Evolutionary Computation with Biogeography-based Optimization

12Hybrid BBO Algorithms

Hybrid evolutionary algorithms (EAs) are attractive alternatives to standard EAs. The combination of several algorithms in hybrid EAs allows it to exploit the strength of each algorithm. It has been shown that by properly selecting the constituent algorithms and hybridization strategies, hybrid EAs can outperform their constituent algorithms due to their synergy. This characteristic is strong motivation for the study of hybrid EAs. Many hybrid EAs have been proposed to improve performance and to find global optima. Although some of these improvements are significant, the development of new hybrid EAs and strategies is worthy of further investigation.

Current research directions in hybrid EAs involve several major areas. The first area is the determination of how to hybridize a given set of EAs into a single algorithm; that is, how to determine the hybridization strategy. The second area is the determination of which EAs to combine in a hybrid algorithm. The third area is the application of hybrid EAs to special types of optimization problems, such as constrained optimization and multi-objective optimization. The fourth area is the application of hybrid EAs to real-world optimization problems. The goal of this chapter is to address the first and second areas; that is, we emphasize the mechanism of hybridization to improve the optimization performance of EAs.

Overview of the chapter

In this chapter, we propose several hybrid EAs by combining some popular ...