神经元网络场电位的迭代学习控制研究

基于映射神经元模型和Hindmarsh-Rose神经元模型构建小世界神经网络，并施加带有遗忘因子的迭代学习控制算法，以实现神经网络的同步控制。仿真结果表明迭代学习控制同时适用于离散的和连续的神经网络模型，可以实现神经网络同步和去同步状态的相互转化，其优势在于随着迭代次数的增加，控制信号强度逐渐减弱，从而保持神经元本身的放电特性不变。所得结果为将非线性控制理论应用于帕金森等神经疾病控制提供了新思路。

Iterative learning control of field potential of neuronal network

This paper constructs two types of small-world network based on mapping neural model and Hindmarsh-Rose neuron model, respectively. The iterative learning control with forgetting factor is used to control the synchrony dynamics of the neuronal networks. Simulation results show the effectiveness of the iterative learning control method. It can realize the transition between synchronization and non-synchronization states of the neuronal network. The advantage of the control method is that the amplitude of control signal is largely decreased with the increase of iteration times and the firing properties of neurons keep unchanged. The obtained results provide a new way for the control of neural diseases, such as Parkinsion, with nonlinear control theories.