How Heat Pumps Push Power Plant Efficiency

12 February 2012

Interview with

Professor Andrew Knox, University of Glasgow

Kat - We've heard how a Thermo-Electric Generators or TEGs could be a useful way to turn waste heat energy back into useful electricity in small scale settings like cars or things like that, but there are also settings on a very big scale, for example a power station which throws away huge amounts of heat up its cooling tower.  So, could Thermo-Electric technology help here?  Now we're going to talk about it with Andrew Knox.  He's Professor of Power Electronics Renewable and Sustainable Energy at Glasgow University and he's looking at feasibility of using this kind of technology, but more or less in reverse.  So good evening, Andrew.

Andrew -   Good evening, Kat.

Kat -   So tell me a little bit about for a start, the kind of scale of heat loss we're talking about in a power station and why it would be attractive to try and use thermo-electric technology?

Andrew -   Much of the UK's power generation comes from what's called thermal power stations and that's where we burn fossil fuels; gas, coal, et cetera.  The generation of electricity from that process is subject to various thermo-dynamic restrictions.  It's the same basic problem that the thermoelectrics have got where you've got a hot and a cold side, and the maximum efficiency is a function of those as determined by the Carnot cycle.

In a conventional thermal power station, typically what happens as you burn the fuel, let's say coal for example, and that raises steam in a boiler.  That steam is then rooted through a turbine and the force of the steam going through the turbine turns an electrical alternator which is what generates electricity.  The steam goes through a series of different turbines, each at progressively lower pressures and at the end of that process, when you've extracted about as much of the useful work as you can out of the steam, what you have to do is recondense the steam back into water and that water is then returned to the boiler to be used again.

The water that's used as the working fluid for this power generation process is exceptionally pure and the reason for that is that any impurities that would be in normal tap water for example, would convert your turbines into Swiss cheese very quickly.  I mean, it would really attack the turbine blades.

So this process that's going on - boiling water, driving it through a turbine and then condensing it back to water, and reboiling it back in the boiler - that process is called the Rankine cycle and it's subject to limitations.

If you drive past a large thermal power station, let's take for example Ferry Bridge down the M1.  When it's working, you will see large quantities of steam being rejected from the cpower stationooling towers and the cooling towers are these big huge concrete structures.  That's basically dropping the energy out of the steam and back to the environment, so that's wasted heat.  And on the best modern thermal power stations, their efficiency is about 46 or 47%, so more than half of the energy that's used is being rejected to the environment.

Kat -   So in a power station we have a lot of things that are very hot and things that you're trying to cool down, so there does seem to be capacity to use thermo-electric technology, but you're proposing a slightly different way, not using the difference between heat and cold to generate electricity but something else.  How do you think we could introduce thermo-electric technology into a power station to make it more efficient.

Andrew -   One of the things that Laurie touched on is the semi-conductor Thermo-Electric Generators.  One of the properties of these Thermo-Electric Generators is that as Laurie described it, if you apply a temperature difference then you will generate a voltage from them, but if you apply a voltage to the same material, the same device, you can actually use it to shift heat from one side to the other.  In the context of the power station, in the condenser of the power station, where the used steam is converted back into water, as that steam gives up its energy and condenses away from the heat of evaporation, by using some thermoelectrics, we can actually capture some of that energy and re-inject it back into the power plant rather than rejecting it to the atmosphere.

Kat -   So you're talking about actually taking the heat and putting the heat back into the system to generate a bit more electricity again.

Andrew -   Yes.  It's very low grade heat but it's conceptually thought about as the energy released as the steam gets converted back into water.

Kat -   So this would mean you get more bang for your buck.  So for the amount of coal that you're burning, you would end up getting more electricity out the other end.

Andrew -   Correct.  You would end up having to burn less coal for a given amount of electrical output from your generators.

Kat -   So this sounds brilliant, but how realistic is it?  What are some of the challenges that are there to try and implement this kind of technology?

Andrew -   There are two big challenges - the first one is the engineering of the condenser itself.  In other words, you need to get tens of thousands of these semi-conductor devices, properly arranged in the steam flow to maximize the heat transfer and the second thing is to optimise the electronics that would be used to drive this process.  The Peltier effect as it's called, which is the property of the semi-conductor material when it's in use like this, that has a co-efficient of performance.  That is determined by the difference in temperature between the hot and the cold side.  In general, as the temperature difference increases, the coefficient of performance decreases.  So this is not something you can use up to a temperature you like.  This is really to be used only at the low temperature and for a typical large scale power plant, the steam coming out of the last stage of the turbine is about 30-35 degrees C.

Kat -   So, do you think the kind of technologies you're talking about could actually be fitted into the power stations we have now or will it take a new breed of power station?

Andrew -   I think it's suitable for both new build and for retro fit.  The efficiency using today's devices with today's materials means that we are just about at the breakeven point.  There's two contributing factors which help us here, one is the power station efficiency itself as we go through supercritical, very high temperature, very high pressure steam, that helps us.  Also, as the materials in the semi-conductor devices improve, that also helps the overall efficiency of the heat pumping and therefore, the coefficient of performance of the heat pump.

Kat -   I know this is kind of "how long is a piece of string" question but if all the research goes well, when do you think this kind of technology might be able to be brought in?

Andrew -   I would think you're probably looking at 10 years.  To do a large scale power station or like Ferry bridge or Drax, that would be ambitious to say the least.  But if you were to go for a relatively modest, 1 megawatt or 5-megawatt power station and I would expect to see in 10 year's time, that would be available, with a modified Rankine cycle.

Kat - And using the benefits that could be gained in efficiency in saving money on coal would actually make it worth investing in this kind of technology.

Andrew - Yes, I do. Even if you're only talking about a couple of percent efficiency, increase in the overall coal to electricity conversion, that couple of percent represents a huge amount of power if you take all of the lifetime of the power station of maybe 25 years, coupled to the fact that energy prices are going to continue to increase in my view.

Kat - So that's definitely something worth investigating and investing in.  Thank you very much, Andrew.  That's Professor Andrew Knox from Glasgow University.

Add a comment

This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.