Study on the Influence of Water Level on Earth Dam Reinforced by Cut-Off Wall: A Case Study in Wujing Reservoir

Abstract

The construction of a cut-off wall is a common reinforcement method for earth rock dams. At present, compared with the in-depth study on homogeneous earth dams, more and more attention is being paid to the stability and deformation of earth dams strengthened by a concrete cut-off wall. In this study, aiming at the Wujing project of the earth dam strengthened by cut-off wall, the influence of the water level rise and fall on the stability of the dam slope, the deformation of the dam body, and the crack width on dam crest were analyzed by numerical calculation and in situ measurement. The analysis results show that when the reservoir encounters a sudden drawdown, the safety factor of the dam slope decreases sharply. The faster the sudden drawdown, the faster the safety factor decreases. When the reservoir water level rises, the dam’s horizontal displacement shifts to the upstream direction, and the change of horizontal displacement of the downstream slope is significantly larger than that at the measuring point of the upstream slope. The water level of the reservoir rises, and the surface of the dam body rises, and the fluctuation of settlement deformation shows that the upstream side is larger than the downstream side, especially during the period of abrupt change in the reservoir water level. The longitudinal cracks on the dam crest show a tendency of shrinkage when the reservoir water level rises, and opening decreases with the decrease of deformation gradient increment and increases with the increase of gradient increment.

1. Introduction

Most of the reservoir earth rock dam projects in China were built in the 1960s–1970s. Due to the limitations of technical and historical conditions at that time, there are many problems, such as poor project quality and many hidden diseases [1,2,3]. Therefore, dam seepage has become the most common hidden danger of earth rock dams [4,5,6]. At present, there are few studies on whether the construction of concrete cut-off walls in the reinforcement project of earth rock dams will affect the stability and the deformation of the dam.

The deformation of earth dams reinforced by concrete cut-off walls has certain particularity. First, most of the reinforced earth dams are old dams that have been in operation for decades. The consolidation of the dam soil has been basically completed, and the consolidation deformation and self-weight load deformation can be ignored [7,8]. However, the additional cut-off wall will change the seepage field and stress field in the dam, resulting in the reconstruction of the stress deformation of the dam [9]. Secondly, the factors that cause deformation are multi-year cyclic loads, such as reservoir water level load, flooding, and drainage processes [10,11,12]. The operation of the reservoir often experiences the working condition of water level fluctuation [13,14,15]. Water level fluctuation will cause the transient change of soil pore water pressure [16,17]. When the reservoir water level drops too fast, the soil drainage time is insufficient, and the pore water pressure in the dam body cannot dissipate quickly. Thus, the dam slope will form a downward trend under the action of seepage force, which causes slope instability, dam deformation, dam crest cracking, and other failure modes [18,19]. Therefore, it is of great significance for the design and application of seepage walls and the operation and management of reservoir dam projects in practical projects, especially for many small- and medium-sized reservoirs under construction or built, to systematically analyze the influence of seepage wall conditions and water level fluctuation on dam slope stability, dam deformation, and dam crest cracks.

At present, the research on slope stability analysis has been relatively mature [20,21,22,23]. The limit equilibrium method is widely used because of its clear mechanical concept and computational stability [24,25], which divides the sliding soil into several soil strips, and each soil strip and the whole meet the conditions of force and moment balance. That is, according to the shear strength criterion, the stability of the slope is solved by analyzing the balance under the limit state. The limit equilibrium method includes two aspects of calculation: calculating the safety factor of anti-sliding stability on a hypothetical sliding surface and finding a critical sliding surface with the lowest safety factor among the possible sliding surfaces [26]. According to the shear strength criterion, the stability of the slope is solved by analyzing the equilibrium under the limit state [27,28].

