节点文献
非线性大系统直接自适应分散控制的神经网络逆系统方法
NEURAL NETWORK BASED INVERSE SYSTEM METHODAPPLIED TO THE DIRECT ADAPTIVE DECENTRALIZED CONTROLOF NONLINEAR LARGE SCALE SYSTEM
【摘要】 针对一类带有不确知关联的非线性大系统,提出了一种新型分散控制方案:以隔离子系统取代原子系统并求其逆系统,从而产生原子系统控制量的一部分;控制量的另一部分由用于补偿误差的在线神经网络形成。此方案的收敛性得到了论证,并在双倒立摆所形成的大系统中得到成功应用。
【Abstract】 A new decentralized control scheme is presented to deal a kind of interconnected uncertain large scale systems. The control signal includes two terms: one is obtained by inverting the separated subsystem instead of the origin subsystem; the other is an adaptive term to compensate the error introduced by the referred method. The convergence of the algorithm is verified by the application to a double inverted pendulum system.
【关键词】 非线性大系统;
分散控制;
神经网络;
自适应;
【Key words】 nonlinear large scale system; decentralized control; neural network; adaptive;
【Key words】 nonlinear large scale system; decentralized control; neural network; adaptive;
【基金】 国家自然科学丛金;航空科学基金资助项自
- 【文献出处】 飞机设计 ,Aircraft Design , 编辑部邮箱 ,2001年04期
- 【分类号】TP18
- 【下载频次】98