Distributed control for reducing carbon footprint in the residential sector

The current power grid is conservatively provisioned for rarely-occurring peaks. Expensive, quickly-ramping generators, typically with high carbon emissions, provide peak power. Therefore, it is possible to reduce both capital cost and carbon footprint by reducing the peak load. We address this issue by proposing to intelligently reduce loads from household appliances during peak times. Our scheme capitalizes on the fact that the power consumed by resistive loads can be reduced, at the cost of a small increase in appliance operation duration, with little impact on perceived user comfort. Specifically, on the receipt of congestion signals from the grid, appliance-based controllers intelligently reduce their load while ensuring that user comfort does not degrade below a pre-specified level. Simulations show that significant gains in energy reduction can be obtained with our scheme. For example, in Quebec, an estimated 12.9 MWh of peak power reduction can be obtained for a maximum of 10% increase in appliance operation duration.