On Friday January 31, 2020, the National Electricity Market narrowly avoided another System Black – the complete shutdown of a state due to catastrophic failure of the system. The last System Black shut down the entire grid across South Australia on September 28, 2016.
This time, as then, the triggering cause was high winds bringing down transmission lines at a critical time in the operation of the grid. On Friday afternoon around 1.25pm high winds brought down six transmission towers in western Victoria, which knocked out the Heywood Interconnector and effectively isolated South Australia from the rest of the NEM.
Friday was a period of record electricity demand in Victoria, with Melbourne recording a maximum temperature of 42 degrees and high humidity. As a result wholesale prices reached their maximum and emergency demand management controls (RERT) were invoked by the Australian Energy Market Operator.
So the failure of the interconnected came at a time when the grid was operating at the edge of its capacity. Which is, of course, the worst time for this to occur.
At the time South Australia was exporting electricity to Victoria, principally because there were reasonable winds blowing in south Australia, while it was relatively still in Victoria. The shut down of the transmission line cut off this supply, causing frequency in Victoria to sag (undersupply) and surge in South Australia (oversupply). This is effectively the exact opposite of what caused the system black in South Australia in 2016.
Electricity market experts Global Roam have provided a rolling series of blogs on the technical details of the trip. The Portland aluminium was isolated from its supply and tripped off, which inadvertently may have saved the day by reducing a large load (but at a cost to Alcoa). The frequency in South Australia surged to 50.96 hertz, which is just inside the upper limit of 51 hertz above which the system is designed to shut itself down automatically. Operating a grid above these frequency limits is dangerous.
In Victoria the frequency sagged as other generators recovered to help cover the load. A bigger system is harder to shift, and it was able to restore adequate frequency levels.
The event will be officially reviewed in detail by AEMO in the coming days. It will be spun in many different directions. The event does highlight the vulnerability of using transmission when it plays a critical role in grid stability, and the increased sensitivity of smaller grids like SA to events like these. Is the solution more transmission lines (to spread the risk) or less reliance on transmission (by using local generation)?