论文部分内容阅读
主要研究一类基于动态输出反馈的环型互联大系统鲁棒重叠分散控制问题.根据大系统包含原理的约束条件,对系统进行特殊重叠结构分解,在扩展空间中分解为一系列两两子系统对,为每个两两子系统对分别设计动态输出反馈控制器,然后收缩回原空间,实现对该互联大系统的重叠分散控制.在控制器设计中,采用了Lyapunov理论和线性矩阵不等式(LMI)方法,推导了系统稳定的充分条件,给出了动态输出反馈控制器的设计方法,并用遗传算法对设计参数进行了优化,从而进一步优化了设计结果.由于控制器设计时仅针对两两子系统,因此该方法避免了直接采用LMI方法因子系统过多,维数过大引起的LMI求解上的困难.最后将该方法应用到一个三区域互联电力系统的自动发电控制(AGC)设计中,仿真结果证明了此方法的可行性及优越性.
This paper mainly studies a class of robust overlapping decentralized control problems for ring-type interconnected large-scale systems based on dynamic output feedback.According to the constraint conditions of large-system inclusion principle, the system is decomposed into a series of two-to-two subsystems in the extended space For each pair of two subsystems, the dynamic output feedback controllers are respectively designed and then contracted back to the original space to achieve the superposition and decentralized control of the interconnected large-scale system.In the controller design, Lyapunov theory and linear matrix inequality LMI) method, the sufficient conditions of system stability are deduced, and the design method of dynamic output feedback controller is given, and the design parameters are optimized by genetic algorithm, so as to further optimize the design results.As the controller design only for two-two Therefore, this method avoids the difficulty of solving LMI caused by too much system dimension and too many dimensions by using LMI method directly.Finally, this method is applied to the AGC design of a three-zone interconnected power system The simulation results prove the feasibility and superiority of this method.