In this study, the Wujing project of the earth dam strengthened by the cut-off wall is taken as the research object, in which the stability of the dam slope, deformation of the dam body, and crack width on the dam crest are analyzed by means of limit equilibrium method and in-situ deformation measurement. It has been found that when the reservoir encounters a sudden drawdown, the safety factor will also decrease sharply. The horizontal displacement and settlements of the dam are affected by the change in the upstream reservoir water level. With the rise in the reservoir water level, the longitudinal cracks on the dam crest show a tendency of shrinkage, while the cracks show a tendency of opening when the reservoir water level drops.

2. Project Overview

Wujing reservoir is located in Wujing village, Zhaoxian Town, Wanli District, Nanchang City, Jiangxi Province, China, 3 km away from the urban area of Wanli, which is a small reservoir mainly for water supply and with comprehensive benefits of flood control. On the basis of the original homogeneous earth dam, high-pressure jet grouting was used to build a concrete cut-off wall in the dam in November 2011 to May 2012. Cracks were first found on the dam crest pavement in 2015, and continued to extend from 2015 to 2018. The cracks are about 132 m long and close to the dam axis (concrete cut-off wall). The average width of the cracks is 3 mm, the widest part is 40 mm, and the maximum dislocation is about 4 mm, as shown in Figure 1.

The dam is a homogeneous earth dam. After the construction of the concrete high-pressure jet grouting cut-off wall, the dam crest elevation is 192.55 m, the maximum dam height is 45.4 m, and the dam crest is 181 m long and 8 m wide. The normal pool level is 190 m, the design flood level is 190 m, the check flood level is 190.16 m, and the total storage capacity is 4.48 million m³. From 183 m elevation to dam crest elevation, precast block slope protection is adopted for the upstream slope of the dam, with a slope ratio of 1:2; From the elevation of 170 to 183 m, the upstream slope of the dam adopts dry masonry slope protection, with a slope ratio of 1:2.5. Below the elevation of 170 m, the upstream slope of the dam is a block stone prism with a slope ratio of 1:1.25. From the elevation of 170 m to the elevation of the dam crest, the downstream slope of the dam is protected by concrete grid turf. Below the elevation of 171 m, the downstream slope of the dam is sloped for drainage, as shown in Figure 2.

3. Study on the Influence of Sudden Drawdown on Dam Slope Stability

3.1. Seepage and Slope Stability Analysis Method

The earth rock dam and foundation are assumed to be isotropic porous media, the seepage in the dam body conforms to Darcy’s law, and the seepage control equation is:

$$\frac{\partial }{\partial x}(k_x\frac{\partial H}{\partial x})+\frac{\partial }{\partial y}(k_y\frac{\partial H}{\partial y})+Q=\frac{\partial \Theta }{\partial t}$$

(1)

where H is the total head, k_x and k_y are the horizontal and vertical permeability coefficients, cm/s or m/d, Q is the flow of water on the boundary, cm³/s or m³/d, and $\Theta$ is the volumetric moisture content.

Among them, the unsaturated percolation model was adopted from the Fredlund and Xing model, and its expression is:

$$\frac{\theta }{{\theta }_s}=\frac{1-\frac{\ln (1+u_s/u_r)}{\ln (1+{10}^6/u_r)}}{{\left\{\ln \left[e+{(u_s/a)}^b{u}_s/u_r\right]\right\}}^c}$$

(2)

where θ is the volume water content, u_se is the saturated volume water content, u_s is the suction, kPa, c is the soil property parameter of the residual water content function, u_r is the matrix suction corresponding to the water content, kPa, and a and b are the fitting parameters.

The limit equilibrium method was used to calculate the stability of dam slope. The solution idea is based on the Mohr Coulomb shear strength theory. The slope within the potential sliding surface is divided into several soil strip according to a certain proportion, and the static balance equation is established according to the limit equilibrium conditions between the soil strips. Considering the overall moment balance of sliding soil, the safety factor F_S of slope is calculated, and the stability of slope is evaluated according to the equation.

