HYDRAULIC RESEARCH ON DEEP-LEVEL OUTLETS OF THE THREE GROGES PROJECT

 

 

Sun Shuangke, Chen Jie, Liu Zhiping and Guo Jun

China Institute of Water Resources and Hydropower Research, A1, Fuxing

Road, Beijing, China, Phone: 86-10-68515511-1917, Fax: 86-10-68538685

 

 

Abstract: The deep-level outlets of the Three Gorges Project (TGP) have two important hydraulic problems as following: The first is the problem of cavitation mitigation. Because the flow velocity in outlet exceeds 35m/s, aeration facility is used to protect the surface, the protecting range and its effect of aeration facility need to be evaluated correctly, and how to control the irregularity in construction work is of vital importance as well. The second is the problem of sediment abrasion. Because the operation mode of ¡°discharging the turbid flow and impounding the clear water¡± is to be adopted in TGP, the problem of the sediment abrasion on deep-level outlet is non-negligible. In this paper, based on the hydraulic model test, the protecting range and the effect of aeration facility are discussed, and the distribution of cavitation number along the flow passage also presented. Furthermore, the quantitative forecast of the sediment abrasion on deep-level outlets carried out by reviewing of the performance experience of some similar hydropower stations. The results show that the aeration facility is effective but the sediment abrasion is so serious that the designer, the builder, and the supervisor of TGP should pay grave attention to it.

 

Keywords: deep-level outlet, aeration facility, cavitation number, sediment abrasion

1     INTRODUCTION

The notable characteristics of the TGP layout is that there are many flood discharge outlets including 22 surface spillways, 23 deep-level outlets and 22 bottom outlets are provided in three layers within the flood discharge dam block of 483m in long. As the most important outlet structures, the deep-level outlets shoulder the main responsibilities of flood discharge for the Three Gorges Project. Owing to its important function in TGP, many schemes are brought forward and studied by the designer and the researcher. As a result, the layout scheme of closed pipe with aerator of vertical drop type is applied as shown in Fig.1. The drop depth is 1.5m, and the diameter of two air vents is 1.4m set in side walls behind the drop section.

Îı¾¿ò:  Fig.1  The scheme of the deep-level outlet

For the deep-level outlet in TGP, the following hydraulic problems such as the air cavity behavior, the protecting range and its effect of the aeration facility, the cavitation number distribution, and the sediment abrasion on the outlet face, should be studied comprehensively. Especially, the sediment abrasion is of serious concern. In order to maintain enough effective storage for the reservoir of TGP, the operation mode of ¡°discharging the turbid flow and impounding the clear water¡± is to be adopted in TGP. Because the sediment concentration of outlet flow will increase gradually with the operating years, so the sediment abrasion on deep-level outlet will be getting more serious. During of the initial operation of TGP, the quantitative analysis is pressing need to be conducted.

2    HYDRAULIC CHARACTERISTICS OF THE DEEP-LEVEL OUTLET

In order to reduce the scale effect of the model test, a model with large scale 1:26 is built. The experimental study is progressed comprehensively including the flow pattern, the air cavity behavior, the distribution of pressure and air entrainment along the flow passage, and so on [1]. The result shows that the aeration facility can ensure an enough size of air cavity which is 23m~54m long. The effect of air entrainment in the hydraulic model is remarkable. Both in upper layer and bottom layer, there have an aeration zone of which the depth increases in the downstream direction, and two aeration zones merge together at the 20+30m on initial water level of 135.0m and at 20+70 on check flood level of 180.4m respectively. Between the two aeration zones, there is a clear water zone without aeration. In the downstream behind clear water zone, the section is fully aerated in vertical direction. 

