A Shared Control Approach for Muti-Area Interconnected Power System via Operational Behaviors Learning

In this study, our focus lies on the issue of optimal load frequency control (LFC) in a multi-area interconnected power system. A shared control strategy that incorporates knowledge of human operational behavior is proposed. Given the difficulty in obtaining analytical models of operators, a Gaussian process regression is introduced to construct a data-driven model for operator behaviors. To improve the performance of the multi- area interconnected power system, a shared control approach with an operational model is presented, which integrates grid state information and operator instructions. Specifically, a GP-MPC is designed to predict the system states by propagating uncertainty and obtain an optimal control profile within the prediction horizon. Several experiments are conducted to validate the efficiency of the proposed shared control approach.