The moment equilibrium relationship is given by:

$$F_S=\frac{{\displaystyle \sum (c\Delta LR+RN\tan \varphi )}}{{\displaystyle \sum W{L}_W-{\displaystyle \sum N{L}_N}}}$$

(3)

Or expressed by horizontal force balance relation:

$$F_S=\frac{{\displaystyle \sum (c\Delta LR\cos \alpha +RN\tan \phi \cos \alpha )}}{{\displaystyle \sum N\sin \alpha }}$$

(4)

where:

$$N=\frac{W+\lambda f(x){\displaystyle \sum (\frac{c\Delta LR\cos \alpha }{F_S})-\frac{c\Delta L\sin \alpha }{F}}}{(\cos \alpha +\frac{\sin \alpha }{F_S})-\lambda f(x)(\frac{\cos \alpha \tan \phi }{F_S}-\sin \alpha )}$$

(5)

where $c$ is the cohesion, kPa, $\varphi$ is internal friction angle of soil mass, °. $\Delta L$ are the respective soil lengths on the sliding surface, m. $L_W$ is the moments of each soil strip against the center of the slip surface, Nm. $L_N$ are the moments corresponding to their normal for the midpoint of each soil strip at the sliding surface, Nm. $\alpha$ is the angle between the tangent lines of each soil and the horizontal plane, $R$ is the moment length to the center of the circle, m. $N$ is the normal phase force of the soil strip, kN. $W$ is the weight of soil strip, kN. λ is coefficient of variation of interaction force between soil strips, and the f(x) in Equation (5) is the function of interaction force between soil strips.

3.2. Numerical Model and Parameters

The axis of cut-off wall of Wujing reservoir dam is located on the dam axis with a thickness of 0.25 m and a wall top elevation of 190.5 m. Stone prism is used to fix the foot in the upstream. According to relevant design data, site investigation report and topographic mapping data of dam slope of Wujing Reservoir, a typical section of design is selected to establish a 2D dam model by GeoStudio software to analyze stability of dam slope with sudden drawdown, as shown in Figure 2. The boundary and initial conditions of the upstream face of the dam are the design water level and steady-state seepage field, and the dam foundation is the fully constrained boundary condition.

The Wujing reservoir dam is filled with silty sand and the foundation is granite rock mass, in which the weathered rock mass of the dam foundation mainly concentrates on the downstream side of the dam. According to the engineering geological report, the calculation material parameters of the dam and rock mass foundation are as follows:

When simulating the saturated–unsaturated seepage flow field, the saturated/unsaturated model is used for the dam body, and the saturated model is used for the dam foundation. The concrete cut-off wall is constructed by high-pressure jet grouting, and the measured anti-seepage coefficient of the cut-off wall is 5 × 10⁻⁷ cm/s.

3.3. Stability Analysis of Dam Slope under Sudden Drawdown

When the reservoir water level drops in a short time, it is easy to cause dam slope collapse, which is an important reason for the instability of earth rock dams. The process of sudden drawdown is analyzed and simulated under the condition of sudden drawdown of Wujing reservoir with the water level under normal storage conditions of 190 m, and the safety factor of dam slope stability before and after adding the cut-off wall is calculated. The simulation process of sudden drawdown is assumed as: water level drops from 190 m (The normal water level) to 160 m, lasting 10, 6, and 3 days, with descending speeds of 3, 5, and 10 m/day, respectively. The initial condition is the steady-state analysis result under long-term immersion. Figure 3 and Figure 4 have shown the typical most dangerous sliding surface calculated by GeoStudio with the parameter in

Table 1. Model calculation parameter.

Material	Permeability Coefficient/cm/s	Density/kg/m3	Cohesion/kPa	Friction Angle
Silty sand of dam body	5.0 × 10−5	1700	20	20
Dam foundation overburden	5.0 × 10−4	1800	15	28
Strongly weathered rock mass of dam foundation	1.0 × 10−4	2400	15	30
Weakly weathered rock mass of dam foundation	1.0 × 10−5	2600	20	33
Basement rock	1.0 × 10−6	2700	30	35

and its safety coefficient before and after sudden drawdown reaches the lowest level respectively with descending speed of 10 m/day by the mean of the limit equilibrium method, and the arrow indicates the effect of upstream reservoir water level.

