Classification and regression tree-based adaptive damping control of inter-area oscillations using wide-area signals

An adaptive damping control scheme based on the classification and regression tree (CART) using wide-area signals is proposed. A family of robust multiple-input multiple-output controllers are designed offline and used in real-time after the current operating point is retrieved from phasor measurement unit (PMU) data with CART interpolation. When the power system is operating close to a previously set operating point, only the corresponding controller is active. When the power system is not close to any previously set operating point, a combined controller is formed. A 16-machine 68-bus system is used for simulation test. Thyristor controlled series compensation, static var compensator and energy storage device are used as actuators and the remote PMUs frequencies are employed as added inputs in the example control. Simulation results demonstrate good damping performance of the proposed control method against a wide range of changes in operating state.