Research Article
A Novel Multi-objective Squirrel Search Algorithm: MOSSA
@INPROCEEDINGS{10.1007/978-3-030-72795-6_15, author={Xinyuan Wang and Fanhao Zhang and Zhuoran Liu and Changsheng Zhang and Qidong Zhao and Bin Zhang}, title={A Novel Multi-objective Squirrel Search Algorithm: MOSSA}, proceedings={Simulation Tools and Techniques. 12th EAI International Conference, SIMUtools 2020, Guiyang, China, August 28-29, 2020, Proceedings, Part II}, proceedings_a={SIMUTOOLS PART 2}, year={2021}, month={4}, keywords={Multi-objective optimization Non-dominated sorting Squirrel search algorithm Mapping fitness evaluation Roulette wheel selection}, doi={10.1007/978-3-030-72795-6_15} }- Xinyuan Wang
Fanhao Zhang
Zhuoran Liu
Changsheng Zhang
Qidong Zhao
Bin Zhang
Year: 2021
A Novel Multi-objective Squirrel Search Algorithm: MOSSA
SIMUTOOLS PART 2
Springer
DOI: 10.1007/978-3-030-72795-6_15
Abstract
This paper suggests a non-dominated sorting genetic algorithm II (NSGA-II) as a multi-objective framework to construct a multi-objective optimization algorithm and uses the squirrel search algorithm (SSA) as the core evolution strategy. And a multi-objective improved squirrel search algorithm (MOSSA) is proposed. MOSSA establishes an external archive of the population to maintain the elitists in the population. The probability density is applied to limit the size of the merged population to maintain population diversity, based on roulette wheel selection. Also, this paper designs a fitness mapping evaluation according to the individual fitness value of each object. Compared with the original SSA, the generational gap is introduced to make the seasonal condition suitable for multi-objective optimization, which could keep the solution from the local con-vergence. This paper simulates MOSSA and other algorithms on multi-objective test functions to analyze the convergence and diversity of PF. It is concluded that MOSSA has a good performance in solving multi-objective problems.
