Renewables not the root cause of SA blackouts but data management key to preventing another black event
When the lights went off and all power went out in South Australia on 28 September last year, public commentary was quick to blame renewables, but the report released yesterday by Australia’s Energy Market Operator (AEMO) tells a different story.
Black System, South Australia, 28 September 2016 gives a detailed account of what happened on the network, at what time and how the services were recovered to South Australian residents and businesses. It’s an energy engineering delight covering technical challenges faced by the network and the difficulties in dealing with less synchronous generation online and increasing periods of voltage fluctuations.
The report sets the facts straight.
This is a welcome report in what is an emotionally charged debate on energy security and reliability. Importantly, the amount of wind and solar renewables was not the root cause of the Black System event, although voltage fluctuations from newer generation did complicate the operation of the network at the time.
Energy networks and grids are going to get a lot more complex in the years ahead.
Real time access to the rapidly growing amounts of network data is essential to minimise failure risk and shorten emergency response times and among the 19 recommendations made by AMEO, this is a key point.
Some of the recommendations were actioned in February 2017 when we came close to having more blackouts during the summer heatwaves. Unsurprisingly, recommendations on national electricity rules and licensing conditions were quickly actioned.
But better data access and management was for me the key theme. AEMO’s new CEO Audrey Zilbelman made this clear at her inaugural CEDA address in Adelaide saying, “we need to use data and infrastructure better and use data tools on rapidly growing networks”.
If market operators and network businesses can get better access to real time data on the devices on the grid, this will help minimise the risk of electricity blackouts and reduce emergency response times.
Is this surprising? No.
As energy networks expand the number of devices on the grid and the networks become more distributed, access to real-time data becomes critical. Accurate asset information is the lifeblood of enabling effective management of the network. As the number of devices exponentially grow we have an opportunity to harness new internet enabled technologies, the IoT (Internet of Things). Collecting data for predictive analytics such as wind speeds and the integration with weather feeds is essential. IoT and predictive analytics will play a vital role in providing real time data on how grid connected equipment is functioning.
To date, utilities have rightly placed significant focus on getting asset information up-to-date and trying to get it 100 percent right. But with a growing number of devices the focus is shifting to assessing risks and failure across a larger population of devices. This is where machine learning, stream analytics and big data tools need to play a greater role in harnessing insights from the ecosystems of data across multiple networks. This also requires secure and resilient network infrastructure to mitigate increasing cyber security threats to networks.
AEMO’s other important call out is the need to do more proof-of-concept trials of promising new technologies. Yes please.
As advocates for innovation, we need to adopt agile testing and learning frameworks more quickly across utilities. Other sectors (banks, telcos) do this well but utilities can do more. For example, banks have to analyse enormous volumes of financial trade data (while in motion) for surveillance purposes in order to create real time responses to financial crime incidents.
Pleasingly, AEMO has begun trials with the Australian Renewable Energy Agency (ARENA) and others on a range of technical trials in grid stabilisation services. Additionally, networks businesses have innovation centres for conducting trials to learn how to adapt their networks for the future. These trials have a great opportunity to test and learn new data management techniques so they can be safely implemented in the field.
There were many lessons from the SA Black System event. We need to test and learn quickly so we do all we can to minimise the risk of interruptions during what remains challenging times ahead.
Better access and data management is a key enabler.