Thread by @whgorman

🔦New article out in The Electricity Journal 🔦

The quest to quantify the value of lost load: A critical review of the economics of power outages

Check-out paper below. Short summarizing thread incoming! 👇
#VoLL #gridreliability #poweroutages

sciencedirect.com/science/articl…

I distinguish applications of VoLL btwn: (1) cost-benefit analysis for centralized reliability-enhancing infrastructure investments & (2) decentralized demand rationing during electric supply shortages (like what’s discussed in @California_ISO this week).

https://twitter.com/California_ISO/status/1568022664580632577?s=20&t=WoYtYEYvxKG928f0tbr6rQ

Power outages resulting from supply shortage can often be mitigated by demand-side solutions like dynamic pricing or demand-response programs (and we need more of them!), which only need loose connection to VoLL estimation

However, what about all of the other types of power outages? Photo: City of Concord NC following Hurricane Irma

Research by colleagues of mine @BerkeleyLabEMP have shown that distribution system outages (NOT the supply shortages that we were experiencing this week @California_ISO) cause >90% of outages on the system. Think damage to T&D by weather events (…or squirrels🐿️)

Read more about their work here: ietresearch.onlinelibrary.wiley.com/doi/full/10.10…

T&D infrastructure investments that make the grid more resilient to factors like extreme weather conditions are not informed by dynamic pricing / demand response. Decision makers use VoLL estimation in order to do those cost-benefit tradeoffs!

However it is HARD to quantify benefits of better grid reliability. I review the various methods applied to do such quantification – and introduce new discussions on how we might use adoption of distributed energy resources for a “revealed preference” VoLL quantification approach

BUT I don’t identify any silver bullets to quantification. I DO provide a compilation of various VoLL estimates by method that have appeared in the literature over the last 40-50 years…As you see, there is a wide range

Though the VoLL is often shown as a point estimate, in reality it is multi-dimensional, i.e. vary by end-user (e.g. industrial vs. residential), end-use (e.g. manufacturing vs. lighting vs. HVAC), consumption context (e.g. community exposure, outage timing/duration/notification)

Such heterogeneity can make it difficult to perform an optimization or cost-benefit analysis that tries to tradeoff the costs of power outages with the costs of generation, transmission, and/or distribution investments across society.

I conclude by arguing that given this difficulty, cost-effectiveness analysis which sets a reliability target (think the 1-in-10 year standard) should remain an important tool for decision makers. Read the paper for the more nuanced narrative!

Much thanks to my PhD advisors @BorensteinS, Duncan Callaway, and Meredith Fowlie for teaching and sharing their wisdom and expertise along the way! Read the full paper: sciencedirect.com/science/articl…