Cyclone Gabrielle has shone a light on the value of electricity network resilience
As the cleanup continues and efforts to restore power in the worst affected communities are ongoing, questions about the resilience of our power infrastructure are front of mind.
Cyclone Gabrielle has torn a path through New Zealand’s infrastructure and the clean up bill is significant. As I write this there are still over 100,000 people without electricity. Transpower has released some updated information this morning about the extent of damage to their assets in the Hawke’s Bay. This includes over 1.5m of water through the Redclyffe high voltage substation, flooding at Whirinaki substation, and one completely destroyed 110 kV tower.
Whilst Transpower is yet to confirm how long the substation restoration will take, it is clear that it is not going to be a matter of just sweeping out the water and turning things on again. The local lines company, Unison is likely to also have suffered significant asset damage, the extent of which is still unknown. As we look beyond the inital cleanup and restoration, how do we ensure that we minimise the risk of something like this happening again?
It has been recognised for some time that the Hawke’s Bay is vulnerable. There is only one 220,000 V line that supplies the region from the central North Island. This line is a double circuit line, meaning that the large pylons that march across the country from Wairakei carry two independent circuits. In the world in which I spend my professional time, this is called “N-1” security. That is, either of the two circuits can fail and the region will still be connected to the rest of the country via the remaining circuit.
Transpower has a requirement to continuously ensure that all of its high voltage substations are operating under N-1 security, or to use the terminology in the Electricity Industry Participation Code (EIPC), a “secure state”. Transpower goes to great lengths to ensure this, right down to ensuring there are things like redundant low voltage supplies should one fail. With regard to the Hawke’s Bay though, the question can rightly be asked, is the system in a “secure state” when the flooding of one substation results in the disconnection of an entire region?
More to the point, can you really claim that the system is N-1 secure if both supplies terminate at the same point? The answer is technically yes, but as we now have first hand knowledge of, pragmatically no.
To be clear, the blame for this situation does not sit with Transpower. They are operating within a regulatory environment where they are required to seek Commerce Commission approval for major grid upgrades, which would be required to increase the resilience of the Hawke’s Bay region.
All major capital upgrades must pass a grid investment test which assesses the net benefit of the proposed upgrade to the electricity consumers in NZ. One aspect of this is the grid reliability standard which states that the minimum reliability standard of the core grid is N-1. Despite the perhaps obvious evidence to the contrary at the moment, the Hawke’s Bay region does meet the existing grid reliability standard. However, should the current grid reliability standards be relooked at in the light of Cyclone Gabrielle? Perhaps to include a probabilistic assessment of the complete failure of a single substation?
Transpower also manages an asset base that has been developed for over a century. Naturally there are different standards for equipment that gets built today versus assets that were constructed 40 or more years ago. Transpower, to their credit, also have an ongoing resilience workstream to identify and improve assets.
What to do with the supply to the Hawke’s Bay then? There is currently an alternative feed to the region through the 110,000 V network via Woodville, Dannevirke and Waipawa, terminating at Fernhill substation which is west of Hastings. This network is not capable of supplying the whole region and Transpower actually manages a permanent operational “split” in the network at Waipawa to avoid overloading this part of the system. This split has been closed to at least allow Unison to supply some of the network through Fernhill substation.
Presumably in the coming days work will continue on bypassing Redclyffe substation to allow Whakatu substation to be energised. Hopefully, with some reconfiguration of Unison’s network, this will then allow supply to be restored to most people in the region.
In the longer term, how can we ensure better redundancy and security for the people of Hawke’s Bay? The obvious solution is another 220,000 V line from the south. This could be connected to Bunnythorpe substation near Palmerston North with the remote end terminating at Fernhill substation, a distance of approximately 125 km. It would also have the side benefit of providing additional resiliency and redunancy to other southern Hawke’s Bay and Wairarapa communities by allowing a 220 kV connection to substations in this region.
What would such a line cost? The most recent example in NZ of a greenfield transmission line build was the Whakamaru to Pakuranga double circuit line which cost $894M in 2012. This line is 186 km and was actually constructed to 400 kV, and had an underground cable section. Given an allowance for inflation over the last ten years, we could expect a new 220 kV line between Bunnythorpe and Fernhill to perhaps cost at least as much on a pro-rata basis, so perhaps $600M. What price for resilience?
As we have seen around the world and in New Zealand, the construction of new high voltage lines is neither easy nor fast. It is likely that such a project would take a minimum of ten years with no guarantee that it would get through the existing resource management act process. Such an upgrade has always been an option for Transpower to pursue, but as mentioned above, due to the current settings of the grid investment test, it would have been unlikely for the Commerce Commission to approve it. Cyclone Gabrielle will perhaps have shifted minds here.
If not a new line, then what other options are there? The existing 110 kV line could be upgraded to 220 kV. This would perhaps limit some of the objections, but it would not necessarily be a more cost effective option and would require more work at the substations through which this line passes. It would also take many years to complete.
I’m a big proponent of electrifying our economy. It is one of the most effective ways that the country can decarbonise. However, it will only happen if the network is resilient in the face of natural disasters that will continue with increasing frequency as the world warms. The case for a new high voltage line into the Hawke’s Bay is clear. But adjustment of the regulations that govern grid investment is also needed, to ensure that resilience of the electricity network is rightly prioritised.
Let’s get on with it.
As I understand what happened in the Hawkes Bay is that the main Transpower GXP substation at Redclyffe flooded due basically due to its position in a flood prone area and the extreme weather event and this was a known risk factor. The existing 220 V transmission circuits did not suffer huge damage - one tower issue ? Unison relies on Redclyffe GXP and did not have much diversity of supply ( at 33kV) to act as back up from other surrounding (lower capacity ) GXPs in their region. The priority to achieve resilience improvements therefore would appear to be making Redclyffe GXP less flood prone and for Unison to look at their sub-transmission system design and build back better with more diversity of supply sources although this may need other GXP capacities to be increased. I think a lot of this resilience improvement work could be achieved with a bill less than $300M.