This post will get into some technical topics. It is important to understand these topics to understand the real solutions to Mon Power’s capacity shortfall.
So, let’s start with how Mon Power gets the electricity it sells in WV. Most people think that power companies just produce electricity at their plants and sell it to their customers. Some companies do this, but Mon Power doesn’t.
Mon Power sells all of the electricity it produces, and all of the production capacity it controls, from WV power plants, including the 20% of the Harrison plant it owns, into the PJM Interconnection regional electricity and capacity markets. Then Mon Power buys back the electricity it needs for its WV customers from those same markets.
So, first, we need to understand the difference between electricity (or energy) and capacity. Electricity is what it says. When customers turn on their switches and appliances, Mon Power provides electricity to meet that demand.
Capacity is a little different. PJM requires a certain amount of power plant capacity to make sure that it can call on a definite amount of electricity during normal times, as well as for short periods of time when a lot of electricity is demanded by customers, as well as a specific margin of safety beyond that capacity in case anything goes wrong. In order to insure that the total capacity it needs is available, PJM holds capacity auctions at regular intervals in which power sellers can reserve capacity on power plants to insure PJM’s capacity requirements are met.
So there is a difference between electricity, which is the actual watts of power sold, and capacity, which is the reservation of production capacity needed to meet PJM’s resource requirements. Companies that own power plants can sell both electricity and capacity, and companies that sell electricity to customers buy that electricity on PJM’s electricity markets, as well as reserving capacity to insure that they can meet future demand and meet PJM’s peak load and reserve requirements.
Which gets us to peak load and base load. Here is how expert Richard Hornby, of Synapse Energy Economics, explained it in his Harrison case testimony last week:
Peaking, load-following and baseload capacity generally have different characteristics in terms of physical operating capabilities, fixed costs and variable production cost. Peaking capacity ideally has the flexibility to operate at very high output levels with short notice for short periods. This segment would ideally be served by capacity with relatively low fixed costs because it only generates energy in a few hours of the year, e.g. less than 5%, and therefore have a very low capacity factor. Load following or intermediate capacity must have the flexibility to increase and decrease its generation substantially and quickly in response to increases and decreases in customer load. Base load capacity generally has high fixed costs and low variable costs relative to load-following and peaking capacity. Base load capacity is cost-effective when it operates at a relatively steady level and high capacity factor because its high fixed costs are recovered over a large
annual quantity of annual energy.
In his testimony, Mr. Hornby points out that FirstEnergy can meet 99% of its electricity needs for its customers for the next 25 years with no problem. FirstEnergy’s only “shortfall” is in peaking capacity, the ability to meet the highest demands for electricity which may only exist for 10% of the year.
The Companies shortfall in capacity coverage is greater than their shortfall in energy coverage because the Companies are short peaking capacity rather than baseload capacity. Peaking capacity plays a key role in meeting customers demand in the few hours of the year when load is highest, However, it plays a tiny role in meeting customers annual energy load, again because it only generates energy in a very few hours of the year. Therefore, the Companies have a small physical shortfall in energy coverage, despite a large physical shortfall in capacity coverage, because they have a shortfall in peaking capacity rather than in load following or baseload capacity.
Peaking capacity typically supplies energy in less than 10% of the hours of the year. For example, the capacity the Company retired in 2012 had an average capacity factor of 11% in 2011 and a projected capacity factor of less than 1% in 2012, indicating that it operating at full capacity in very few hours each year.
The “capacity factor” which Mr. Hornby expresses as a percentage, is simply the percent of the time that a power plant is actually producing electricity. The point he makes in the last paragraph is that the three old, expensive coal burners that FirstEnergy shut down in WV were only running about 11% of the time, because they could only operate profitably at peak times when PJM’s electricity prices were the highest.
So the only thing that Mon Power needs is peak capacity, not base load electricity production. The Harrison Power Station is the worst kind of power plant capacity that a rational person would use to meet peak demand only 10% of the year.
