Abstract: In immune optimization algorithm with multi-learning mechanism,the selection probability of mutation operators plays a vital role in the effectiveness of the algorithm．If the choice is not appropriate,the algorithm is easy to fall into local optimum,to a certain extent,affects the quality of the solution and reduces the rate of convergence．According to the characteristics of the multi-learning mechanism,the dependence and correlation between mutation operators were discussed．A new mutation operator selection strategy was proposed and four benchmark functions were selected as test functions to verify the performance of the strategy．The result shows that the quality of the solution and the convergence speed have obvious improvement.

Key words: Multi-learning mechanism,Gaussian mutation,Cauchy mutation,Lateral mutation,Baldwinian mutation