Multi-Agent based Dynamic Stability Control for Low-Frequency Global Mode of Oscillations

This paper presents a multi-agent based dynamic stability control of electric power systems especially for low-frequency global mode of oscillations. Different types of intelligent agents are also proposed to realize the proposed wide area stability control system: monitoring agents for gathering required information to evaluate the dynamic stability of the study system, control agents which perform the actual control action, and a supervisor agent for the real time monitoring of eigenvalue based dynamic stability and the decision of the required dynamic stability control action to keep the pre-specified dynamic stability margin. The supervisor agent sends commands to a selected unit to keep the stability margin within the pre-specified range, whenever the stability margin is violated in the study system. To demonstrate the efficiency of the proposed multi-agent based dynamic stability control system, real time non-linear simulations have been performed on the Analog Power System Simulator at the Research Laboratory of Kyushu Electric Power Co.