The safety factor of upstream dam slope decreases with the sudden drawdown. When the water level drops to the lowest level, the safety factor reaches the lowest level, then shows a slow upward trend and finally remains stable. The reason is that the water head difference changes sharply due to the decrease in water level, which makes the stability of the slope decrease. At this time, the suction of the inner matrix of the rocks and soils on the dam slope does not rise in time. With the water level no longer dropping, the suction of the matrix inside the dam slope gradually recovers as the pore water pressure dissipates and the shear strength of the soil body gradually rises, thus the safety factor of stability increases. The stability safety factor duration curve of upstream and downstream dam slopes is shown in Figure 5.

According to the calculation parameters in the Wujing engineering geological report, the stability safety factor of the dam slope is about 2.0 without water level change, which is generally safe and stable. Only at the extreme sudden drawdown can the dam slope be unsafe. The faster the speed of sudden drawdown speed is, the faster the safety factor of upstream dam slope stability will decrease, which will be disadvantageous to the stability of the dam slope. With the end of the sudden drop, the pore water pressure of the dam body soil gradually dissipates, which makes the coefficient of stability safety rise again and finally tends to a stable value.

4. Measuring and Analysis of Dam Deformation under Water Level Rise and Fall

4.1. Arrangement of Measuring Points

Deformation measurements are carried out on the existing observation piers for surface displacement observation of the dam. Surface displacement and settlement of three transverse sections of the dam are observed, and each transverse section has 2 observation points, a total of 6 observation points, as shown in Figure 6.

4.2. Horizontal Displacement

The horizontal displacement of the dam is measured by observing piers. The results of horizontal displacement measurement for three measuring sections are given in Figure 7, Figure 8 and Figure 9, in which data on reservoir water level and rainfall are also given. The horizontal displacement obtained by the monitoring is listed in

Table 2. Horizontal displacement.

Monitoring Time	1# Section		2# Section		3# Section
	S11	S12	S21	S22	S31	S32
5 March 2020	0.00	/	0.00	0.00	0.00	0.00
23 March 2020	−2.75		3.07	1.17	−0.99	−0.67
7 April 2020	−2.66		0.38	32.05	−0.97	13.70
6 May 2020	0.84		1.31	−31.72	0.69	−12.90
21 May 2020	0.53		0.62	−1.33	2.58	0.55
15 June 2020	−5.46		−2.93	3.22	−1.31	1.03
28 June 2020	8.95		−3.04	−2.01	−1.23	−0.96
3 August 2020	−17.46		−9.96	−4.01	−11.80	−6.05
17 August 2020	3.13		−0.02	0.32	0.99	0.06
4 September 2020	0.41		−5.00	0.71	−2.37	1.26
12 October 2020	−3.88		5.52	−0.21	0.10	−0.28
4 December 2020	11.93		4.90	1.83	4.90	1.91
28 December 2020	1.57		−0.46	4.81	1.36	0.07
18 March 2021	4.04		7.98	−1.01	6.04	0.76
26 March 2021	−2.17		−2.50	0.98	1.94	2.25
2 April 2021	−1.54		0.54	−1.95	−3.96	−1.54

.

The horizontal displacement of the dam body at Section 1 # (S11–S21) is only the data on the upstream side, and the downstream measurement data are missing due to the failure of observation pier S21. The maximum displacement amplitude of water level at the upstream measuring point occurred from the end of June to the beginning of August 2020. From the deformation of measuring points and the variation law of the reservoir water level, as the water level rises from March 2020 to August 2020, except for the 8.95 value monitored on June 28, the horizontal displacement of the dam was almost smaller than 0, which generally tended to change to the upstream. As the water level drops from August 2020 to April 2021, the horizontal displacement of the dam was almost all greater than 0, and the horizontal displacement of the dam generally tended to change to the downstream.

