The proving of Bugok

20 July 2001

Korea’s first independent power project (IPP) has proved a triumph for the Sigma CAD tool and a validation of the 3A turbine. Staff report

The first IPP power plant in Korea, constructed by LG Engineering & Construction, has now come fully on line. After the start of commercial operation, on 1 April, it was inaugurated on 4 June, in the presence of the Korean commerce minister Chang Che-shik, after a year of testing and commissioning.

Bugok 1 is a 537 MWe (net) seawater cooled multi-shaft combined cycle plant consisting of 2x179 MWe (net) gas turbines of the Siemens V84.3A type, and a 193 MWe (net) Siemens condensing steam turbine, type KN, provided with steam from two HRSGs. About 14 MW is consumed on site. The low pressure steam turbine was supplied by Fuji Electric, under a Siemens licence, and the high/medium pressure, K type, turbine by Siemens-Parsons of Newcastle. Primary fuel is LNG supplied by KOGAS via pipeline; emergency fuel is low sulphur distillate supplied by LG Caltex.

Plant design (see MPS, November 1998) was by LG E&C, working with KOPEC, based on Siemens standard design, but the power island (PI) was entirely by Siemens. LG Energy owns the plant, whose entire output is contracted to KEPCO for 20 years. Siemens was the main equipment supplier, and supervised erection, most of which was carried out by local companies. Siemens also had the power block commissioning contract and has the maintenance contract. The balance of plant and 25 per cent of the PI were supplied locally, primarily auxiliary boiler, water treatment systems, air intakes (AAF Korea), bypass stack (Hyundai), HRSG, condensers and piping systems (Doosan).

The entire plant was installed, commissioned and handed over within the contracted schedule and performance requirements.

Bugok’s 550 MW brings to 4845 MW Siemens’ installed capacity in Korea, of a national total of 47980 MW.

The capacity of the Bugok site is large – provision has been made to add a second power island, to create a compact 1000 MW block. The site has room for three more such blocks.

But continuation to unit 2 is not certain. The general economic situation is still recovering from the 1997 slump. In 1998 GDP actually went down 5.8 per cent while energy demand dropped by approximately the same figure. But the growth in supply in 2001 is expected to be 4.3 per cent, and to continue at that rate, to meet demand forecasts suggesting an increase of 2 GW per year for the foreseeable future. Projections imply that four plants per year on the scale of Bugok would have to be built during the next 10 years.

Korea is tackling the situation by selling off its state generation company KEPCO, currently in control of 90 per cent of energy production, and inviting investment from foreign power generating companies and developers. Altready in the field are El Paso, Mirant, Texaco and Tractebel, among others. And the government plans increasingly to exploit the coal firing option. At present about 43 per cent of production is coal fired – a figure expected to rise to nearly 47 per cent by 2015.

Testing ground

Inevitably the Bugok site has become something of a proving ground for Siemens, in particular the 3A series turbine but also its Sigma CAD system, being used for the first time on this scale as a complete design and specification tool. Used in this way it incorporates wider system considerations, in order to avoid physical or logistical conflict with civil and steel structures, cabling, catwalks etc.

It was the first time every aspect of design of the power island site had been covered, down to 15 mm piping, in a single system, and it worked well. Usually there is a huge number of design errors that have to be corrected, sometimes costing up to millions of dollars; in this case only minor changes were required, at a cost of less than $50 000.

Design changes

The project was LG E&C’s first large power station design, and KOPEC’s first experience of working with German standards. For this and other reasons, of the two and a half year period from commencement (November 1997) to start of simple cycle commercial operation (June 2000), detailed design took up nineteen months, overlapping only with procurement and manufacturing. By contrast, construction time was only six months.

This resulted in some design changes to the standard kit.For example, it is usual to site the steam condensers directly below the steam turbine within a hot well, which increases plant height and may mean constructing a basement floor. In this case a lower cost option has been chosen. The condensers are sited side by side with the turbine, most of the increase in footprint size being offset by savings in the space consumed by cooling water lines, and better access.

The contract required introduction of steam blowing and chemical cleaning. Siemens took on the project, introducing a type of chemical cleaning (from Beiztechnik) not previously employed in a Korean power station, and saved one month on the critical path during combined cycle commissioning.

Performance testing

Operational, performance and safety tests were required by KEPRI, KEPCO and by LG Energy. The performance tests confirmed the high standard of power output (over 550 MW (ASME)) and efficiency, over 58 per cent (LHV). For combined cycle operation, 32 tests were required by a combination of the requirements of KEPCO, LG Energy and the Korean government. For simple cycle operation the testing sequence was shorter.

The charts above come from that series of tests. They show reliability testing, especially important during a fast start up. Again, the chart shows the significance of load rejection tests. It is necessary that turbine speed does not rise to the level that trips the overspeed control, because local regulations demand that the overspeed trip is the last to go.

3A issues

During the build period for Bugok, development of the 3A series gas turbine was continuing at the company’s test rig in Berlin for 50 Hz and 60 Hz models and particularly on the combustion side where problems had been encountered. The facility employs scaled down (3 burners from a ring of 24) burner rings and one full scale burner – of the V84.3A type, the 60 Hz version of the 3A.

The original aim was to develop a high rating GT by scaling up above 50 MW (into the 100 MW plus zone attractive to utilities) from a proven straightforward and robust design with a moderate pressure ratio and only one combustion chamber at a competitive price and with low life cycle costs. But early designs proved problematic in terms of blade corrosion and resonant burner oscillations combined with flame instability.

Before 1997 gradual scaling up involved greater mass flow, blade cooling, and the use of heat resistant materials. Later work on compressor blading resulted in a new 15 stage variant developed with Pratt and Whitney. Other work on improved aerodynamics, film cooling and the blade material (monocrystalline alloy PW1483) as well as the thermal barrier coating in cooperation with PWC resulted in further V84.3A efficiency improvements also incorporated in the V94.3A.

In 1997 hot gas path inspection revealed coatings on first blade rows being eroded, especially at hot spots; improvements to the film cooling and coating application procedure significantly extended the service life of coatings. Further work on hollow rotors, allowing second and subsequent stage cooling from extraction points at lower temperatures and pressures, thus saving compression work improving efficiency. The creep strength of the central hollow shaft was increased by the use of an improved alloy, now used in all 3As.

The resonance problem was solved on the test rig in Berlin using an annular hybrid burner, type 84.3A. at one eighth size (ie 3 burners). At lean burn, no characteristic frequency was detectable. But on enriching the fuel mix, humming suddenly occurred at a single frequency and high amplitude. Siemens called this 'fluid dynamic feedback'.

The most successful prevention method was achieved by modifying the outlet, adding an extension, a cylindrical burner outlet (CBO), to the normal conical shape. Flame instability was reduced to a minimum by experimentation. Results showed that the number of modified CBO burners is important, and on the full size HBR rig, baseload was reached at high and low ambient temperature with NOx and CO well below authorised levels. But output was limited by oscillation. Experiment showed that fitting 20 burners with cylindrical burner oulets gave the optimum increase in power, 7 per cent above nominal load, and this has become the standard design.

Bugok, which represents Siemens’ 17th and 18th 3A units, features the design arrived at in 2000, regarded as the definitive improved version in the series.

Linkedin Linkedin   
Privacy Policy
We have updated our privacy policy. In the latest update it explains what cookies are and how we use them on our site. To learn more about cookies and their benefits, please view our privacy policy. Please be aware that parts of this site will not function correctly if you disable cookies. By continuing to use this site, you consent to our use of cookies in accordance with our privacy policy unless you have disabled them.