论文部分内容阅读
对一类具有未建模动态结构相似形的严格反馈非线性互联大系统,提出一种基于神经网络的分散自适应动态面控制方案.该方案引入Lyapunov函数来约束未建模动态,利用神经网络逼近理论分析中所产生的未知非线性连续函数.通过Young’s不等式和三重求和项的分解,有效地处理了耦合作用项,并利用动态面控制技术,实现了系统的分散控制.与现有研究结果相比,所设计的分散控制律中不含有控制增益下界常数.通过构造的方法,利用动态面控制设计中引入的紧集有效地处理了未建模动态和分析中产生的不确定连续函数.理论分析证明了闭环控制系统中所有信号半全局一致终结有界,且跟踪误差收敛到原点的一个小邻域内.两个数值算例的仿真结果表明所提控制方案的有效性.
For a class of strict feedback nonlinear interconnected systems with unmodeled dynamic structure similarity, this paper proposes a decentralized adaptive dynamic surface control scheme based on neural network. This scheme introduces Lyapunov function to constrain unmodeled dynamics, and uses neural network Approximates the unknown nonlinear continuous function generated in the theoretical analysis. Through the decomposition of Young’s inequality and triple summation term, the coupling action term is effectively processed and the decentralized control of the system is realized by dynamic surface control technology. Compared with the existing research Compared with the results obtained, the decentralized control law does not contain the lower bound of the control gain.By the constructed method, the tight set introduced in the dynamic surface control design is used to deal with the un-modeled dynamics and the uncertain continuous functions The theoretical analysis proves that all the signals in the closed - loop control system are semi - globally congruent, and the tracking error converges to a small neighborhood of the origin.The simulation results of two numerical examples show the effectiveness of the proposed control scheme.