From the deformation laws of the three sections, it can be seen that the deformation laws of the measuring points of the 2 # and 3 # sections are similar. As the water level rises from March 2020 to August 2020, if we do not consider the point with huge extreme deformation, it may be caused by measurement error, i.e., 32.05 and 13.70 mm measured on April 7 and 31.72 mm measured on May 6 in S22 and S32, the horizontal displacement of the dam was almost smaller than 0, which generally tended to change to the upstream. As the water level dropped from August 2020 to April 2021, the horizontal displacement of the dam was almost all greater than 0, and the horizontal displacement of the dam generally tends to change to the downstream. On the whole, the horizontal displacement of the dam body was affected by the change of the water level of the upstream reservoir, the water level of the upstream reservoir rises, and the dam body shifted to the upstream direction (the measured value is negative). On the contrary, the water level of the upstream reservoir decreased, and the dam body generally tended to shift to the downstream (the measured value was positive), and the horizontal displacement amplitude of the downstream dam slope measuring points of the two sections was significantly greater than that of the upstream slope measuring points.

4.3. Settlements

The results of settlements measurement for three measuring sections are given in the Figure 10, Figure 11 and Figure 12. The horizontal displacement obtained by monitoring are listed in

Table 3. Settlement.

Monitoring Time	1# Section		2# Section		3# Section
	S11	S12	S21	S22	S31	S32
5 March 2020	0.00	0.00	0.00	0.00	0.00	0.00
23 March 2020	−34.50	−38.30	−34.20	−25.19	−32.40	−28.60
7 April 2020	15.50	4.40	7.70	4.79	4.10	7.90
21 May 2020	−9.30	−1.70	−3.50	−2.90	1.70	−6.60
15 June 2020	−4.80	−9.80	−9.00	−16.00	−8.20	−5.30
28 June 2020	−5.90	6.80	−0.10	9.50	1.10	1.60
3 August 2020	−6.20	−11.20	−4.40	−4.10	−4.70	−1.90
17 August 2020	6.80	5.40	−1.50	0.80	2.60	1.80
4 September 2020	−2.30	8.70	4.30	6.70	−4.20	0.50
12 October 2020	3.10	−9.90	−2.50	−8.90	3.20	−3.00
4 December 2020	6.90	6.30	2.60	4.70	4.60	1.70
28 December 2020	4.00	−0.10	5.90	3.50	2.60	2.30
18 March 2021	−3.70	3.50	−3.10	−0.70	−4.30	−1.00
26 March 2021	−5.20	−1.60	−1.90	−2.50	1.70	−0.30
2 April 2021	8.60	−1.00	6.00	−3.10	2.20	1.90

.

From the settlement laws of the three sections, it can be seen that as the water level rises from March 2020 to August 2020, except for the settlement monitored on April 7, the horizontal displacement of the dam was almost smaller than 0, which generally tended to change to the upward. As the water level drops from August 2020 to April 2021, the horizontal displacement of the dam was almost all greater than 0, and the horizontal displacement of the dam generally tended to change to the downward. During the continuous change of reservoir water level, the two monitoring points in the same section basically show the same change trend. The change in the dam surface settlement was related to the fluctuation of the upstream reservoir water level. When the upstream reservoir water level rises, the dam surface settlement value was -, indicating that the surface had a trend of rising. On the contrary, when the water level of the upstream reservoir dropped, the surface settlement value of the dam body was +, showing a downward subsidence trend.