3    THE PROTECTING RANGE AND ITS EFFECT OF AERATION

Table 1  The distribution of the aeration concentration

No

Elevation

Chainage

The upstream water level (m)

 

(m)

(20+  m)

135.0

145.0

166.9

175.0

180.4

1

85.12

43.17

19.5

/

/

/

/

2

84.14

47.11

21.0

38.0

/

/

/

3

83.15

51.05

7.8

30.2

/

/

/

4

82.23

54.75

5.1

12.8

11.5

/

/

5

80.99

59.69

3.1

6.0

16.4

11.5

/

6

80.13

63.16

2.0

3.4

15.0

14.8

19.7

7

79.14

67.09

1.6

2.5

14.9

17.7

18.5

8

77.83

72.34

1.2

2.1

11.9

11.4

11.3

9

76.29

78.77

0.9

1.5

4.5

4.8

4.4

10

75.50

86.34

0.9

0.8

2.1

2.0

2.2

the symbol/means the measuring point is in the air cavity

From the hydraulic model test [1], it is clear that the main protecting area is the range from 20+42m to 20+81m. Table 1 gives the air entrainment distribution along the flow surface, this result shows that the air concentration reaches 3%~5% at the end of air cavity and gradually decreases downstream. The minimum value is about 1%. In the prototype the actual air concentration should be higher due to the influence of scale effect. It shows that the aeration facility is able to practically prevent the flow passage from the cavitation erosion.

4    THE DISTRIBUTION OF CAVITATION NUMBER

Besides using the aeration facility, controlling the irregularity in construction is the other effective means to prevent suffering from cavitation erosion. In order to control the irregularity effectively, it is necessary to learn about the distribution of the cavitation number that is defined as following,

                                (1)

here, h0  is the atmospheric pressure£¬h0=10.332mH2O£»hv is the saturated vaporous pressure£¬ hv=0.238mH2O ( the water temperature is 200C)£»h¡¢V express the pressure and average velocity of the corresponding point respectively; g is the acceleration of gravity.

According to Eq.1, we can use the experimental results to calculate the cavitation number along the bottom surface and side walls respectively. On the bottom surface, the cavitation number is small such as in the front of drop, the end of air cavity, and the end of outlet, therefore the protecting emphases of bottom surface is mainly in the region from 20+40m~20+70m. By the way, there is a little of reversing water with a slow velocity in the air cavity section, the cavitation erosion is impossible to occur. Similar to the bottom surface, the low value of cavitation number in side wall is in the range of 20+40m~20+70m too, and its value of 0.13~0.3 is smaller than in the bottom surface. It is clear that the value of s changes with the reservoir water level obviously: under the level of 135.0m or 145.0m, it is mostly larger than 0.3; over the water level of 166.9m, it is mostly in the range of 0.13~0.2.

In summary, the side walls are more liable than the bottom surface to suffer from cavitation, so it is necessary to control the irregularity strictly. According to the relationship curve between the irregularity and the incipient cavitation number[2], we can give the standard of irregularity to be controlled for the deep-level outlet as following: for the bottom surface, the width-to-depth ratio on the inclining convex should be bigger than 13, and about trigonal convex it should be larger than 25; for the side wall, the corresponding value are 50 and 70 respectively. It seems to be very difficult to govern the irregularity of the outlet widely. Fortunately, as the flow pattern mentioned above, it is evident that only the clear water zone in side walls is easy to be damaged by cavitation, in other areas, in virtue of the aeration protection, the governing criteria can be moderated in certain extent. In addition, the rounding is necessary to be handled, because it is one of the effective techniques to decrease the incipient cavitation number of convex.

5    QUANTITATIVE FORECAST OF THE SEDIMENT ABRASION

The sediment problem, which has attracted great attention since the feasibility study, is one of the most important problems in Three Gorges Project. Up to now, the sediment research for TGP is mainly to resolve the sediment deposition in reservoir and approach channel, less attention is paid for the problem of sediment abrasion on the outlet structure, especially on the outlets. But in fact, the sediment abrasion on outlet structures is an unsolved and important problem from the view of operating management. It is well known, many hydropower stations in China have encountered fearful sediment abrasion [3], for example, the Liujiaxia Hydropower Station in the Yellow River, the Gongzui Hydropower Station in Daduhe River, and so on. Because of the complexity of the mechanism and the variety of the factors, the quantitative forecast on the sediment abrasion is quite difficult for the time of being, and the findings are seldom reported. On the basis of the operation experiences of Liujiaxia spillway, we try to forecast quantitatively the sediment abrasion on deep-level outlets of TGP.

