What Can IoT Learn From Smart Grid?
Silver Springs Networks, October 30, 2015. Image credit: Kayaker~commonswiki
Data Data Data
Day-to-day operations for electricity networks are generally run through centralised network operations centres with powerful software systems seamlessly linking monitoring, communications and control functionality together. Smart grid applications have taken this further with additional network monitoring reaching out to the furthest branches of the system. The consolidated control ethos naturally led to smart grid data being gathered back to a centralised location. The problem has then been, what do we do with all that data?
This has led to two notable trends in smart grid data applications. Firstly DNOs have begun to invest in specialist analytics tools that draw on the expertise of third party developed algorithms to unpack the bits and bytes and reveal hidden truths about the system. Confidence in these outputs and smart grid tools has resulted in a paring back of data gathered, meaning only the essential data sets are being collated and stored.
Secondly there is an emerging trend towards distributed intelligence. Processing data in the field reduces the burden on communications infrastructure and database storage as mundane data is discarded and important events flagged. A good example of this is the use of teaming in remote control switching where devices are able to work together to make joint decisions to restore electricity supplies following a fault. No centralised control system is required and subsequently power can be restored much faster.
Similar trends are also appearing in smart cities, as organisations used to managing static infrastructure begin to delve into the world of analytics. Cities are beginning to collect vast amounts of records and huge data stores are taking shape. However, there is a subtly different approach being applied by some organisations, with some data sets being made publically available through open data platforms. It is hoped that this will lead to innovative applications drawing on a wider creativity pool, and delivering further benefits.
So smart grid projects have shown that a maturity regarding data can be achieved through application of analytics, thus informing what data is essential. The application of distributed intelligence improves this efficiency further reducing the reliance on human intervention. For cities, a wider pool of analytics resources will be available to help understand the data, but the learning should be applied to inform further decisions about data gathering and where it is ultimately consumed. Without this feedback loop, smart cities infrastructure risk becoming bogged down in superfluous monitoring.