5. Crack Propagation Law Based on Deformation Gradient Method

5.1. Deformation Gradient Method

The deformation gradient method was to predict dam cracks based on dam settlement observation data. As shown in Figure 13, if there were two observation points a and b at the same elevation of the dam body, and the horizontal distance between the two points was ∆y, and if the cumulative settlement measured by T_i on a certain calculation date was Z_a and Z_b, respectively, then the deformation gradient of a and b on the date T_i was defined as $\gamma$, asshown in Equation (5).

$$\gamma \approx \tan \gamma =\frac{\Delta Z}{\Delta y}\times 100=\frac{Z_a-Z_b}{\left|y_a-y_b\right|}\times 100$$

(6)

When the monitoring point of the same cross-section was not at an elevation, the deformation gradient could be modified. As shown in Figure 14, assuming 2 observation points a and b at different elevations of the dam, the horizontal distance between the two points was ∆y, the initial difference in height between the two points was ∆Z₁, and the difference in height on a certain calculation date was ∆Z₂, then the modified deformation gradient of points a and b was defined as:

$$\Delta \gamma \approx \left|\tan {\gamma }_1-\tan {\gamma }_2\right|\times 100$$

(7)

$$\tan {\gamma }_1=\frac{\Delta Z_1}{\Delta y}$$

(8)

$$\tan {\gamma }_2=\frac{\Delta Z_2}{\Delta y}$$

(9)

$$\Delta Z_2=Z_a-Z_b+\Delta Z_1$$

(10)

It is worth noting that when the initial elevation difference ∆Z₁ = 0, i.e., a and b are at the same elevation, the modified deformational gradient agrees with the calculation result of Equation (5).

5.2. Analysis of Crack Width Propagation Law

Longitudinal cracks of dam crest are distributed along dam axis in plane in the range of dam 0+005~dam 0+148 stake, as shown in Figure 6. The crack width was measured with the vernier caliper. The measured crack width changes with time as shown in Figure 15.

Measuring data from 5 March 2020 to 2 April 2020 show that the overall change of crack width is relatively stable. During the rising process of reservoir water level, the development and change of crack width showed a certain contraction trend, and then the development and change of crack width showed a slowing trend of opening during the falling process of reservoir water level.

According to the monitoring results of Wujing reservoir, the longitudinal crack of the dam crest was analyzed by the deformation gradient method.

Table 4. Gradient calculation results.

Monitoring Time	Settlement Z/mm						Horizontal Distance of Observation Points on the Same Section/m	Deformation Gradient/%
	1# Section		2# Section		3# Section			γ1	γ2	γ3
	S31	S32	S21	S22	S11	S12
5 March 2020	0.0	0.0	0.0	0.0	0.0	0.0	9.00	0	0	0
23 March 2020	−32.4	−28.6	−34.2	−25.19	−34.5	−38.3		0.0422	0.1001	0.0422
7 April 2020	4.1	7.9	7.7	4.79	15.5	4.4		0.0422	0.0323	0.1233
21 May 2020	1.7	−6.6	−3.5	−2.9	−9.3	−1.7		0.0922	0.0067	0.0844
15 June 2020	−8.2	−5.3	−9	−16	−4.8	−9.8		0.0322	0.0778	0.0556
28 June 2020	1.1	1.6	−0.1	9.5	−5.9	6.8		0.0056	0.1067	0.1411
3 August 2020	−4.7	−1.9	−4.4	−4.1	−6.2	−11.2		0.0311	0.0033	0.0556
17 August 2020	2.6	1.8	−1.5	0.8	6.8	5.4		0.0089	0.0256	0.0156
4 September 2020	−4.2	0.5	4.3	6.7	−2.3	8.7		0.0522	0.0267	0.1222
12 October 2020	3.2	−3	−2.5	−8.9	3.1	−9.9		0.0689	0.0711	0.1444
4 December 2020	4.6	1.7	2.6	4.7	6.9	6.3		0.0322	0.0233	0.0067
28 December 2020	2.6	2.3	5.9	3.5	4.0	−0.1		0.0033	0.0267	0.0456
18 March 2021	−4.3	−1.0	−3.1	−0.7	−3.7	3.5		0.0367	0.0267	0.0800

Notes: The initial elevation difference is 4.3198 m for 1# section, 4.4942 m for 2# section, and 4.1936 m for 3# section.

is the result of calculating the deformation gradient according to the monitoring data from March 2020 to March 2021.

