A Cognitive Power Grid Recovery Strategy in the Smart Grid Context
power grid, smart grid
The reconfiguration of electricity distribution networks has become quite attractive due to the process of automation and incorporation of electronic devices, which enable remote maneuvers. The reconfiguration process consists of changing the distribution network topology by closing or opening the interconnect switches. The distribution network reconfiguration process aims to support decision-making, planning and / or real-time control of grid operation aiming at minimizing active losses, load balancing and fault isolation and improving voltage levels. . The task of managing and making decisions to change the power grid topology is a very complex task due to the diversity of configuration possibilities. In this context, Autonomic Management Systems (AMS) are being investigated as a workable solution to the power grid reconfiguration problem. Thus, it is expected that human intervention in management can be replaced by autonomously generated solutions, preferably dynamically. This thesis proposes the use of Case-Based Reasoning (CBR) coupled with the HATSGA algorithm for the rapid reconfiguration of large power distribution networks. The suitability and scalability of the CBR-based reconfiguration strategy using the HATSGA algorithm are evaluated. Performance evaluation indicates that the HATSGA algorithm calculates new reconfiguration topologies with viable computational time for large network topologies. The CBR strategy looks for acceptable management reconfiguration solutions in the CBR database and, as such, contributes to reducing the required number of reconfiguration calculations using HATSGA. This suggests that CBR can be applied with a quick reconfiguration algorithm, resulting in a more efficient, dynamic and cognitive network recovery strategy.