Of the 172 supercritical-presssure (SCP) power units (predominantly coal or lignite fired) put into operation since the beginning of 1998 and currently under construction (due to be commissioned by the end of 2014), around 70 are in China, 31 in Europe, 19 in Japan, 15 in Southern Korea, and 23 in other countries of South-Eastern Asia, India, and Australia, while 14 have been ordered for the USA and Canada.

As of 2004, there were over 600 SCP units operating in the world, including some 240 in the countries of the former USSR, nearly 170 in the USA, about 100 in Japan, and approximately 60 in west, central, and southern Europe.1,2

In the 1990s, the pace in world power engineering progress was set by Japan and Germany. But in the 2000s, China and South Korea joined the club and even though their recently completed SCP projects have included imports of key equipment from European, Japanese and US manufacturers, this equipment is now mainly produced by domestic (Chinese and Korean) manufacturers, albeit largely on the basis of foreign licenses.

The first new SCP power unit in the USA for over fifteen years was put into operation in 2007 – MidAmerican’s Walter Scott Jr (formerly Council Bluffs) Energy Center unit 4, with a rated output of 790 MW and steam conditions of 25.3 MPa, 566/593 ºC. Instructively, the main equipment for this unit, as well as all the other SCP power units to be launched in the US in the near term, is produced by Japanese or, in a few cases, Western European manufacturers.

In the early 1990s, on a wave of success emanating from the putting into operation of two 700 MW ultra-supercritical (USC)* units at Kawagoe in Japan (steam conditions: 31 MPa, 566/566/566 ºC), it was assumed that further evolution of steam-turbine-based power generation technology would mainly follow the direction of building USC power units with double steam reheat and a concurrent rise of both the steam pressure and temperatures.

In reality, the main direction of progress has turned out to be different.

Evolution of steam conditions

Figures 1a, 1b and 1c show the evolution of rated steam conditions for SCP power units with entry-into-service dates from 1998 to 2014. Among these, the only USC (according to my definition, see footnote*) unit is Avedøre 2 in Denmark, with rated capacity 390 MW and steam conditions of 30 MPa, 580/600 ºC, and the only coal-fired double-reheat unit is Nordjylland 3 (29 MPa, 582/580/580 ºC), also in Denmark, both developed by the same Danish electric power utility Elsam (now Dong). A year earlier, Elsam had commissioned an oil/gas-fired double-reheat SCP unit, Skærbæk 3. Compared with other SCP power units of the same vintage, these Danish units rather stand out as the exceptions.

In practice the SCP power units of the “new generation” feature relatively moderate main steam pressure values – see Figure 1a.

Take the three new 660 MW SCP units of the repowered Torrevaldaliga Nord in Italy, which was converted from oil to coal. The original units had worked out their rated lifetimes and the repowered units preserved the former main steam pressure, 24 MPa, but with elevated main and reheat steam temperatures – to 600/610 ºC. Similarly, practically all the Japanese SCP power units put into operation in the last decade have been designed for a main steam pressure of 24-25 MPa, while rated steam temperatures have been progressively increased to 600/610 ºC.

With the increase in the main steam pressure level, its further rise (with the same steam temperatures) gives less and less increase in the efficiency and available energy of the thermodynamic cycle. Also the higher steam pressure demands thicker boiler tubes, manifolds, steam-lines, and turbine casings, as well as feedwater boiler pumps of larger capacities. All these factors make the units more expensive and less flexible – that is, coping with transients becomes more difficult because of higher and fluctuating thermal stresses in the thicker components. So there is a downside in pushing on with a major shift to USC steam pressures. Nevertheless, many of the European SCP power units due to enter operation in the early 2010s are designed with main steam pressures in the range 27 MPa to 29 MPa, with a rising trend over time.

As to main and reheat steam temperatures (see Figures 1b and 1c), if a country has not had previous experience with SCP technology, the first units are commonly built with moderate steam temperatures, in the region of 538-566 ºC, as in the case of Australia, India, Thailand, Mexico, and South Africa, for example. The first SCP units constructed in China (for example, Shidongkou II, 2 3 600 MW, built in 1992) were designed for steam parameters of 24.2 MPa, 540/566 ºC, and Suizhong (4 3 800 MW), put into operation in 2002-2003, employs steam parameters of 24 MPa, 540/540 ºC.