Quantitative forecast method

The sediment in flow has two existing form of suspension load and bed load, in which the mechanism and the characteristics of abrasion are different each other. The former is always uniform in the whole section, and the latter is usually heavy in the bottom surface because of the gravity action. At present, the quantitative study is impossible because lack of performance data on bed load abrasion. In addition, the bed load in TGP is only in the order of 2% of the suspension load, so the abrasion is mainly dependent on the amount of suspension load. Therefore, in our study, only the suspension bed is considered to quantitatively forecast the sediment abrasion.

The abrasion intensity depends upon the hydraulic and sediment factors, the former includes the flow velocity, the flow duration, aeration characteristics, and flow pattern; the latter includes the sediment concentration, grain size and gradation, shape, mineral components and its hardness, as well as the abrasive resistance of the material. In general, the abrasion depth is in direct proportion to some parameter as following [3]:

                           (2)

In which, d is the abrasion depth (mm); S is the silt concentration (kg/m3); V is the flow velocity (m/s); D50 is the grain diameter (mm); T is the duration (hr.); R is the concrete strength; and K is the coefficient indicating the effect of grain properties. Referring to the previous study [3], the value of above coefficients is around m=0.7~1.0, n=2.7~3.2, p=-1.105, a=b=1.0, it indicates the flow velocity is the main factor. With the use of m=1.0, n=3.0, p=

¨C1.105, a=b=1.0, the Eq.2 can be derived as

                        (3)

In general, the abrasion resistance material is often used to prevent the surface of important area from the sediment abrasion, different material has different value of R. If the effect of R is counted into the coefficient of K, the simple form of Eq.3 can be expressed as following,

                           (4)

In which, the coefficient C indicates the combined effect of grain property and the abrasive resistant capability of the outlet. Using this Eq.4, we are able to quantitatively forecast the sediment abrasion on TGP¡¯ outlets.

Estimation of the annual mean operating duration

Îı¾¿ò:  
Fig.2?The regulation diagram in moderate flow year

Based on the regulation diagram in moderate flow year as shown in Fig.2, the unit  capacity of generating unit (700MW) and annual energy output (846.8 ¡Á108 kWh), it is easy to estimate the annual water amount through the power set as 4028¡Á108m3. Since the yearly runoff through TGP is about 4500¡Á108 m3, the yearly runoff through the outlet should be about 472¡Á108 m3 if the runoff through the surface spillway and the ship lock is ignored. Furthermore, the annual mean operating duration of one outlet can be estimated as 346hr. approximately.

Silt property and sediment concentration

The sediment in TGP is mainly from the Jinshajiang River, the annual sediment transport is 5.3¡Á108 t, and the mean sediment concentration is 1.20kg/m3. The field data shows that most of suspension load is fine in grain size, the median diameter is 0.036mm, the content of quartz and feldspar with the hardness over 6.0 is about 75%~90%, and the grain morphology is mainly in prism shape. The TGP¡¯ reservoir has about 600km long, being the type of fluvial channel. To reduce the silt accumulation, the operating mode of ¡°discharging the turbid flow and impounding the clear water¡± is to be used in TGP, it means that when the flood season comes let the water level down below 145.0m to draw the sediment. The hydraulic model tests[4] show that the sediment concentration and the median diameter in flood season will increase with the operating time, after 30 years these values are about 1.62kg/m3 and 0.0131mm respectively, after 50 years, these values will increase to 3.02kg/ m3 and 0.0227mm, after 89 years when the reservoir has reached the equilibrium period of sediment transport and deposition, the corresponding values will be 5.12kg/ m3 and 0.0034mm, and in the far future the bed load will begin to pass through the deep-level outlets.

Determination of coefficient

C In contrast with other factors, how to determine the value of coefficient C is more difficult. At present, the analogy analysis may be the feasible method to solve this problem. Fortunately, we have the performance experience of Liujiaxia Hydropower Station in the Yellow River that the silt property is similar to the TGP. Although it is in Yellow River basin, the suspension load in Liujiaxia outlet sluice is quite similar to the TGP. For example, the median diameter of the sediment coming from upstream is 0.025mm, mean sediment concetration is 3.31kg/m3, the content of quartz and feldspar with the hardness over 6.0 is about 96%, and the grain shape is mainly in prism form also. So we can assume that if the abrasive resistant material used in TGP outlet has the same abrasive resistance as in Liujiaxia spillway, the value of coefficient C may be similar to each other. Therefore, with the use of the C in Liujiaxia outlet, we can quantitatively forecast the sediment abrasion in TGP outlet. According to the operating data from 1975 to 1988 in Liujiaxia outlet and the inspection result about the yearly abrasion depth [5], we can use Eq.4 to calculate the value of coefficient C in Liujiaxia outlet. It is about (3~9)¡Á10¨C6 as shown in Table 2.