Figure 16 shows the variation curve of dam deformation gradient increment and crack width. The monitoring results showed that the variation rule of 3 cracks with deformation gradient is basically the same. Except for the monitoring in June, other monitoring calculation results showed that the change of increment of deformation gradient was basically consistent with that of crack width, i.e., the crack width decreased with the decrease in gradient and increased with the increase in gradient increment.

The relationship between the expansion of dam crest cracks and the uneven settlement of the dam body on both sides of the cracks is shown in Figure 16. The longitudinal cracks on the dam crest of Wujing Reservoir close with the rise of reservoir water and open with the fall of reservoir water, indicating that the soil mass on the upstream slope of the reservoir water fluctuation dam undergoes cyclic expansion and contraction deformation, thus affecting the opening and closing deformation of existing cracks.

6. Discussion and Conclusions

The deformation of the dam is closely related to the change in the reservoir water level, whether for horizontal deformation or settlement deformation. However, due to the different states of different dams, the impact of the dam on the change in the water level is different, as shown in

Table 5. Study on the effect of water level fluctuation on dam deformation.

Dam	Dam Type	Research Method	Whether the Water Level Fluctuation Affects the Dam Deformation	How Water Level Fluctuation Affects Dam Deformation
Chengbihe [7]	Earth dam	In situ measurements	Yes	Settlement on the dam’s upstream side
Yamula [29]	Earth dam	In situ measurements	Yes	Vertical deformation. The rising water level increases the subsidence velocity
An ultra-high arch dam [30]	Concrete arch dam	In situ measurements	Yes, but the influence of water level on dam deformation is hysteretic.	With the rise of the water level, the cluster center area also rises, indicating that the trend of the expansion and upward movement of the maximum deformation area of the dam body
Jiangya [31]	Gravity dam	In situ measurements	Yes	The reservoir impoundment is the predominant cause of the uplift of dam foundation.
An earth embankments of dams [32]	Earth dam.	Numerical simulation	Yes	The horizontal deformation of the dam toe is higher for a higher rising rate
Princeville [33]	Earth levee	Numerical simulation	Yes	The Factor of safety is affected by rate of rise/drawdown of the water level

.

In this study, by means of numerical analysis and in situ measuring, the influence of water level rise and fall on the Wujing earth dam reinforced by the cut-off wall has been analyzed, and the following conclusions can be drawn:

• Based on the established numerical calculation model of Wujing reservoir, and the geological report parameters, the safety factor of dam slope stability was obtained by the limit equilibrium method, which was about 2.0. When the reservoir encountered a sudden drawdown, the safety factor also decreased sharply. The faster the sudden drawdown was, the faster the safety factor decreased, and the more unfavorable it was to dam slope stability. With the end of the sudden drawdown, the pore water pressure of the dam body soil gradually dissipated, so that the safety factor of stability increased again and finally tended to be stable;
• The horizontal displacement of the dam was affected by the change in the upstream reservoir water level. It was found that the upstream reservoir water level rose and the horizontal displacement of dam shifted to the upstream direction. The upstream reservoir water level dropped, the horizontal displacement of the dam shifted to downstream, and the change of horizontal displacement of downstream slope was significantly larger than that at measuring point of upstream slope;
• The settlement deformation of the dam body was related to the fluctuation of the reservoir water level, in which the water level of upstream reservoir rose, and as surface of the dam body rose, conversely, it tended to sink. The fluctuation of the settlement deformation shows that the upstream side was larger than the downstream side, especially during the period of abrupt change in reservoir water level;
• With the rise in the reservoir water level, the longitudinal cracks on the dam crest showed a tendency of shrinkage, while the cracks showed a tendency of opening when the reservoir water level dropped. The change in the deformation gradient increment was basically consistent with the change in the crack opening, that is, the crack opening decreased with the decrease in the deformation gradient increment and increased with the increase in the gradient increment.