Sensing in power distribution networks via large numbers of smart meters

The electricity grid is undergoing significant changes as it paves its way towards a smarter grid. New technologies provide a much clearer view on the electricity grid and its processes, mainly due to the fine-grained and near real-time acquisition of data as well as assessment of it. The huge amount of information will include additional technical data beyond the traditional metering information for billing purposes. This may give rise to a new generation of tools that rigorously monitor residential endpoints. Such monitoring may assist towards rapidly identifying problems and decrease the time to resolve them. Currently, a real-time view of the grid state is done by specialized equipment, but their costs combined with their need to be widely deployed, is a limiting factor. Today's modern communication approaches coupled with high-performance analytics can be used in complex system monitoring and assessment. We argue how a large network of distributed smart meters can be used for analytics of the smart grid, based on our experiences in developing a smart metering platform that is currently monitoring approximately 5000 real smart meters. We analyse and discuss the results stemming from the data being streamed to the platform, and show how new insights can be obtained with respect to power quality.