The electricity distribution industry has been one of the few sectors that has had no need to develop comprehensive or sophisticated data management capabilities. Industry regulation and legislation have offered no incentives to invest in business-led schemes such as enterprise data management.
But times are changing with the arrival of smart grids. Traditional power systems are converging with modern information, control and communications technologies, so DNOs increasingly have to handle huge amounts of information from new data sources and distribute data securely to a new set of stakeholders. Add to this a record increase in the number of grid sensor and monitoring devices, grid management applications and advanced analytics, and better integration with back-office IT systems, and the scale of the data challenge becomes clear.
Sound like a tall order? The good news is other industries that have done this very successfully. Financial trading firms faced a similar challenge after the “Big Bang” of 27 October 1986, and the telecommunications sector had to manage domestic/end-node devices and data from broadband service providers, telcos and re-sellers during the 1990s when the explosion of the internet triggered mass demand for new services.
So what can DNOs learn from these experiences? Obtaining and managing useful information from cleansed, consistent and secure data – and identifying distinct classes of data to establish an initial data management framework – should be the main aim of any future and enduring smart grid data management strategy. This will involve managing upfront investment and controlling ongoing costs. Industries that have opted for piecemeal or reactive data initiatives have multiple iterations of failed IT-led programmes to demonstrate why establishing a robust framework from the outset is so important.
Making sure that network data is visible at the level of detail needed by those responsible for operating the network is a must. This is particularly relevant when, for example, connecting new devices to live network capacity limits offered by dynamic ratings technologies for overhead lines, underground cables or MV/LV transformers.
Some good work is under way in individual electricity distribution companies, but a cross-industry, consolidated strategy will be the only way to ensure advanced grid management applications perform consistently to provide the data required. The industry will need a thorough grasp of master, or reference, data, together with an understanding of data flow streams across the distribution network.
In the past, electricity grid data has been considered homogeneous. Now firms must identify distinct classes of data. At a minimum, these might include:
· operational network data;
· non-operational data, such as business support;
· fiscal and sub-metered data, accumulated/volume-based;
· event-driven data;
· metadata, which defines/describes and structures the first four.
Then a new data architecture for the smart grid must be defined. There are many issues to tackle: data volume, information exchange, interoperability and integration, data access and usage, and security. But these issues will be far easier to manage in a formalised architecture than on an ad hoc basis at a later stage.
Many DNOs see data management as a minefield. But realisation is growing that a timely, holistic and considered approach – that can flex to accommodate changing smart grid requirements – is needed. Better to address data management proactively, comprehensively and cost-effectively now than with an uncoordinated series of reactive and expensive initiatives later on.
Neil McNeill is head of solutions at Smarter Grid Solutions
This article first appeared in Utility Week’s print edition of 28th June 2013.
Get Utility Week’s expert news and comment – unique and indispensible – direct to your desk. Sign up for a trial subscription here: http://bit.ly/zzxQxx