In June, I wrote how intermittent
power sources such as photovoltaics and wind would have to compete with baseload
technologies such as IGCC "Clean Coal" and nuclear for capacity on the
grid. The key problem is that neither baseload technologies nor
intermittent technologies are able to match themselves to the fluctuations of
demand. This creates a need for technologies which can fill the varying
gaps between supply from these sources, and normal energy use. From the
comments, it seems like I was not completely clear how intermittent and baseload
power cause problems for each other, so I will start with a simplified example,
which I will use to illustrate the various strategies for dealing with the
problem. I see investment potential in all of these strategies.
The One-House Grid: An Illustration
Suppose that the entire grid were just one house, and it was the utility's
job to make sure that there was always enough power to run all the gadgets that
anyone in the house was using. Even in the middle of the night when
everyone is asleep, there will still be some power usage: running clocks, the
VCR, charging cell phones for use the next day, and maybe the porch light.
That is the minimum load of the house, and traditionally utilities have met this
demand with baseload power. In contrast, there will probably also be a
moment on hot summer afternoon when the air conditioner is running full blast,
the refrigerator kicks on, dad is watching football on his 60" plasma TV,
dinner is cooking in the electric oven, and 15 other appliances are on somewhere
or other. This is peak load, and the difference between the minimum load
and peak would traditionally be met with dispatchable generation, which, until
recently, mostly means gas turbines.
In addition, some dispatchable generation will always be kept running below
full capacity in order to maintain power quality and availability as appliances
are turned on and off throughout the day. These ancillary
services [pdf] are called load-following reserves (maintaining availability)
and voltage and frequency regulation (power quality,) and both require fuel,
even if the actual energy provided is negligible. Ancillary services are
like your car's engine idling at a stop light so that you can start quickly when
the light changes. They're necessary to keep the system running, and they
use fuel, but they don't actually get you anywhere. Also like idling
engines, options like hybrids exist which can save much of the energy cost (see
below.)
Add a Solar Panel
Suppose we now add a photovoltaic system and a wind turbine on the roof. Most people with solar systems know, that if you want to spin
your meter backwards (i.e. produce more energy than you are using) the best
time to do it will be in the late morning, while it is still cool, but it's
bright enough that the panels (which actually produce
more power from the same amount of light when they are cool) are producing
near their peak output.
With grid-connected solar, spinning your meter backwards may be fun, or at
least get you bragging rights. However, in my fictional one-house grid, we
now have a new minimum demand: demand will be negative (we're going to have to
find something to do with the excess electricity) because there is no other grid
to sell it back to. Peak demand will also be reduced, because on the hot
summer day, the PV will also be producing power. The result is that the
one-house gird with a PV system will no longer need any baseload generation
(since minimum demand is now negative), and it will probably also need less
dispatchable generation, because peak demand will also have been reduced, most likely
by more than minimum load. Not only will peak demand have been reduced, but it
will also have shifted to the early evening when the PV is producing little
electricity, but cooling, cooking, and football watching needs are still high.
Adding a wind turbine to the roof has a similar effect. Now, the meter
will also be spinning backwards on windy nights, and demand is reduced whenever
it's windy, which will in turn save fuel and reduce the need to run the
remaining dispatchable generation.. However, if the climate is similar to
that here in Denver, on the hottest days of the year, the wind will typically be
minimal, so there will be little further reduction in peak load, so nearly the same
total amount of dispatchable generation will be needed, although it will not be
in use as often.
Consequences
As the above illustration shows, the oft-repeated shibboleth that we
"need" baseload generation is not only misleading, but also
counter-productive. Adding baseload generation will simply increase the
number of hours per year that intermittent sources of power exceed net
demand. I too, formerly believed
we needed baseload. I no longer do, although some level of baseload
power in the grid is no doubt inevitable, at the very least produced by
renewable sources such as geothermal and electricity generation from industrial
waste heat.
Solutions
Returning to our one-house grid thought experiment, a number of options
present themselves.
- Storage. In the real world, if you build a house off the grid,
you will add batteries so that you can still run your lights when the sun
isn't shining and the wind isn't blowing.
- Transmission. Suppose our one-house grid has a neighbor,
running his own one-house grid. While generation from their PV and
wind systems will be similar (but not identical), demand at the two houses
is likely to be different. By diversifying the electric demand,
average demand will double, but peak load will increase by somewhat less,
and minimum load will more than double. This reduced volatility
of electrical load brought by connecting two homes is analogous to the
reduced volatility of a portfolio of two securities, rather than just
one. Unless the electrical load of the two homes is perfectly
correlated, there will be benefits in terms of a reduction in the overall
amount of dispatchable generation needed to service the same total
load. Our knowledge of the principles of diversification will
correctly lead us to the intuition that connecting dissimilar users of
electricity will lead to greater diversification benefits than similar
users. If residential, commercial, and industrial users are all on the
same grid, the same average electric demand will be easier to serve than if
only residential or only industrial customers were connected, because a
residential user will have lower correlation of demand with most industrial
users than with other residential users.