Similarly, the first series of South Korean 500 MW SCP units were designed with steam conditions of 24.9 MPa, 538/538 ºC, and the first SCP power unit built by Hitachi in Canada (Genesee 3, with rated output of 495 MW, launched in 2005) also has pretty moderate steam conditions of 25.1 MPa, 570/568 ºC, even though Hitachi already had rich experience in designing and delivering SCP units with steam temperatures of up to 600/610 ºC.

As SCP experience is amassed, the succeeding units in a country tend to be designed with higher steam temperatures, the rate of increase depending on the extent of conservatism – that is, how long the process of accumulating experience lasts. So in South Korea, after a large series of 500 MW SCP units with steam parameters of 24.9 MPa, 538/538 ºC, the main and reheat steam temperatures for new 500 MW and 800 MW units were raised to 566/566 ºC and then to 566/593 ºC. Similarly in China, a series of SCP power units with a steam temperature of 540 ºC was followed by a series of SCP plants with steam temperatures of 566/566 ºC and then 600/600 ºC. Of significance is that these increases were not embarked upon until a good deal of work had been put into mastering the higher steam temperature levels by Japanese and Western-European power equipment manufacturers at power plants in Japan, Germany, and Denmark.

According to calculations by Siemens, the second steam reheat increases a power unit’s efficiency by approximately 0.8%. However, the arrangement of the boiler superheaters becomes more complicated, with the addition of a high-temperature section in the turbine and, what is most significant, additional high-temperature steam-lines of a large diameter. As a result capital costs rise markedly. That is why, for relatively low supercritical steam pressures (ie below 30 MPa) and where there is potential for raising steam temperatures, the abandonment of second reheat looks quite justified.

A side view of a typical SCP boiler produced by Babcock-Hitachi KK (BHK) for a 1000 MW Japanese unit, Haramachi 2, with steam conditions of 24.5 MPa, 600/600 ºC, is shown in Figure 2. The superheater and reheater pendant sections are manufactured from austenitic heat-resistant 18% Cr steels, but all the remaining boiler tubes, pipes, and steam-lines are made of advanced ferritic-class steels with 9% Cr content. The design features were firmly based on experience with the 1000 MW Matsuura 2 SCP unit, which has steam parameters of 24.1 MPa, 593/593 ºC, and then repeatedly reproduced at other Japanese SCP units with higher steam temperatures, including Tachibana-wan unit 2, with rated output of 1050 MW and record-breaking (for that time) steam temperatures of 600/610 ºC.

Turbine technology trends

Raising the rated steam temperatures demands a transition to new materials for steam turbines, too. In doing so, cast casings are replaced where possible with forged ones. Design materials used by Japanese manufacturers for major components of modern steam turbines with rated steam temperatures of 593-600 ºC are shown in Table 1 and compared with those used in steam turbines with rated steam temperatures of 538-566 ºC.

New materials for supercritical-pressure power units with advanced steam temperatures have also been developed and researched under the European COST (Co-Operation in the field of Scientific and Technical research) programme.

The results of these developments and research programmes, as well as those in Japan, should make it possible to commission the first SCP units with steam conditions of 28 MPa, 630/650 ºC in the early 2010s, and then to begin mastering USC units with steam parameters of 35 MPa, 700/720 ºC.

According to the European AD700 project, pilot coal-fired power units with these parameters and rated gross output of 550 MW should be launched in Denmark and Germany in 2014-2015. The net efficiency of such units is expected to be around 50-51%. Compared with the existing German fossil-fuel power plants with their average efficiency of 38%, these new units will produce electricity with about 16% less CO2 emissions. Compared to power plants with the world average efficiency, about 30%, the CO2 emission decrease is about 40%.

The rated output for modern SCP power units already launched or to be launched in 1998-2014 varies from 400 MW up to 1000-1100 MW. In this range, the turbine configuration can vary widely and it seems reasonable to select turbine unit sizes that minimise capital costs for the turbine island.

For large steam turbines, the problem of coping with thermal expansion restricts the maximum number of cylinders that can be used to five: the HP cylinder; double-flow IP cylinder; and three double-flow low-pressure (LP) cylinders.2,3 This is the scheme adopted for modern tandem-compound (TC) 1000-MW-class SCP steam turbines currently in service that have been produced by the transnational turbine manufacturers Alstom and Siemens for Western European power units with a grid frequency of 50 Hz (ie 3000 rpm). Examples include Lippendorf R and S, Boxberg Q, and Niederaussem K.