There are lots of ways to meet peak demand without buying a huge power plant with very high fixed costs that must be run constantly to produce a profit. The best way is to shave the peaks off of demand, shifting electricity use to other times of day. This is called demand management. There are lots of techniques that companies use to manage demand.
Many companies specialize in managing demand for power companies. These companies provide what is called demand response mechanisms designed to shift demand away from peak periods. WV Citizens Action Group expert Cathy Kunkel explained in her testimony:
First, there are some demand response programs – direct load control programs and interruptible tariffs, for example – that can be offered by utilities independent of third party curtailment service providers. Direct load control programs are common throughout the country; a recent study documented 52 utilities running residential load control programs in 2010. Typically in such programs, the utility installs (at the customer’s request) a switch on central air conditioning units that allows the units to be remotely cycled; the customer is provided an installation incentive and annual bill credits. Since 2008, all Maryland utilities except Potomac Edison have implemented residential direct load control programs that were cost-effective even though the PJM capacity price in Maryland is less than the capacity cost of the Harrison plant.24 Interruptible tariffs are also commonly used by utilities nationwide, including Appalachian Power in West Virginia. The Federal Energy Regulatory Commission’s 2009 state-by-state assessment of demand response potential found the potential for 4.5% savings in West Virginia from direct load control programs and interruptible tariffs in the absence of advanced metering infrastrscture. This is likely an underestimate given that it doesnot include any potential DR from small commercial direct load control programs.
Earlier in her testimony, Ms. Kunkel also pointed out that simply reducing demand by investing in energy efficiency would also reduce peak demand.
Q. WHAT FRACTION OF THE COMPANIES’ CAPACITY SHORTFALL COULD BE MET THROUGH ENERGY EFFICIENCY BY 2026?
A. Under this scenario, energy efficiency would save 3 18 MW UCAP by 2026 – or more than a quarter of the Companies’ actual capacity shortfall of 1211 MW.
Q. HOW WOULD THE COST OF EE/DR COMPARE TO THE COST OF THE PROPOSED RESOURCE TRANSACTION?
A. It would be less expensive. The levelized cost of energy efficiency is about a third to a half of the cost of Harrison or market purchases (7.4 cents per kWh and 7.5 cents per kWh, respectively, according to the Companies’ analysis). As a result, any scenario which uses energy efficiency to substitute for some fraction of the proposed resource transaction or market purchases will have a lower levelized cost.
Mr. Hornby shows clearly that buying capacity from other companies on the PJM capacity markets or entering into a power purchase agreement directly with a gas-fired power plant or plants would be the best way for Mon Power to acquire any additional generating capacity it needed for peaking capacity.
Mr. Hornby’s testimony begins on page 40 at this link. His testimony is well worth the read if you want to see how FirstEnergy misrepresented every aspect of the current situation in its thoroughly bogus “resource plan” that it filed at the WV PSC last fall. I covered this so-called plan in a post at the time.
Mr. Hornby goes through the FE “resource plan” with a fine tooth comb and points out that the company completely distorted numbers and scenarios to fit its own desired conclusion that the Harrison transfer was the best option. In many cases, they simply made up numbers that can be easily checked, such as capacity factors of natural gas power plants and the capacity factors of the Harrison plant itself. They just made stuff up.
So FirstEnergy needs peaking capacity, not base load capacity like the Harrison plant. I’ll let Mr. Hornby have the last word:
First, the Companies have a shortfall in peaking capacity; they do not have a shortfall in baseload capacity, and they do not have a reliability problem. The Companies have time to find a reasonable strategy to address that strategy.
Second, acquiring 1,576 MW of Harrison capacity will limit the Companies’
ability to take advantage of other, less expensive with less fixed cost risk over the next several years.
Demand response, efficiency investments and purchase of power from natural gas power plants were never considered by FirstEnergy. They were certainly never considered as a flexible, mixed portfolio that could be used to minimize risk for WV rate payers. Risk is the next big issue that I will address in tomorrow’s post.