Purpose

This paper aims to focus on the design of a decentralized observation and control method for a class of large-scale systems characterized by nonlinear interconnected functions that are assumed to be uncertain but quadratically bounded.

Design/methodology/approach

Sufficient conditions, under which the designed control scheme can achieve the asymptotic stabilization of the augmented system, are developed within the Lyapunov theory in the framework of linear matrix inequalities (LMIs).

Findings

The derived LMIs are formulated under the form of an optimization problem whose resolution allows the concurrent computation of the decentralized control and observation gains and the maximization of the nonlinearity coverage tolerated by the system without becoming unstable. The reliable performances of the designed control scheme, compared to a distinguished decentralized guaranteed cost control strategy issued from the literature, are demonstrated by numerical simulations on an extensive application of a three-generator infinite bus power system.

Originality/value

The developed optimization problem subject to LMI constraints is efficiently solved by a one-step procedure to analyze the asymptotic stability and to synthesize all the control and observation parameters. Therefore, such a procedure enables to cope with the conservatism and suboptimal solutions procreated by optimization problems based on iterative algorithms with multi-step procedures usually used in the problem of dynamic output feedback decentralized control of nonlinear interconnected systems.

中文翻译：

---

非线性大规模互联系统的动态输出反馈控制

目的

本文旨在针对一类大型系统的分散观测和控制方法的设计，该系统具有非线性互连函数，这些函数被确定为不确定但具有二次边界。

设计/方法/方法

在线性矩阵不等式（LMI）的框架内的Lyapunov理论中，开发了充分的条件，在此条件下设计的控制方案可以实现增强系统的渐近稳定。

发现

得出的LMI以优化问题的形式制定，其分辨率允许并发计算分散的控制和观测增益，并且系统可以容忍的非线性覆盖范围最大化而不会变得不稳定。与在文献中发表的杰出的分散式保证成本控制策略相比，所设计的控制方案的可靠性能在三发电机无限母线电力系统的广泛应用中通过数值模拟得到了证明。

创意/价值

通过一步步骤分析渐近稳定性并综合所有控制和观测参数，可以有效解决受LMI约束的已开发优化问题。因此，这样的过程使得能够应对基于迭代算法的优化问题所产生的保守性和次优解，其中该迭代算法具有通常用于非线性互连系统的动态输出反馈分散控制问题的多步骤过程。