基于牵制控制的HR生物神经网络的线性自抗扰同步

Pinning Control for Synchronization of HR Biological Neural Networks: Linear Active Disturbance Rejection Approach

考虑以Hindmarsh-Rose生物神经元为节点的复杂网络同步问题。鉴于同步算法通常调节节点间的耦合强度,而耦合强度可能无法改变;此外,同步受到各种干扰的影响,基于牵制控制策略,设计线性自抗扰算法,依据受控节点的同步偏差以及网络节点间的耦合实现网络同步。仿真中,固定耦合强度为较小的数值,分别获得了干扰不存在和存在时的网络同步。结果表明,线性自抗扰控制能够实时估计和补偿干扰,保证网络同步性能,避免了同步算法过于依赖网络模型、抗扰性能不佳的不足。

Synchronization of a complex network whose nodes are Hindmarsh-Rose biological neurons was considered. Coupling strengths of the whole network were always taken as the tuning variables. However, coupling strengths may not be changeable. In addition, synchronization was affected by different disturbances. In view of above factors, pinning control was utilized and linear active disturbance rejection control(LADRC) was designed. By this approach, synchronization was achieved in according to synchronization errors of the controlled nodes and the couplings among nodes. In the simulations, coupling strength was fixed at a relative small value, and no external disturbance case and external disturbance existing case were considered. Simulation results confirm that LADRC is able to estimate and compensate disturbances in real time, which guarantee nice performance of synchronization. Compared with other reported synchronization approaches, LADRC is less dependent of accurate model of networks, and robust enough to disturbances.