For 60 Hz (3600 rpm), because of the bigger circumferential forces acting on the LP last stage blades (LSBs), it was impossible until recently to provide the necessary annular exit area using three full-speed LP cylinders, that is, preserving the previously mentioned five-cylinder configuration.

That is why all the 60 Hz Japanese SCP units of 900 to 1050 MW capacity incorporate cross-compound (CC) steam turbines, with the HP and IP cylinders located on the full-speed (3600 rpm) shaft-line and two LP cylinders on the low-speed (1800 rpm) shaft-line.

With the emergence of longer full-speed LSBs with bigger annular exit areas (see below), it became possible to design 1000-MW-class TC steam turbines with a five-cylinder configuration for 60 Hz as well.

All three leading Japanese producers of large steam turbines – Hitachi, MHI, and Toshiba – have such turbines at their disposal.2

For 50 Hz and relatively low vacuum in the turbine condenser, it became possible to have a 1000 MW TC steam turbine with four cylinders, that is, with two double-flow LP cylinders. Just this configuration was realised by Siemens at Yuhuan (four 1000 MW SCP power units).

The same scheme is adopted for turbines produced by Alstom for German SCP units Neurath F and G, Boxberg R, Moorburg A and B, Datteln 4, and Staudinger 6, as well as Port of Antwerp in Belgium – all with a gross capacity of about 1100 MW, Figure 3. A similar SCP unit for Maasvlakte in the Netherlands will employ a five-cylinder version same Alstom steam turbine, with three double-flow LP cylinders, making use of colder cooling water and, as a result, deeper vacuum in the condenser.

If the turbine capacity does not exceed 700 MW, the HP and IP sections can be united in an integrated HP-IP cylinder, so the turbine’s total number of cylinders decreases to three (with two LP cylinders). Such a configuration is employed, for example, at Japanese SCP unit Tomato-Atsuma 4, with a turbine manufactured by Hitachi – Figure 4. If the set capacity decreases to 600 MW with a rotation speed of 3000 rpm, the turbine could be designed (according to MHI) with just two cylinders: one HP-IP cylinder and one double-flow LP cylinder.

Despite the many indisputable advantages of an integrated HP-IP structure, turbine manufacturers such as Siemens and Alstom still prefer to design their turbines with separate HP and IP cylinders, even if this makes for longer machines. To decrease the overall turbine length, both Siemens and Alstom design their large turbines with shared journal bearings for adjacent cylinders (as well as for the turbine and generator) – see Figure 3.

The most efficient steam turbines for SCP units put into operation by the beginning of this century are presented in Table 2. Of significance is that the very similar efficiency values were achieved with very different steam conditions. This suggests that, at least for some turbine types, there are considerable reserves for raising efficiency by reducing energy losses in the steam path. Moreover, there exists a reserve for raising the turbine efficiency to 50% and beyond, even without transition to USC steam parameters. Table 3 presents the guaranteed heat-rate data for steam turbines manufactured by Chinese turbine works based on foreign licences.

Along with the efficiency (or heat-rate) data for turbines as a whole, also of interest are the internal efficiency values for individual turbine cylinders (sections). So, for the 900 MW Siemens turbine of the Boxberg Q SCP unit, the acceptance field tests gave internal efficiencies for the HP and IP cylinders of 94.2% and 96.1%, respectively. These figures, even though record-breaking, can be considered consistent with the evolutionary stage that modern turbomachinery finds itself in.2,4

Noteworthy is that the internal efficiencies of individual turbine sections depend only to a small extent on steam parameters, more important is the voluminous amount of steam flow through the sections. This makes it possible to meaningfully compare the data for turbines with different steam conditions.

So, for example, after refurbishment by Alstom, the HP-IP cylinder of the GE steam turbine at the J. K. Spruce power plant in the USA, the acceptance tests gave internal efficiencies for the HP and IP sections of 93% and 95.7% respectively, whereas for the best steam turbines of this type manufactured in the 1980s the corresponding values were 90% and 93%. Another example is the 750 MW turbine of the Mehrum unit in Germany, where, after refurbishment by Siemens, the internal efficiencies for the HP and LP cylinders were 93.6% and 89.9%, respectively, compared with the pre-refurbishment figures of 85.5% and 87.2%.

