Design of Mutation Operator Based on Information Entropy

This paper analyze the traditional mutation operator of GAs in design idea of mutation operator,and show that the design idea has some disadvantages. That is, the design idea of mutation operator that is stochastically independent and occurs with fixed probability is not perfect. then, mutation operator based on information entropy is presented to take the place of the traditional one. The function of mutation operator based on information entropy to prevent premature convergence is also discussed. Using new mutation operator to solve the typical function optimization problem, the experimental results show that the new GAs can converge quickly and prevent the premature convergence effectively. This shows that the design idea is validity.