Abstract: This paper presents a kind of adaptive neural network(NN) control method based on genetic algorithms(GA). In order to overcome the defects such as slow convergent speed and unsatisfied dynamic character when backpropagation(BP) is used to train the NN control system,the method using a modified GA to train parameters of NN is proposed to improve the performance of the control system. Simulation shows that the controller has fine accuracy,robustness and dynamic character when it is used to control nonlinear,time variable and time delayed systems.