Abstract: An improved multi-strategy sparrow search algorithm based on iterative chaotic mapping(IMSSA-ICM) is designed to address the problems of low accuracy and susceptibility to local optima in sparrow search algorithm.In the sparrow search algorithm,iterative chaotic mapping is used to initialize the sparrow population,and elite opposition-based learning and Levy flight strategy combined with roulette wheel selection mechanism are introduced to perturb the sparrow update position to optimize the traditional sparrow search algorithm's insufficient search accuracy and susceptibility to local extremum problems.On 12 benchmark test functions,IMSSA-ICM is compared with 5 classic swarm intelligence optimization algorithms.Simulation results show that IMSSA-ICM outperform the other 5 algorithms in terms of solving accuracy and escaping local optima.

Key words: sparrow search algorithm, iterative chaotic mapping, elite opposition-based learning, Levy flight