (03)

[12]

Datta R, 2011, GECCO-2011: PROCEEDINGS OF THE 13TH ANNUAL GENETIC AND EVOLUTIONARY COMPUTATION CONFERENCE, P1843

[13] A fast and elitist multiobjective genetic algorithm: NSGA-II [J].

Deb, K ;

Pratap, A ;

Agarwal, S ;

Meyarivan, T .

IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION, 2002, 6 (02) :182-197

[14]

Deb K., 2001, MULTIOBJECTIVE OPTIM, DOI [10.1002/9780470496947, DOI 10.1002/9780470496947]

[15] An Evolutionary Many-Objective Optimization Algorithm Using Reference-Point-Based Nondominated Sorting Approach, Part I: Solving Problems With Box Constraints [J].

Deb, Kalyanmoy ;

Jain, Himanshu .

IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION, 2014, 18 (04) :577-601

[16] A practical tutorial on the use of nonparametric statistical tests as a methodology for comparing evolutionary and swarm intelligence algorithms [J].

Derrac, Joaquin ;

Garcia, Salvador ;

Molina, Daniel ;

Herrera, Francisco .

SWARM AND EVOLUTIONARY COMPUTATION, 2011, 1 (01) :3-18

[17] MULTIPLE COMPARISONS AMONG MEANS [J].

DUNN, OJ .

JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1961, 56 (293) :52-&

[18]

Fan Z., 2016, ARXIV161207603

[19] ON A MONOTONICITY PROBLEM IN STEP-DOWN MULTIPLE TEST PROCEDURES [J].

FINNER, H .

JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1993, 88 (423) :920-923

[20]

HOCHBERG Y, 1988, BIOMETRIKA, V75, P800, DOI 10.1093/biomet/75.4.800

← 1 2 3 4 5 →