A comparative study of the multi-objective optimization algorithms for coal-fired boilers

被引：19

作者：

Wu, Feng ^{[1
]}

Zhou, Hao ^{[1
]}

Zhao, Jia-Pei ^{[1
]}

Cen, Ke-Fa ^{[1
]}

机构：

[1] Zhejiang Univ, Inst Thermal Power Engn, State Key Lab Clean Energy Utilizat, Hangzhou 310027, Zhejiang, Peoples R China

来源：

EXPERT SYSTEMS WITH APPLICATIONS | 2011年 / 38卷 / 06期

基金：

中国国家自然科学基金;

关键词：

Combustion; Multi-objective optimization; Support vector regression; SPEA2; OMOPSO; AbYSS; MOCell; SUPPORT VECTOR REGRESSION; EVOLUTIONARY ALGORITHMS; GENETIC ALGORITHM;

D O I：

10.1016/j.eswa.2010.12.042

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Combustion optimization has been proved to be an effective way to reduce the NOx emissions and unburned carbon in fly ash by carefully setting the operational parameters of boilers. However, there is a trade-off relationship between NOx emissions and the boiler economy, which could be expressed by Pareto solutions. The aim of this work is to achieve multi-objective optimization of the coal-fired boiler to obtain well distributed Pareto solutions. In this study, support vector regression (SVR) was employed to build NOx emissions and carbon burnout models. Thereafter, the improved Strength Pareto Evolutionary Algorithm (SPEA2), the new Multi-Objective Particle Swarm Optimizer (OMOPSO), the Archive-Based hybrid Scatter Search method (AbYSS), and the cellular genetic algorithm for multi-objective optimization (MOCell) were used for this purpose. The results show that the hybrid algorithms by combining SVR can obtain well distributed Pareto solutions for multi-objective optimization of the boiler. Comparison of various algorithms shows MOCell overwhelms the others in terms of the quality of solutions and convergence rate. (C) 2010 Elsevier Ltd. All rights reserved.

引用

页码：7179 / 7185

页数：7

共 30 条

[1] The exploration/exploitation tradeoff in dynamic cellular genetic algorithms [J].