Better network resilience for Hawkes Bay. What options and for what price?
Some ideas for making the Hawke's Bay electrical network more resilient.
My last post on focusing on resilience of the electrical network in the Hawke’s Bay following cyclone Gabrielle generated a bit of interest and some thoughtful questions and comments. I felt there were sufficient “gaps” in the original post to warrant a follow up. Here goes.
I mentioned, somewhat provocatively perhaps, that the need for a new 220 kV line from Bunnythorpe - Fernhill was clear. What I really meant, was that improved resilience for the Hawkes Bay power system is a reasonable goal. A new 220 kV line is one option, but there is no silver bullet. For instance, Transpower has restored supply to Redclyffe substation, yet there are still customers without power in the Hawkes Bay. Why? Because most residential and commercial power outages are caused by faults in the distribution system, not the transmission system.
What proportion of faults are attributable to the transmission system versus the distribution system? A recent study in the US identified that around 90% of all interruptions are primarily caused by the distribution system. That is, for every outage experienced by a customer, 9 out of 10 are caused by some issue in the distribution system. This may be counter-intuitive, probably because transmission outages are generally more widespread, whereas distribution outages are generally localised. Your local “car vs pole” doesn’t tend to make the mainstream news. This strongly suggests that the focus for improving network resilience should be on the distribution system.
I tend to agree that we need to improve the distribution system reliability, but that doesn’t mean we shouldn’t also try to improve the transmission system. This is a false dichotomy.
There have been some excellent articles covering practical things that can be done to improve Electricity Distribution Business (EDB) resilience, including:
Undergrounding of overhead lines; and
Better “resilience routing” to provide less single failure points
Undergrounding cables is somewhat of a controversial topic; yet it doesn’t need to be. There is no question it costs more to bury cables, than run a comparable overhead line. An old “rule of thumb” that I used to work by was roughly 10x the cost. I think that was more based around HV transmission, rather than distribution though. Costs are also unfairly weighted to initial capital expenditure, rather than ongoing operational expenditure, for which overhead lines is considerably higher than underground.
In most circumstances, underground cables are more reliable than overhead lines. The reasons include less exposure to weather and vegetation related problems. Unison’s most recent asset management plan reports that 25% of outages in 21/22 were attributable to vegetation related causes. Undergrounding would reduce many of these problems.
It’s not always appropriate to underground electricity lines. For remote communities where the density of power connections is low, it is very hard to justify the additional expense. There are also legitimate concerns about cable resilience in earthquakes and during slips. Another difficulty with underground cables is that any damage takes longer to locate and repair than for overhead lines, increasing the outage duration. This offsets a little the benefit from reduced outage frequency.
Earthquakes that damage cables are far less common than weather incidents that damage overhead lines. There are extensive underground cable networks in earthquake prone areas like California and Japan. Japan recently announced a plan to bury 4,000 km of powerlines by 2025. One utility in California plans to spend US$10 billion burying power lines to prevent fires.
So more undergrounding at the local level then. What about for the transmission supply? I mentioned in my last article that the cost of a new 220 kV line would be approximately $600M based on Transpower’s most recent experience, approximately $5M/km. After further investigation, there is actually a better example, which is the more recently constructed Wairakei-Whakamaru-C 220 kV line. This 38 km line was built for approximately $1.7M/km. Allowing for ten years of inflation, today’s cost is probably more like $2.3M/km. This suggests the cost of a Bunnythorpe-Fernhill line might be in the vicinity of $300M.
It doesn’t necessarily need to be a double circuit line, although this would be better from a “core grid” point of view. A single circuit line using steel monopoles might be cheaper and require a smaller easement.
Connecting Bunnythorpe through to Fernhill at 220 kV would provide an alternative “eastern” route for bulk transmission through the North Island, so it has the side benefit of improving core grid resilience in addition to improving Hawke’s Bay security of supply. This is not a minor consideration, especially with the potential for additional renewable energy connection in the Southern Hawke’s Bay and Wairarapa, which would require transfer to load centres in the North.
Transpower is already proposing to spend $128M on thermally upgrading other transmission circuits in the central North Island as part of their Net Zero Grid Pathways (NZGP) work. Would this money be better spent contributing to a more ambitious project for a diverse transmission route up the east coast, thereby improving the Hawke’s Bay security of supply, but also providing wider system benefits?
Transpower have already looked at this option as part of their “long list” of transmission options for NZGP. It was identified as option “D3” and appears to have been discarded due to a longer lead time than other items and the need for significant upgrades to other parts of the grid. Transpower has rightly prioritised least cost and low regrets options, which was a reasonable thing to do. However, given what has transpired with Redclyffe Substation and the Hawke’s Bay, would it be reasonable to relook at the long list with more emphasis on resilience?
Unison’s letter as part of the consultation process for the longlist options, with uncanny foresight, listed four reasons why option D3 should be preferred, including:
Access to the grid for the significant solar and wind resource in the central and southern Hawke’s Bay
Diversity of transmission routes including reduced exposure of the transmission system to central North Island volcanic hazard
Improved security of supply to the Hawke’s Bay.
Potential to expand the route through to Haywards through the Wairapapa, potentially alleviating some longer term constraints.
Could we achieve the same benefits as a new line with non-transmission alternatives? Here is where we start getting into murky water about non-transmission alternatives such as batteries and demand management. There is going to be an increasing role for batteries as they continue to become more economic. They can provide important backup. However, typically, they are sized to provide hours of storage. As we have seen from Gabrielle, we may require solutions that can provide power without a grid connection for days. It is hard to beat liquid fuels for emergency resilience and I believe there will continue to be a need for diesel gensets for some time to come.
Demand management is a useful tool to reduce strain on the network during peak times and avoid over-building our distribution and transmission networks. It doesn’t provide tools to help with resilience during large scale loss of supply events.
What about micro-grids and self-sufficiency? Yes, those communities that build small resilient networks that are somewhat self-sufficient will fare better in emergencies. However, most communities are not going to have the will or resources to do this. There is power in the scale of our distribution utilities and they have shown that they can provide resilient networks within NZ and overseas.
Let’s carry on the resilience conversation, and the trade-offs we must make. It is a conversation worth having.