Two-population Shuffled Frog Leaping Algorithm Based on Bio-parasitic Behavior

摘要

A two-population shuffled frog leaping algorithm based on bio-parasitic behavior is proposed to solve the problems of the original algorithm, such as non-uniform initial population, low search precision and easiness in falling into local optimum. It consists of the host and the parasite population. The mechanism of facultative parasitic behavior is incorporated into the algorithm. The two populations exchange particles according to individuals fitness value at intervals of a certain number of iterations, so as to realize the parallel evolution and cooperative search. The algorithm constructs the initial population with chaos theory. Furthermore, an adaptive factor is designed to adjust the moving step of the frogs. Experimental results show that the new algorithm has a better convergence result and higher solution accuracy.