Output details
11 - Computer Science and Informatics
University of Southampton
Agent-based control for decentralised demand side management in the smart grid
Significance of output:
<22>Describes the first model of agent-based demand-response algorithms for smart homes to control the schedule of deferrable devices and heating.
This work was nominated for best application award at the main Agents conference (AAMAS 2011) out of 575 submissions. This work pioneered the use of agent-based optimization and learning algorithms to control devices in the home and contributed to the development of an open-source smart home simulation framework.
The properties of the algorithm were validated empirically on data from 1000 homes and point to reductions in peak electricity demand of 17% and carbon emissions of up to 6%.