Such increases in internal efficiency are attained by reducing all the various kinds of energy losses in the turbine steam path.2 Apart from using new blade profiles, traditional cylindrical blades are replaced with “three-dimensional (3D)” blades, bowed and twisted throughout the entire steam path.

In addition, in the late 1990s Siemens developed a new concept for shaping the turbine steam path: setting the degree of reaction for each stage individually, varying it from 10% to 60%, to minimise energy losses.

This approach was first applied in full measure to the 1000 MW class SCP turbine at Niederaussem K and since then has been implemented on all Siemens new-build projects.

While Siemens came to this concept from designing reaction-type turbines, GE traced a similar path for impulse-type turbines. Their concept of “Dense Packing” means increasing the number of stages and their reaction degrees compared with a purely impulse-type steam path. In particular, this idea is realised in steam turbines designed and manufactured by the Korean company Doosan Heavy Industries for the newest Korean 500 MW and 800 MW SCP power units, based on the GE licence.

Three-dimensional design is applied not only to the turbine’s steam path, but also to all the bladeless channels and paths: steam-admission and exhaust sockets and chambers; intercasing and bleeding chambers; crossover pipes; and so on. Decreases in steam pressure drop in these parts have a considerable influence on turbine efficiency.

Various types of parasitic steam leakages are reduced without increasing the danger of rubbing in the steam path by using different types of advanced gland seals, including so called “retractable packings.” For such seals, at the initial start-up stages, when the probability of rubbing is maximum (because of vibration, thermal bowing of the turbine shaft and/or casing, and other possible circumstances), the radial clearances in the seals are maximum. And only when the turbine begins operating under load, does the steam pressure difference across the seal segments displace them inward, that is, into the working, “closed” position, decreasing clearances to the set values. Gland seals of this type were first tested at steam turbines in operation in the 1980s. Based on more than two decades of operating experience with hundreds of different types of steam turbines, retractable packings have shown themselves to be a very cost-effective and reliable way of preventing wear of seals and maintaining high steam turbine efficiency.

These gland seal systems have been further improved by using built-in brush bristles sandwiched between two solid faceplates. The brush material is a cobalt-based super-alloy. Several thousand extremely fine bristles, with the wire diameter in the range of 0.1 mm to 0.15 mm, are packed together, forming a barrier against the leakage steam flow. The bristles are inclined radially in the direction of the shaft rotation to prevent them from picking up on the rotor. Such seals are widely used, in particular, by Alstom.

Another type of contact gland seal was developed by MHI and passed long-term field tests under realistic operating conditions at the experimental 105 MW steam turbines of the combined-cycle unit at MHI’s Takasago works.

A further key factor in raising turbine efficiency is to decrease the energy loss by reducing the exit steam velocity through use of longer LSBs, with bigger annular exit area. Nowadays, practically all the world’s leading turbine producers have at their disposal titanium LSBs that make it possible to increase considerably the exit area. An additional advantage of these blades is their higher resistance to erosion-corrosion compared with steel blades.

In the case of low-speed LP shaft-lines of cross-compound steam turbines, there still exists significant margin for increasing the length of the steel LSBs, but the concept of cross-compound steam turbines as such has essentially become obsolete. Tables 4-6 present data for the LSBs used by the leading turbine producers for newly designed and refurbished steam turbines with rotation speeds of 3000, 3600, and 1800 rpm (the latter as applied to large cross-compound turbines with grid frequency of 60 Hz).

Supercritical-pressure turbines are usually equipped with advanced regenerative systems that heat the boiler feedwater up to the final temperature of 280-290 ºC and for USC units (for example, Avedøre 2) up to as high as 320 ºC. The number of feedwater heaters (FWHs) varies from eight (four LP FWHs + deaerator + three HP FWHs) to 11.

Boiler-feed pumps (BFPs) in modern SCP units are typically driven by condensing turbines with a unit capacity of up to 20 MW. Many SCP power units have three turbine-driven BFPs each providing 50% of the maximum rated feed capacity or two 50% turbine-driven BFPs and one start-up motor-driven BFP with 30% of maximum capacity. Some SCP power units (mainly those in the lower power range, 500-600 MW) are equipped with three motor-driven BFPs with 50% capacity each.