  Table 2  The calculation of C in Liujiaxia outlet sluice

year

T(h)

S(kg/m3)

D50(mm)

V(m/s)

¦Ä(mm)

C(¡Á10¨C6)

1975-1976

4599.5

0.88

0.027

26.6

10

4.86

1977-1978

668.9

9.95

0.024

27.4

10

3.03

1979

987.5

3.96

0.024

27.1

15

8.03

1984

1085.2

2.76

0.024

23.7

8

8.36

1985

945.4

3.60

0.069

26.9

20

4.38

1986

187.0

27.45

0.023

30.9

10

2.87

1988

174.5

24.82

0.021

22.8

10

9.28

Quantitative forecast

For the deep-level outlet in TGP, because its operating water level is often on 145.0m, hence the outlet flow velocity is about 27.5m/s. With the use of the following data: V=27.5m/s, T=346hr., C=6¡Á10¨C6, and the sediment concentration  recited above, the quantitative results of sediment abrasion on the deep-level outlet are given in Table 3, the abrasion depth is about 1.0mm/year, 3.2mm/year, and 8.0mm/year corresponding to the different operating years such as 30 years, 50 years, and 89 years respectively. In addition, this forecast is merely based on the assumption that the material used in TGP outlet is same as in Liujiaxia outlet. In fact, as better material is to be used, the value of C in TGP should be lower than in Liujiaxia, hence the actual results will be smaller than the forecast results. But if considered the effect of bed load, the abrasion will be more serious. Therefore, the above results can be seen as a reference information. It indicates that the sediment abrasion on the outlet increases with the operating years, which will attract more attention by the designers, constructors, and managers of TGP.

Table 3  Quantitative forecast to the sediment abrasion on TGP outlet

Operating ear

Silt oncentration

(kg/m3)

Flow elocity

(m/s)

the median diameter

(mm)

The abrasion pepth

(mm/year)

After 30 years

1.62

27.5

0.0131

1.0

After 50 years

3.02

27.5

0.0227

3.2

After 89 years

5.12

27.5

0.0340

8.0

6    CONCLUSION

To prevent the flow passage surface from cavitation erosion, aeration facility is to be used in deep-level outlets of TGP. By hydraulic model test, it indicates that the main protecting section is in the range from 20+42m to 20+81m, and the aeration facility can form an enough size of air cavity which is 23m~54m long to prevent the flow passage from cavitation erosion. The distribution of cavitation number shows that it is a hard-nut problem to govern the irregularity of the outlet. Fortunately, only the clear water zone in side walls is easy to be eroded by cavitation, in other areas, in virtue of the aeration protecting, the governing irregularity can be moderated in certain extent. The quantitative forecast of the sediment abrasion on deep-level outlets is carried out by the use of the operation experience of Liujiaxia hydropower station. As a result, the abrasion depth is about 1.0mm/year, 3.2mm/year, and 8.0mm/year corresponding to the different operating years such as 30 years, 50 years, and 89 years respectively. It is so serious that it should attract serious attention by the designers, constructors, and managers of TGP.

References

[1]    Sun Shuangke, Liu Zhiping & Zhou Sheng, The hydraulic model test on the deep-level outlet with aerator in TGP, IWHR report, No: HY-2000-03-021, 2000.6 (in Chinese).

[2]    Wu Chigong, Hydraulics, Publishing Company of Higher Education, 1983(in Chinese).

[3]    The damage on the outlet structure and its preclusion, ed. by Northeast China Investigation & Design Institute, publishing company of Chengdu University of Science and Technology, 1996 (in Chinese)

[4]    Study on the key problem of Three Gorges Project, IWHR & Changjiang Institute of Science report, No: 85-16-03-01, 1995.11 (in Chinese).

[5]    Liujiaxia Hydropower Plant, The cavitation erosion on Liujiaxia outlet and its repair work, The third symposium on the repair technology of hydro concrete, 1992.6 (in Chinese).