Knowledge Problem

Michael Giberson

When I read someone suggesting that “vehicle-to-grid” (V2G) operations will make money for owners of plug-in hybrid vehicles (PHEV), I wonder how carefully they’ve thought through all the implications.*  The analyst might assume a particular battery technology and characteristics, for example, and then run a simulation against market data to see how much value can be produced in energy arbitrage.

V2G energy arbitrage requires a real-time rate (or at least a variable rate of some sort), and the profit depends on the round-trip efficiency of the battery system and the difference in peak and off-peak prices.  In the simplest case, the consumer uses the stored power to offset her own consumption at peak and isn’t selling power back to the grid (avoiding the related transaction costs).

I see two related problems with the modeling approach.  First, it may be reasonable to assume historic differences between peak and off-peak prices if few PHEV will engage in arbitrage, but even modest penetration of the market will add enough storage capability to start equalizing prices.  The more dramatic peak locational prices are driven by transmission constraints and the cost of starting up high-cost peak power plants, and it just wouldn’t take much in the way of PHEV storage to reduce or avoid many transmission constraints.   A good estimation should include consideration of the elasticity of supply and demand in order to assess the degree to which arbitrage will tend to reduce arbitrage opportunities.

Second, when battery technology advances sufficient to make PHEV economical, won’t have battery technology also have advanced enough to make grid-dedicated battery storage applications economical too?  A PHEV battery will be optimized for vehicle operations, with energy arbitrage sort of an afterthought, while grid-dedicated batteries will obviously be optimized for providing grid services.  And PHEV technology is the much harder problem because of the size and weight constraints.  By comparison, grid-dedicated energy story is easy.  Is it likely that V2G can out-arbitrage grid dedicated devices?

(Yes, V2G-based arbitrage has one big advantage if you assume that the cost of the energy storage device is fully justified by transportation, and is in effect freely available while parked to engage in some energy day-trading.  Maybe this consideration saves V2G.  But dedicated grid-storage devices are available 24/7, while PHEV will have competing uses.)

Some V2G analysis also proposes that PHEV supply high value grid services like reserves, frequency control, voltage support, and so on.  I think the same considerations apply. If there are thousands of little PHEV energy storage devices connected to the grid supplying these services at very little additional cost to the owners, then the price will fall.  If PHEV devices can provide these services at low cost, then dedicated grid-connected devices should be cheaper.

In short, some of these PHEV V2G value calculations are at best numbers for the early movers.  Once everybody is doing it, it won’t pay to do it anymore.

*Of course I could actually read the more serious reports on V2G** and see for myself how carefully they’ve thought through all the implications, but it is late Friday afternoon and with classes beginning next week I have a million other things to do.

**Actually, I was just reading a new working paper from the CMU Carnegie Mellon Electricity Industry Center, but since the paper was prominently stamped “DRAFT: Do Not Cite Or Quote” on every page, I am not citing or quoting it.  Technically speaking, it doesn’t say “Do Not Link To The Abstract.”