Boiler technology trends

The SCP steam generator depicted in Figure 2 is of the two-pass type, with spirally wound water walls of multi-ribbed tubes. The boiler operates with sliding pressure along the whole water/steam path and features large-capacity low-NOx burners and two-stage fuel combustion. It is equipped with large-capacity roller-type pulverisers with rotating classifiers that significantly improve the pulverised coal fineness. Some boilers of this type also feature simplified connection, without headers, between the water walls and subsequent heated surfaces.

Some SCP boilers produced by Mitsubishi Heavy Industries (MHI) employ a similar design concept (two-pass steam generator with spirally wound water walls) – for example, as deployed in the Japanese 700 MW Tsuruga 2 unit, whereas other MHI SCP boilers use a vertical tube water wall furnace – for example, as at the 600 MW unit Hirono 5 or the 900 MW Maizuuru 1 unit. Major advantages of this system, according to MHI, are the following:

• Because of the lower mass velocity in the vertical tubes, the pressure loss in the furnace system is reduced leading to savings in auxiliary power consumption.

• Thanks to the greater simplicity of the vertical tube arrangement, the furnace can be supported more easily, with fewer fittings needing to be attached to provide support.

• For coal-fired boilers, ash adhesion to the water wall tubes cannot be avoided over time, however the vertical arrangement helps the slag fall naturally and reduces the amount of ash that adheres.

• Because the hydraulic friction loss in vertical tubes is small compared with the total pressure loss, the flow variations across the furnace perimeter are also small.

Similar considerations have been articulated by Mitsui Babcock, Siemens, and others.

As distinct from two-pass boilers designed and manufactured by Babcock & Wilcox, BHK, MHI, etc, some other boiler manufacturers produce tower boilers. Their furnace water walls employ spirally wound tubes in the lower part and vertical panels higher up. A typical example is the Alstom boiler for the 965 MW Niederaussem K. There are similar boilers at Waigaoqiao II in China as well as in the new 500 MW and 800 MW SCP units at Tangjin and Yonghung in South Korea.

Just as in the case of turbines, minimising dimensions is an important factor in reducing boiler capital costs.

In the Reference Power Plant North Rhine Westphalia (RPP NRW) project, with a net output of 550 MW, the dimensions of three different boiler types were compared, assuming the same coal mills and pulverisers, the same low-emission burners, and the same final feedwater and flue gas temperatures. On economic and engineering grounds the tower-type boiler emerged as the preferred alternative. But for other SCP projects two-pass boilers may exhibit better economics.

In the case of the RPP NRW exercise, it was an innovative horizontally fired Benson-type boiler that was considered as the alternative to the traditional tower-type and two-pass boilers. This was initially developed by Siemens for combined cycle applications. Then it was targeted at the AD700 project (USC steam-turbine unit with the steam parameters of 35 MPa, 700/720 ºC). The pilot, subcritical-pressure, HRSG (heat recovery steam generator) of this type was supplied by Siemens and is in operation at the Cottam combined cycle plant in the UK. It could be considered a possible basis for developing supercritical-pressure combined-cycle units in the future.

The low mass flux vertical tube Benson technology, with rifled tubing, suggested by Siemens, is also being used in the boiler produced by Foster Wheeler for Longview in the USA, due to start up in 2011.

This technology is also used in the world’s first supercritical-pressure CFB (circulating fluidised bed) boiler, also supplied by Foster Wheeler, which entered commercial operation at Lagisza in Poland in July 2009. This power unit is rated at 460 MW gross, with steam conditions of 27.5 MPa, 560/580 ºC.

According to acceptance field tests, the boiler efficiency of modern large SCP units is typically around 92-94%. The field tests also demonstrate that such units have high flexibility. Figure 5 depicts a hot start-up at the 1000 MW Matsuura 2 SCP unit (with BHK boiler and MHI turbine), using automated control (Self-Tuning ART System (STARTS)).

The boiler’s minimal stable load is equal to 30% of the rated output. Above 50% of the rated output, the load change rate can be 4%/min, which means that the unit can be operated in semi-peaking mode, with regular deep unloading or shut downs during periods of low demand at night and during weekends.

Some SCP units allow load change rates as high as 7-8% of the rated output per minute. For other SCP units, intended for operating in baseload mode, the permissible load change rates are usually set lower: typically up to 2 %/min for scheduled load changes and up to 5 %/min for unplanned load changes (eg, as in the case of Genesee 3, Canada’s first SCP unit).

