Abstract:

This paper proposed an artificial fish school algorithm (AFSA) based on adaptive t distribution mixed mutation, aiming at some defects of traditional AFSAs，such as low optimization precision and long running time. Gauss optimization adjustment mutation is executed on the excellent fish. Adaptive t distribution mutation is executed on the no excellent fish. In this algorithm, a new mutation operator following the t distribution is used to integrate the advantages of Gaussian and Cauchy mutation. Emulation test of four representative test functions, many application examples and a large number of test data show that this algorithm is more reliable and efficient. And this algorithm is better than ACMAFSA algorithm and AGMAFSA algorithm in the stability aspect and precision aspect.           

Key words: artificial fishschool algorithm(AFSA)； adaptive t distribution mutation；Gauss optimization adjustment mutation