论文部分内容阅读
为提高神经网络对未知非线性大滞后动态系统的泛化能力,提出一种基于高斯微粒群优化的自适应动态前馈神经网络.在输入层与隐含层之间、隐含层与输出层之间分别加入动态延迟算子,可以高效地辨识出系统纯滞后时间,建立精确系统模型.此外,采用高斯函数和混沌映射方法平衡微粒群算法全局寻优能力,以克服提前收敛的缺陷,从而快速有效地自适应优化网络中的参数.仿真实验表明了该方法在非线性大滞后系统辨识中的有效性.
In order to improve the generalization ability of neural networks for unknown large lag dynamic systems, an adaptive dynamic feedforward neural network based on Gaussian particle swarm optimization is proposed. Between the input layer and the hidden layer, the hidden layer and the output layer , The dynamic delay operator can be added separately to effectively identify the system lag time and establish an accurate system model.In addition, Gaussian function and chaos mapping are used to balance the global optimization ability of particle swarm optimization to overcome the shortcomings of early convergence The parameters in the network are adaptively optimized quickly and effectively.The simulation results show the effectiveness of the proposed method in the identification of nonlinear large lag systems.