Pseudo-parallel chaotic self-learning antelope migration algorithm based on mobility models

摘要

Self-learning Antelopes Migration Algorithm (SAMA) is a self-heuristic algorithm that simulates the local exploitation of ordinary antelopes and the global exploration of scout antelopes. Aiming at the imbalance between exploitation and exploration of SAMA, it is easy to fall into local optimum, and the convergence speed and precision will be affected. A pseudo-parallel chaotic self-learning antelope migration algorithm based on mobility model is proposed. Nine chaotic self-learning antelope migration algorithms are formed by incorporating nine kinds of chaotic local searching operators, and the total population is divided into several sub-populations through meme grouping, and the internal circulation optimization is carried out by using nine chaotic SAMAs. After the internal cycle is completed, the five migration models will perform their migration operations and mutation operations to form the pseudo-parallel chaotic SAMA to increase the diversity of the population, improve the optimization accuracy and the ability of the algorithm to balance exploitation and exploration. Three simulation experiments are carried out to verify the effectiveness of the proposed algorithm. Firstly, the chaotic SAMA and the pseudo-parallel chaotic SAMA based on the mobility models are used to optimize 17 benchmark test functions. Secondly, 25 test functions in CEC-BC-2017 are optimized respectively. Finally, the four engineering design problems are optimized, including three-bar truss design, welded beam design, pressure vessel design and spring design problems. Experimental results show that the improved algorithm can better solve the function optimization and engineering optimization problems. The pseudo-parallel chaotic SAMA based on the mobility models has the advantage of balancing exploitation and exploration in the optimization process, and improves the convergence accuracy.