The Chinese market presents a variety of challenges to turbine manufacturers. Operating conditions (head and flow) vary significantly at site due to the Chinese flood season. This article deals with the particular challenges of four important Chinese projects - Geheyan, LiJiaXia, Ertan and Jiang Ya.
GE Hydro has been involved in the Chinese market since the early 1980s. The first contract was obtained in 1988, to supply the design of a large ring gate to Dong Fang Electrical Machinery Works (DFEM) for the ManWan project in southern China.
The four most recent contracts, described in this article, deal with the turbine portions, and in particular medium head Francis turbines.
The turbines for all four projects have rated heads ranging from 80 to 165 meters. The model development work for the first project (Geheyan) took place in 1989, while the model acceptance tests for the most recent project (JiangYa) took place in June 1997.
The evolution in design technology over this 8 year period, in terms of cavitation behaviour, stability of operation at off-design conditions, as well as overall performance, can be clearly seen in the results of the four projects.
The Projects
Supply of the Geheyan units was shared evenly between a Canadian consortium and a Chinese manufacturer, Harbin Electrical Machinery Works (HEMW). GE Hydro designed the hydraulic designs.
For the LiJiaXia project, DFEM subcontracted the hydraulic design of the turbines to GE Hydro. All the manufacturing of these turbines was performed by DFEM in China. The first unit became operational in early 1997, and is now the most powerful unit in China.
The Ertan units were split between GE Hydro, DFEM and HEMW. The first units will be put into operation this year, and will succeed those of LiJiaXia as the most powerful in China.
Jiang Ya follows a similar approach to LiJiaXia in that the hydraulic design will be supplied by GE Hydro, while the prototypes will be supplied entirely by DFEM.
Site Locations
The two most important rivers in China are the Huang He (or Yellow River, so named because of the high silt concentration ), and the famous Yangtze river (Fig.1). Most of the projects we have been involved with are concentrated around these two rivers and their tributaries. The Yellow river gets its name from the high silt content in its lower regions, giving the water a distinctive yellow-brown colour. LiJiaXia is situated on the upper Yellow river, where the water is still relatively clean. The Yangtze also has some silt in the water. It rises in the flood season, but never reaches the high levels found in the Yellow river.
The Challenges
Hydroelectric projects in China pose significant challenges to the hydraulic design of turbines. They are often subjected to wide variations in head and flow due to severe flood seasons. This means that the turbines will rarely be running at their peak efficiency points. High efficiency over a broad range of operating conditions becomes more important than just a high peak efficiency. Running at off design conditions can also lead to stability problems and increased potential for cavitation damage. What do I mean by these terms?
Instability in a hydraulic turbine usually results from the strong vortex created in the center of the flow at the outlet of the runner. For part load conditions, the vortex spirals outward and precesses in the direction of the turbine's rotation at about one third of the turbine's rotational speed. This "rope", as it is called, can cause large pressure fluctuations, low frequency vibrations and undesirable variations in turbine output called "power swings." The full load rope remains almost centered in the flow at the outlet of the turbine, and rotates in a direction opposite to that of the turbine runner. This vortex is usually much more stable, but occasionally can cause problems if the full load efficiency is too low, or if it triggers a resonant frequency in the machine.
Cavitation in a Francis turbine runner comes in many different forms. With modern design technology it is not difficult to avoid cavitation at the design point (best efficiency point at design head); however, when the turbine is operated far from its design point, different cavitation phenomena can occur.
Inflow cavitation is the most aggressive type, often causing significant erosion damage to the runner blades, and results from a mismatch in flow angles and blade angles at the runner inlet. It is often present when the machine is being operated at heads which are either well above, or well below, the design head. As mentioned earlier, the operational head ranges of turbines in China is typically much larger than those of comparable size units in North America. The runner designs for the Chinese market must be adapted to these demanding conditions.
Silt is also a concern for many project engineers in China. Especially when composed of quartz (as is the case in many Chinese rivers) it can cause severe damage to turbine components. Often, one bad flood season is enough to necessitate the shutdown of a machine for extensive repairs.
