An improved genetic algorithm(GA)-Simulated Annealing Genetic Algorithms-is proposed by combining the thoughts of Simulated Annealing Algorithms and Genetic Algorithms to solve the problem of the parameter optimization of PID controller．The new GA has been improvemed in algorithm by the use of adaptive crossover，mutation operators，and scaling or stretching．This algorithm avoids premature convergence，and has a quick convergence performance．Besides，it has both the capability of optimization in the global area as well as in the local area．The simulation results shows that the PID controller designed upon the improved GA is of good dynamic quality and stability．