MicroGrids Dynamic Security Assessment

Abstract

MicroGrids comprise low voltage distribution systems with distributed energy sources, storage devices and controllable loads, operated interconnected with the main power grid or autonomously, in a controlled coordinated way. Storage devices are absolutely essential for MicroGrid autonomous operation. However, depending on the operating conditions, their limited storage capacity is a major drawback and might compromise successful MicroGrid islanded operation. In this paper is proposed an index to evaluate MicroGrids security regarding an unplanned transition to islanded operation due to disturbances in the upstream Medium Voltage network. An Artificial Neural Network tool is proposed to forecast the security index for a credible range of operating scenarios. The usage of Artificial Neural Networks is emphasized due to its computational speed for on-line performance and its flexibility for providing corrective actions for insecure operating conditions in order to achieve a seamless transition from interconnected to islanded operation.