Air compressor as energy storage? - Page 10

 
you might want to read this...
It's a a wikipedia article on grid energy storage.
==================================
http://en.wikipedia.org/wiki/Grid_energy_storage

  drydem wrote:


The pumped water example given goes like this:  At night, buy
electricity at 1.5 c/kwh to pump some water to a high elevation, then
during the day use that water to generate electricity to sell at 4
c/kwh.

(the example given says that the pumping process is 75% efficient,
which I'd like to know how they got that efficiency so high, but
that's another matter).

So even with pumping loss, you're still making maybe 2+ c/kwh.

But the flaw is this.  What is the other guy doing burning (or
consuming) his fuel and only getting 1.5 cents at night in the first
place?  Why doesn't he conserve it too so that he can make more $$$
during the day by selling it at 4 cents like you're doing?

 HVAC Guy wrote:

The other guy might still be making a profit at 1.5 cents if
his costs to produce that kWh is significantly less than that.
The other guy might even be using a generator that takes many
hours to change power levels, works best at a specific level
or which can be damaged by too often changing levels.

Anthony

 Anthony Matonak wrote:


Which means that other types of plants that can't scale their output
within a few hours would presumably be looking for technology to be
able to do so, because it's in their interest to be able to scale-back
production so as not to over-supply the market which leads to a low
price for their output.

The supply of cheap electricity at night is a by-product of
operational limitations of current plant design and I'm sure it's not
intentional and perhaps future plants will not have this limitation.

The other strategy to always run the big plants at full output and
convert excess demand into other forms of energy storage (like
batteries, generate hydrogen, etc) is so stupid because of the energy
conversion losses and the infrastructure needed for society to use the
alternate forms.  This strategy again is based on the assumtion that
the big plants can't, or won't, or don't want to scale back output to
meet demand.  The best solution IS to scale back output so that you're
not burning your main fuel stock when you don't have to.

 

Wrong.  Total production *always* equals total demand.  Plants *do* scale
down production at night.  They have to, because outside of a few
technologies like pumped-storage, electricity cannot be stored.  If
production isn't reduced at night to match the night time load, then the
grid frequency and voltage would rise sharply and plants would trip off
anyway.

The low cost of electricity at night is not a by-product of limitations in
design, it is a direct result of the varying cost of production using
different plant types.

A simplified example:
You have four different power plants, each 1000 MW.  Each has a
fuel/operating cost and a fixed cost.

Plant A burns coal.  It has a fuel/operating cost of $4/MWhr and a fixed
cost of $240,000 / day.  If it runs all day and night, its production cost
is about $432,000 / day and it produces 24,000 MWhr for a rate of about
$14/MWhr (1.4 cents/kwh).

Plant B is nuclear.  It has a fuel/operating cost of only $0.85/MWhr but a
fixed cost of $600,000 / day.  If it runs all day and night, its rate is
about $25.8/MWhr (2.58 cents/kwh)

Plant C is hydro.  Although its fuel is 'free', the operating cost is about
$1/MWhr.  The fixed costs might be $100,000 /day so it's overall rate is
only $5.17/MWhr (0.517 cents/kwh)

Pland D is natural gas.  It's fuel is very expensive at upwards of $30/MWhr
but it has a low fixed cost of only $200,000 / day.  So if it runs all the
time, its rate is $28/MWhr (2.8 cents/kwh).

So, perhaps during the day, the load is 3800 MW.  You run plants A, B, and C
full output and D at 80%.  If someone turns on one more light somewhere, the
plant that picks up the additional load is the expensive natural gas plant
at $360/MWh (36 cents/kwh) for that additional load.

But at night, the load is only 1500 MW.  Obviously we shutdown plant D and
avoid the natural gas bill.  But which of the others do we curtail?  You
might be tempted to shutdown the nuc as well (assume that all plants are
equally flexible in this example) and just run the hydro at 100% and the
coal plant at 50%.  It comes as quite a surprise to a lot of folks in the
industry when they work out the answer and it turns out to be 'run the nuc
at 100% and the hydro at 50%, shutting down the coal'.  Don't believe me?
Work out the total cost of producing those 1500 MW for 12 hours using the
different 'mixes' of which plants to run and you'll see.  (once you know
what to look for, determining the cheapest mix is easy :-) look at the
variable costs of each plant )

Turn on one more light at night and the cost of powering it costs just
$12/MWhr (1.2 cents / kwh).  This is all assuming that each plant is able to
change load at will, any amount at any time.  So the daytime versus
nighttime marginal cost is either 36. cents/kwh or 1.2 cents/kwh.  It isn't
the plant's operational capabilities, it's all in the fixed and marginal
costs for each plant.

Now, another question is, "Does it make sense to run the hydro unit at a
higher power level so it can be used to pump water into storage?"  Probably
not.  But there are 'hydro' units that are known as 'run-of-river'.  They
don't store up water.  If they don't use the water through a turbine, it
just spills over the top of the dam and runs down the river anyway.  If
that's what you have and you have a storage facility somewhere else, then
yes, it does make sense to store the energy from the river water in the
pumped storage facility.  But if you have a reservoir behind a dam, it makes
more sense to just hold the water behind the dam and let the reservoir level
rise for 12 hours, then draw it down during the day.

But remember that many hydro facilities have additional constraints besides
just power production.  They may be constrained in how fast they can draw
down the reservoir (limit downstream flooding), and how high/low they can
vary the reservoir level (irrigation, sporting, erosion).  For example, in
late summer when day-time peak may be very high, hydro may not be allowed to
run at full power due to low water levels.  But pumped storage can be
'charged' at night by coal and nuc, then add it's output to the total during
the day making up shortfalls in generation.

If, in my example above, demand keeps growing every year, pretty soon you
will need a new plant to supply the increase in day time load.  But a pumped
storage facility could be the answer.  Run the nuc, coal, and 'run-of-river'
hydro plants all night to supply 2800 MW (1500 MW load plus 1300 MW pumping)
so you're pumping water to a 12000 MWhr pumped storage facility for 12 hours
(a 1000 MW unit with 12 hour storage) .  Then during the day, run the nuc,
coal, 'run-of-river' hydro and pumped storage facility at full power and run
the natural gas plant just as needed for when demand exceeds 4000 MW.


NOT burning fuel is always an option.  But if you need power beyond what
your cheaper plants can produce during the day, you have two choices.  Burn
a different, more expensive fuel to meet the intermittent demand, or find a
way to store energy from cheaper fuel in some inexpensive manner so it can
supplement the cheap plant output when demand exceeds the cheap plant's
output.  About the only storage that fits the bill today is pumped water.

Of course, if you could supply *all* your energy needs from just the
cheapest type of plant (i.e. build nothing but hydro), that would be better.
But obviously there are limits as to how many such plants can be built.

daestrom