本文根据神经网络在非线性函数中的逼近能力,发挥遗传算法(GA)全局寻优和BP算法局部收敛结合的优势,将GA—BP算法应用于神经网络,完成系统的非线性辨识功能。仿真结果表明:基于GA—BP算法的神经网络具有良好的非线性逼近能力和泛化能力,具有较高的系统辨识精度。 Based on the approximation ability of neural network in nonlinear function, this paper takes advantage of the combination of global optimization of genetic algorithm (GA) and local convergence of BP algorithm, and applies GA-BP algorithm to neural network to complete the system’s nonlinear identification function. The simulation results show that the neural network based on GA-BP algorithm has good nonlinear approximation ability and generalization ability, and has high system identification precision.