- Demand-Response. My sister lives in an old house, and the
kitchen is on an old, low amperage circuit breaker. If she ran both
the microwave and the toaster at the same time, it would trip the breaker
and she would have to trudge outside to turn it back on. Needless to
say, she quickly stopped using the toaster and the microwave at the same
time, and thereby reduced the peak load in her kitchen. Demand
response involves getting electric customers to agree ahead of time to
refrain from using high-wattage appliances during times of high electric
demand. In the one-house grid example above, dad might choose to
record the football game and watch it later in that evening.
- Energy Efficiency. Another way to reduce volatility of demand
is simply to reduce overall demand. If dad had decided to buy an LCD
TV rather than a Plasma TV, the demand from his 60" TV might have
been reduced by as much as 200-300 watts, depending on the models, and this
in turn would have reduced peak load.
Investments
Each of the above solutions leads to an investment, and as intermittent power
sources grow as a percentage of total generation, the needs for these solutions
will increase. Below is a selection of companies working to provide each
of the above solutions to the overall problem of matching electrical supply and
demand.
Electricity Storage
Electricity storage can serve several related needs of the grid. First,
it can absorb excess supply of power at times of otherwise low demand, which
means that intermittent and baseload sources of power do not need to be
curtailed, even though they are producing power at near zero marginal
cost. Second, when charged, energy storage can provide ancillary services
to the grid, by supplying power to meet short term spikes in demand or drops in
supply, and absorbing power if intermittent generation ramps up unexpectedly, or
demand suddenly drops. According to Paul
Denholm of the National Renewable Energy Lab, the revenues from these
ancillary services are significant, and should not be discounted in any economic
assessment of an energy storage technology. Finally, storage can help to
shave peak load by supplying power from off-peak charging.
I have previously written about investments
in large scale batteries for the electric grid, but when I did so I
neglected to consider the value of ancillary services. Since I wrote that
article, both VRB
Power (VRBPF.PK) and NGK Insulators have continued to sell their respective
solutions to utilities,
telecoms,
and other
consortia. However, these technologies are still searching for general
market acceptance. Beacon Power
(BCON) recently commissioned a 20
MW flywheel based plant to supply frequency regulation services to the New York
grid, which will primarily be used for frequency regulation. Given the
enormous potential of demand response and electricity transmission to improve
long-term electricity price volatility, I am currently much more bullish about
companies using energy storage primarily to provide ancillary services over
large scale storage. Because of that, I have recently increased my
investments in Beacon, Maxwell
Technologies (MXWL) and Active Power
(ACPW).
Maxwell's ultracapacitors can be used in various power
quality applications, as well as a high power, low energy supplement to
batteries in hybrid electric vehicles. (As a side note, high power is more of a
concern in hybrids than pure electric vehicles, because the smaller battery pack
has difficulty producing enough power for rapid acceleration.)
Active Power, like Beacon, uses flywheel technology, selling mostly into the
customer side, rather than utility side of the market. However, as the
market for ancillary services grows and becomes more sophisticated, I could see
Active Power's UPS systems selling ancillary services to the grid, in addition
to their primary function of protecting data centers and other sensitive
equipment from temporary power outages.
Transmission
I've written extensively about investments in electricity transmission and
distribution. My top picks are ABB
Group (ABB) and Siemens (SI), Composite
Technology Corporation (CPTC.OB), ITC
Holdings Corp (ITC), Quanta
Services (PWR), General
Cable (BGC), and National
Grid (NGG). Geographic diversification of electric supply and demand
is as essential as financial diversification in your portfolio.
Demand-Response
I haven't written about demand-response
aggregator EnerNOC (ENOC) since before its IPO in March 2007, but that
doesn't mean I'm no longer interested. EnerNOC,
along with Demand-Response/Smartgrid companies Comverge
(COMV) and Echelon
(ELON) all became quite
expensive on a wave of investor euphoria in 2007, which is why I was not buying
or writing about them much at the time. That has now changed, with all
three losing about 70% from their peaks, and making them look relatively
valuable. I have been taking positions in all three over the last few
months.
Energy Efficiency
Unfortunately, few pure-play energy efficiency companies exist. The recently
named Waterfurnace
Renewable Energy (WFIFF.PK)
is one I've recently been adding to my portfolios. I've previously written
about Flir,
Inc (FLIR), a thermal imaging company which I do not currently own due to
valuation concerns, a pair of LED
companies, Cree (CREE) and Lighting Science Group (LSCG.OB) , and a number
of energy
efficiency related
conglomerates.
DISCLOSURE: Tom Konrad and/or his clients have long positions
in VRBPF, BCON, ACPW, ABB, SI, CPTC, ITC, PWR, BGC, NGG, ENCO, COMV, ELON, WFIFF,
CREE, LSCG.
DISCLAIMER: The information and trades provided here are for informational
purposes only and are not a solicitation to buy or sell any of these securities.
Investing involves substantial risk and you should evaluate your own risk levels
before you make any investment. Past results are not an indication of future
performance. Please take the time to read the full disclaimer here.