Climate change and the grid

December 30, 2015 2:03 am0 commentsViews: 101

It’s the information that counts
The second key element in a smarter grid is the leveraging of the huge volumes of data collected – the cyber approach. “We’re talking here about exabytes of data. The largest producers and consumers of power grid data are the hundreds of millions of sensors and controls embedded in smart devices installed in buildings, substations, generators, transformers and other equipment in the transmission and distribution networks,” says Dr Jones.
“There is expanding data from the increasing amount of variable renewable generation resources, demand response programmes, and distributed energy resources such as electric cars and energy storage. Grid operators today and more so in the future will have more access to external data sources such as weather agencies, etc. Extracting actionable information from this avalanche of data will help identify and predict physical phenomena.”

From reactive to predictive operation
This interdependence of the physical and cyber domains is undoubtedly one of the salient challenges of the industry. But, this coupling could also present opportunities in different ways to operate the grid when faced with severe weather events. “Instead of the conventional reactive mode of operation, we are at the beginning of the new age of applying more predictive techniques. Operators will have to keep the lights on while coping with the uncertainty due to climate change. Dr Jones gives examples: “In the case of the tornado that struck Oklahoma in May 2013, it is reported to have rapidly intensified to an EF-5 level tornado in less than half an hour. Grid operators need to be able to simulate such climate-related anomalies and run ‘what-if’ scenarios to better anticipate how the grid reacts and what actions to take.

Similarly, in wind farms across Denmark, the wind speed can go from 0 to maximum in 10 minutes. With integrated forecasting technology and ultra-fast computation, the control centre can calculate what will happen in the next five minutes. This capability enables a predictive mode of grid operation – and is indeed a requisite for what has become known as a selfhealing grid – that anticipates events and responds to them to mitigate their negative impact on the network. This can help to make the system more resilient.”
In distribution systems, Volt Var Optimisation (VVO) optimises power flow using real-time information and online system modelling. “Probably one of the most valuable applications of predictive tools is in asset management. We are now entering what I call the age of ‘hybridity’. For at least the next 30 years, power grids, especially in OECD countries, will consist of both old and new devices and equipment. While utilities will have to replace old assets, there are many assets with more than a decade left in their operational lifespan. Smart condition monitoring devices can be integrated into the grid and asset control rooms for analysis and improved grid operation. Interoperability of the old and new devices is a priority.”

On-going investment in IT solutions
All this will require a major investment in Information and Communications Technology (ICT) solutions. “A particular emphasis will be on advanced grid and asset analytics as well as decision-support systems to harness all the data. The new emerging operational paradigm will require the creation of information flows that allow operators to take appropriate action in real time – or perhaps rather ahead of time. Some applications already exist, but the effort will continue for five to 10 years to come.” To corroborate Dr Jones’ prediction, Navigant Research, a market research and consulting company with special expertise in the energy sector, forecasts that worldwide spending by utilities for smart grid IT systems will more than double in the next 10 years. As the climate changes, the electricity grid will adapt and become more resilient.
Based on inputs from Alstom

Pages: 1 2

Leave a Reply



× 5 = fifteen

shared on wplocker.com