Lower emissions and higher efficiencies

For modern SCP units the main measures to reduce NOx emissions are two-stage fuel combustion and use of low NOx burners. Many boilers are in addition equipped with selective catalytic reduction (SCR) systems that reduce emissions immediately downstream of the boiler (as in Figure 2).

The air quality control system for WSEC 4’s BHK boiler is presented in Figure 6. With a projected NOx concentration in the flue gas downstream of the boiler of 128 ppm (based on 6% O2), the concentration at the SCR outlet is 42 ppm. The outlet dust concentration is 11.6 g/Nm3. The flue gas is also desulphurised using a spray dry absorber.

The guaranteed emission values differ widely among SCP units depending on coal type. So, for example, for the 460 MW Polish plant Patnow II, with steam conditions of 25 MPa, 540/565 ºC, the maximum allowable emissions of both NOx and SO2 are 200 mg/Nm3 and the ash content must not exceed

30 mg/Nm3. This meets EU directives. The three 660 MW Torrevaldaliga Nord SCP units in Italy have steam conditions of 25 MPa, 600/610 ºC and the guaranteed emissions for both NOx and SO2 are less than 100 mg/Nm3 on an hourly basis, with particulates not exceeding 15 mg/ Nm3.

For pilot 550 MW USC power units envisaged in the AD700 project the expected emission values for NOx, SO2 and particulates are 80, 70, and 10 mg/Nm3, respectively.

To shed light on underlying trends in efficiencies for modern SCP power units (which have either entered operation since 1998 or are about to enter operation) published net LHV efficiencies have been plotted against “effective” steam temperature, tef. This is defined as follows:

tef = (tsh + 0.78 x trh)/1.78 + 3.4 x (psh – pshav)

where tsh is the main steam temperature, trh is the reheat steam temperature, weighted with a coefficient of 0.78 based on the assumption that the share of the turbine output generated in the HP section (before reheat) is about 22%, and psh and pshav are, respectively, the main steam pressure and its average value for the sample (26.4 MPa for this set of units). When correcting for the difference between psh and pshav it was assumed that a difference of 5 MPa causes the same change in efficiency as a difference between main and reheat steam temperatures of 17 ºC. This assumption is based on the efficiency data for the newest Japanese SCP power units given by Ishikawajima-Harima Heavy Industries (IHI) – see Figure 7.

The results of processing the efficiency data in this way are presented in Figure 8. This shows that with main and reheat steam temperatures rising to 600-620 ºC the power unit net LHV efficiency can reach 46-48% without the use of USC steam pressure and double reheat.

Towards 50% efficiency

During the last twenty years or so, it is really only Japan, Germany, and Denmark that have consistently kept on implementing new advanced steam-turbine-based SCP coal and lignite fired power units, with South Korea and China, now following them. After a long stagnation period, new SCP power units are presently being commissioned in such countries as the USA and Italy that once were among the leaders, but now major power equipment for these units is being delivered by foreign manufacturers. The first SCP power units are also being put (or shortly to be put) into operation in Canada, Australia, India. Compared with the best SCP power units of the previous vintage, with their LHV net efficiencies of around 41-42%, modern SCP coal-fired units look to achieve efficiencies of 46-47%.

Modern SCP power units have relatively low main steam pressure values (around 24-28 MPa) even with main and reheat steam temperatures of up to 600-610 ºC. The use of double reheat schemes is generally considered impractical because of high capital costs and increased complexity.

The current phase of rising main and reheat steam temperatures is expected to end in the early 2010s at the level of about 620-630 ºC.

Newly developed materials should make it possible to launch the first commercial plants with steam conditions of 28 MPa, 630/650 ºC in the middle of the next decade and then pilot USC units with steam parameters of 35 MPa, 700/720 ºC, promising plant efficiencies of 50-51%.

References

1. P Luby, “Supercritical systems”, MPS , August 2003, pp 27-32.

2. A Leyzerovich, Steam turbines for modern fossil-fuel power plants, The Fairmont Press, 2007.

3. A Leyzerovich, “Steam turbines: how big can they get?”, MPS, May 2007, pp 50-55.

4. A Leyzerovich, “New benchmarks for steam turbine efficiency”, Power Engineering, August 2002, pp 37-42.