Evaluating a MAS architecture for flexible distribution power flow management

The challenges associated with controlling power networks are increasing due to integration of renewables, the potential use of storage technologies, and new dynamics introduced via technologies such as electric vehicles and demand side management. Active Network Management (ANM) is a key area to ensure that voltage and thermal limits are not breached. There has been a drive to explore flexible and extensible approaches to delivering enhanced active network management. One suitable approach is Multi-Agent System (MAS) technology that can provide a distributed solution to Smart Grid control. This paper focuses on the design and testing of a MAS able to deliver the ANM function, covering thermal constraint management as an initial application with the ability to integrate voltage control algorithms in the future. It is used to determine whether intelligent agents are the key to addressing these problems in ways that are flexible, extensible and robust. This system is built on the Presage2 MAS platform, which offers enhanced features and functionality to support the development of MAS for network control, and case studies of thermal overload management are provided for two different distribution networks.