Transportation limitations are another problem plaguing large projects in many countries, and China is no exception. Turbine runners for both the LiJiaXia and Ertan projects had to be split in half in order to fit through the railway tunnels (with little room to spare) on the way to the site. The runner halves would then be reassembled and bolted together at site.
High Weighted (Overall) Efficiency
GE Hydro began an aggressive research and development program in the 1980s, encompassing everything from design software to extensive model test developments, in an effort to ensure that its technology would outpace the ever increasing demands of these large projects. The evolution of the technology can be clearly seen in their results. The design parameters which affect the flatness of performance curves for both wide head and flow variations have been identified and can now be easily controlled.
Stability at Part Load
The most common solution to hydraulic instability is to admit air through the center of the turbine shaft, in order to dampen the fluctuations from the central rope. Unfortunately, this is not always an effective solution.
The best solution remains in designing specialized turbine runners which will remain stable over a wide range of operating conditions. The measured model pressure fluctuations caused by the JiangYa runner (designed just this year) are less than half the amplitude of those of the Geheyan runner (designed in 1989). The physical phenomena and corresponding design parameters responsible for part load pressure fluctuations have been identified and can now be controlled by specialized blade designs. Provisions for central air admission are still a standard practice for all of our large prototype turbines. But if the air is not required to stabilize the machine at site, it can easily be shut off to ensure the maximum possible efficiency for all operating points.
Cavitation Erosion
For new machines, the ratio of maximum to minimum head gives good indication of the challenge a designer will face to avoid any problems with inflow cavitation. As a general rule, ratios above 1.4 require specialized runner designs. In the case of JiangYa, the maximum head is 109.5m, while the minimum head is 62.1m. This gives a respectable ratio of 1.76.
The JiangYa project benefitted from about eight years of development work to extend the safe operating limits through improved blade designs. It is very difficult to move the inflow cavitation limit at heads higher than the best efficiency head compared to the late 1980s technology, but significant improvement has been made for low head operation.
Soil Erosion
Although the water for the four projects is relatively clean (by Chinese standards), a significant amount of research into both hydraulic and mechanical design of turbines for silt laden water applications is being done by GE Hydro.
The erosion resistance of many different alloys has been studied in the laboratory to determine the best case for critical turbine components in high silt environments although a good (cavitation free) hydraulic design still takes precedence over exotic materials.
GE Hydro also recently developed in cooperation with the Centre de recherche en calcul (CERCA) a software package which can compare the potential for silt-erosion damage of different hydraulic profiles. The package integrates with our 3-d viscous flow analysis software and can evaluate all hydraulic components of the turbine. The most important components to optimize are the stay vines, wicket gates and runner. This tool allows the designer to minimize the potential for erosion damage, and to thicken up the profiles in critical areas where silt erosion damage is anticipated at the design stage.
Transportation Limitations
The most significant transportation limitation for large projects in China is often the size of the railway tunnels (particularly when there is no other way to get the components to a remote site). This limitation was a major factor in the sizing of the turbines for both the LiJiaXia and Ertan projects. Half of the 6.3 m diameter runner for the LiJiaXia project has little room to spare on its way to the site.
Fortunately, circumstances allowed us to cut the blades in very low stress areas. The outflow edge of edge of a third blade, although crossing the diametrical plane, would fit inside the tunnel, and could be left intact and later welded to the adjoining half crown at site.
Conclusion
The Chinese market presents a variety of challenges to turbine manufacturers. Turbine performance for Francis runners has been improved to cover wide variations in flow and head at high efficiency levels. Cavitation-free operating ranges have been expanded significantly to cover these very large variations in head and flow. Part load stability has been greatly improved without air admission by designing specialized blade profiles.
All these improvements are the result of continuous research and development efforts. Ultimately, however, the success of these four projects is owing to the close cooperation between GE Hydro and the Chinese manufacturers DFEM and HEMW in an effort to provide end users with a high quality product at an affordable cost.
Stuart Coulson is a Hydraulic Design Engineer with GE Hydro. This paper was originally presented at CEA's Electricity '97. ET