Clean Power Plan and the Heat Rate Fairy

As you must know, the EPA’s release of its Clean Power Plan on June 2nd of this year includes four “building blocks” to achieve a 30% reduction of carbon dioxide emissions by 2030, using a reference year of 2005.  Those building blocks are as follows:

1. Improve the heat rate (efficiency) of coal-fired power plants by 6%
2. “Re-dispatch” natural-gas generators to achieve a capacity factor of 70%
3. Development and preservation of clean sources, including nuclear, hydro, and renewable sources
4. Demand side energy efficiency

Does the EPA have any engineers on staff?  Did any of them provide input and/or oversight for the Clean Power Plan, specifically for 1 above?

I am not a power plant genius, but I took many post-graduate courses in thermodymics, heat transfer, fluid dynamics, and studied thermal efficiency and optimization of power plant efficiency for my thesis.  My first reaction to a six percent increase in efficiency is they will need heat rate fairies, unicorns, and pixy dust to pull this off.  Six percent may seem reasonable to the uninformed simpleton, but it is simply not achievable for a number of constraining factors.

Any credible first-semester course in thermodynamics includes a shot of thermal efficiency in the first week.  There are many power-generating cycles including Otto (gasoline engine), Diesel, Brayton (jet engine), Stirling (external combustion engine), Rankine (power plants), and Carnot (the perfect, unachievable engine used for academia only).

In my undergrad thermo class, we started with the Carnot cycle.  But it doesn’t really matter because all cycles (all engines, including power plants) have their efficiencies dominated and limited by the difference in their high temperature and their low temperature.  For example, a typical coal-fired plant uses steam at 1000 degrees F (1000F) and a pressure of about 2400 pounds per square inch (psi).  The low temperature is dictated by the type of heat sink used: river, lake (natural or impoundment), ocean, or air.

We can’t do anything meaningful about the temperatures of these heat sinks, realistically.  On the other end, where heat is added, we are limited by the characteristics of the materials used; namely old fashioned carbon steel.

Thermodynamic tricks have been deployed over the years to improve power plant efficiency.  An analogy might be a turbocharger or a supercharger for a gasoline or diesel engine.  For power plants, rather than using one turbine, two are used with a reheat between them, and bleeding some steam off to preheat boiler feed water.  This is shown in the diagram (Figure 1).

You may think, “Boy, that is over my head.”  I would respond by saying, “You should see an energy-efficiency-program logic model.”  I provide one of those courtesy of Kansas PBS (Figure 2).  Not to worry – EE program versions look the same.

The addition of these efficiency enhancers is what I studied for my thesis.  More turbines and more reheats can be added.  More feed water heaters can be added.  They can be moved around: further upstream (hotter steam), further downstream (cooler steam), etc.

Like most things with efficiency, the laws of diminishing returns apply.  The second reheat or feed water heater adds less benefit, and much less benefit per dollar invested, compared to the first.

The moral of this story is, power plants are designed as they are for a reason; at 35-40% efficiency, they are already optimized from a cost/benefit perspective.  For reference, a 100F rise in steam temperature (a big deal) produces one percentage point improvement in efficiency.  Increasing from 1000 psi to 1500 psi improves efficiency by about 1.5 percentage points.  These are relatively small increases for monumental leaps in operating parameters – which cannot just be “flipped on”.  The physics behind all this is very stubborn.

Theoretically, adjusting these parameters produces the greatest impact, but it is only theoretical.  Realistically, plant operators are stuck with comparative crumbs.  Indeed, this letter from the Wyoming PSC to Gina McCarthy, the EPA administrator, outlines the (non) impacts of the EPAs proposed improvements, plus some provided by Sargent and Lundy.

To achieve the six percent improvement, the EPA targets four percent of that to come from “best practices” and two percent from equipment upgrades.  This is utterly impossible.  The Sargent and Lundy report referenced in the Wyoming letter includes impacts from neural network, intelligent soot blowers, air heater and duct leakage control, condenser cleaning, boiler feed pump rebuild, and some other best practices.  Equipment upgrades include economizer replacement – which would be like replacing all ductwork in your house – a monumental feat – a turbine overhaul, and “combined VFD and fan”.  These practices and upgrades are nibbling around the edges compared to magically increasing steam pressure by 500 psi, which produces a paltry 1.5% improvement.

Moreover, the North American Electric Reliability Corporation – NERC – reports that when developing the six percent target, the EPA ignored the following very relevant considerations:

• Generation-reducing effects of post-combustion environmental controls (yes, adding scrubbers degrades efficiency/heat rate – huge fans must be added)
• Generating characteristics like supercritical boilers, fluidized bed, gasification combined cycle, capacity and age, fuel characteristics including moisture content
• Degrading impacts of lower load factors – namely that caused by “re-dispatching” to natural gas plants – see number 2 above.

For a great description of coal plant operation, see this website, citzendium.org.

Considering all this, the coal-burning fleet will be doing very well just to maintain the heat rates as they have today.  It is absurd to think a tune-up, turbine overhaul, and some “smart” or “intelligent” controls are going to achieve a six percent improvement.