In the complex conversion analysis of multi-degrees of freedom，large calculation count needed in each calculation step of Genetic Algorithm limits the running speed of genetic algorithm. So the positive calculation count to be reduced is an effective method to enlarge the range of GA’s application. Empirical Genetic-Simplex Algorithm (EGSA) proposed in this paper is one of the effective methods to solve the problem. This method is applied to structural physical parameters identification based on the structural dynamic response. The result shows that EGSA has many advantages on precision, efficiency in searching, strong to resist the noise, and good adaptation to the incomplete information.