Abstract: In the process of inverting evaporation duct profile from radar sea clutter, it presents a strong ability to grasp the overall situation but poor local search ability when using the Genetic Algorithms. On the contrary, simulated annealing algorithm has strong local search ability, but poor global search capability, and the computing efficiency is not high. In this paper, we combine the two algorithms, and then analyze the parameter selection and anti-noise performance of the genetic/simulated annealing hybrid algorithm. Finally, it is emulated according to the analog sea clutter power. The results show that the genetic/simulated annealing hybrid algorithm can complement each other, and can be applied to solve the inversion problem of evaporation duct profile in a certain noise level.