论文部分内容阅读
提出了一种动态留数法,对含有分散非线性控制器的多机电力系统的主导模态进行动态扫描,根据分散非线性控制器安装前后系统主导模态的变化来确定系统中非线性控制器的最佳配置地点。当在线应用本方法时,建议构造一种基于多Agent的电力系统智能非线性控制的体系结构。它是一个由协调层和执行层组成的多Agent系统(Multi-AgentSystem),协调层Agent采用“动态留数”法确定系统中非线性控制器的最佳配置地点,执行层Agent负责非线性控制器参数的优化和控制器的切换。一个10机系统的离线仿真结果表明,这种方法能有效地消除不稳定零动态特性,提高多机电力系统的暂态稳定性。
A dynamic residue method is proposed to dynamically scan the dominant modes of a multi-machine power system with decentralized nonlinear controllers. Based on the changes of dominant modes before and after the decentralized nonlinear controller is installed, the nonlinear control The best place to configure your device. When applying this method online, it is suggested to construct an architecture of intelligent non-linear control of power system based on multi-agent. It is a multi-agent system composed of the coordination layer and the executive layer. The coordination layer Agent uses the “dynamic residue” method to determine the optimal configuration place of the nonlinear controller in the system. The executive layer Agent is responsible for the nonlinear control Optimization of the parameters of the controller and switching of the controller. The offline simulation results of a 10-machine system show that this method can effectively eliminate the unstable zero dynamic characteristics and improve the transient stability of multi-machine power system.