Preventing Cascading Failures in Power Grids
Power grids around the world are decentralizing at a rapid pace as renewable energy production vastly reshapes where electricity is produced. This will require a massive systems overhaul in the way that our power grids are designed, maintained, and protected. Instead of maintaining grid management at a central hub, more and more software and hardware developments are taking place at the grid edge.
According to Wood Macenzie, ‘grid edge’ is “an umbrella term to cover all the distributed hardware, software and business innovations that exist in proximity to the end user, rather than at the centre of a traditional generation network.” Industry experts contend that moving grid intelligence away from central hubs and toward the grid edge is a necessity to adapt and respond to what Greentech Media calls the “undeniable fact that we are moving from a centralized energy system to a next-generation distributed energy system.”
Infrastructure doesn’t only need to move to the edge to keep grids stable, it needs to get much more intelligent too. Demands on our power grid are majorly ramping up with the additional power demand from electric vehicles and data centers, and becoming more complicated thanks to new two-way flows of energy. Historically, energy flowed from utilities to end-users on demand. But now, residential solar panels have turned many consumers into producer-consumers, with two-way energy flows to the grid. What’s more, renewable resources produce energy when the sun is shining and the wind is blowing, not in response to consumer demand. All of this adds up to significant vulnerabilities in our aging grid system.
While artificial intelligence has posed a threat to power grids in recent years thanks to its runaway energy demand growth, it could also provide a critical solution to some of these grid pressures, especially employed at the grid edge, where split-second sensor detection and decision-making abilities has become paramount.
“Milliseconds matter,” reports Power, in an article about grid-edge computing. “The speed of decision-making at the grid edge has become critical for maintaining stability, preventing cascading failures, optimizing efficiency and integrating intermittent renewable resources. With the proliferation of distributed energy resources (DERs), electric vehicles and smart loads, grid edge intelligence has moved from a luxury to a necessity.”
This is leading to a revolution in real-time computing powers placed at the grid edge that can detect and respond to grid anomalies and emergencies like outages and power surges and address them before the utility’s central SCADA system even registers the event, protecting the broader grid from cascading failures.
The United States Department of Energy (DoE) has stated that intelligent computing could be a critical tool in managing smart grids capable of handling huge inflows and outflows of variable energies like wind and solar, and says that “machine learning could help electric utilities improve permitting and siting, reliability, resilience and grid planning,” the DoE report posits. However, the government also notes that the use of artificial intelligence in and beyond our electric systems introduces significant risks if deployed ‘naïvely.’
There are inherently going to be numerous risks and lessons to be learned in a system that is so nascent and so disruptive. As Power notes, “the transition from centralized to distributed intelligence represents a fundamental change in thinking” and an essential shift in the design of massive, entrenched systems. But change is coming whether we like it or not.
“Our electric infrastructure is aging and it is being pushed to do more than it was originally designed to do,” the United States government acknowledges. “Modernizing the grid to make it ‘smarter’ and more resilient through the use of cutting-edge technologies, equipment, and controls that communicate and work together to deliver electricity more reliably and efficiently can greatly reduce the frequency and duration of power outages, reduce storm impacts, and restore service faster when outages occur.” And those cutting-edge technologies will have to be developed at the edge, where split-second decisions need to happen most.
By Haley Zaremba for Oilprice.com
More Top Reads From Oilprice.com