The Modified Shuffled Frog Leapping Algorithm and Dynamic Behavior Analysis of Local Search Process

Shuffled frog leaping algorithm (SFLA) is one of promising optimistic methods which are based on swarm intelligence. SFLA combines the advantages of memetic algorithm and particle swarm optimization. It has been widely used to solve complex optimization problems in engineering fields. To enhance the convergence and effectiveness of local search process in SFLA, a novel individual update method is presented and the dynamic behaviors of local search are analyzed in detail. On the basis of the proposed approach, the modified SFLA is presented. Simulation experimental results show that the modified SFLA achieves more efficiency than the classical shuffled frog leaping algorithm.