A reservoir landslide surge wave probability calculation method based on energy equivalence

Abstract

The application discloses a reservoir landslide surge probability calculation method based on energy equivalence, comprising the following steps: (1) a complex-shaped landslide body is generalized into a landslide body with the same volume stretched by a main section, and the surge characteristics at the water inlet of the landslide body are set to be consistent; (2) a landslide surge main section calculation model is established to calculate the sliding speed of the landslide body and the initial surge height; (3) a surge propagation calculation model is established; (4) the initial surge is input into the surge propagation calculation model to calculate the regional surge propagation; (5) the accuracy of the surge propagation calculation model is verified; (6) a calculation sample of different sliding speeds and instability volumes is established; (7) an LSTM agent model is established; and (8) Monte Carlo simulation is carried out to determine the probability distribution of the landslide surge. The application solves the problems of low calculation precision of the formula method and difficulty in effectively adjusting the instability volume of the landslide body by using the numerical method, and is convenient for quickly and accurately performing the probability analysis of the landslide surge.

Description

Technical Field

[0001] This invention relates to a method for simulating landslide processes in the field of landslide surge research, and more particularly to a method for calculating the probability of reservoir landslide surges based on energy equivalence. Background Technology

[0002] Due to the uncertainties in instability volume and sliding velocity, the landslide surge in reservoir areas is often difficult to determine accurately. Probabilistic analysis of landslide surges is of great significance for accurately assessing landslide surge hazards.

[0003] Formulaic methods, as an effective approach to assessing landslide surge hazards, can quickly perform probabilistic analysis of landslide surges by considering the distribution characteristics of computational parameters in the formula. However, due to significant simplifications, the results of formulaic methods are difficult to accurately characterize surge features under complex terrain conditions, and can only be used for preliminary landslide surge probability analysis. Numerical simulation methods are an effective means of studying landslide surges, and combining numerical simulation with machine learning is an effective method for conducting probabilistic analysis of landslide surges. In this case, the quantity and quality of landslide surge samples are crucial to the accuracy of the machine learning model calculations.

[0004] While computational models based on the three-dimensional Navier-Stokes equations can accurately simulate the evolution of landslide surge motion, the impact range of landslide surges is vast, often extending tens of kilometers. In such cases, the computational efficiency of these models significantly decreases, becoming slow and unable to quickly obtain results under various conditions, hindering the rapid construction of large-scale computational samples, and making probabilistic analysis difficult. While the depth integral two-layer model offers some improvement in computational efficiency, the instability volume of the landslide remains undetermined, making it difficult to quickly adjust the landslide volume, and thus hindering successful probabilistic analysis of landslide surges.

[0005] In response to the current problems, how to quickly and accurately analyze the probability of reservoir landslide surges is an urgent technical issue that needs to be addressed. Summary of the Invention

[0006] Purpose of the invention: This invention proposes a method for calculating the probability of landslide surge in reservoirs based on energy equivalence. This method solves the problems of low calculation accuracy in formula methods and difficulty in effectively adjusting the instability volume of landslide bodies in numerical methods, providing a useful approach for rapid and accurate probability analysis of landslide surge.

[0007] Technical solution: This invention is based on a method for calculating the probability of reservoir landslide surge waves using energy equivalence, comprising the following steps:

[0008] Step (1) generalizes the complex-shaped landslide into a landslide with the same volume formed by stretching the main profile, and sets the surge characteristics at the water entry point of the landslide to be consistent.

[0009] Step (2): Based on the main engineering geological profile of the landslide, establish a main landslide surge profile calculation model including the sliding surface, landslide body, and water body, and calculate the sliding velocity of the landslide body and the initial surge height; the process is as follows:

[0010] Step (2.1) considers the lithological distribution characteristics of the landslide body, uses the Mohr-Coulomb constitutive model to characterize the mechanical characteristics of the landslide body, considers the weakening effect of the mechanical parameters of the slip zone, and establishes the water body according to the water level conditions;

[0011] Step (2.2): Set up velocity monitoring points inside the landslide body and set up initial surge monitoring points in front of the landslide body;

[0012] Step (2.3) sets the calculation parameters, which include the cohesion, internal friction angle and density of the landslide body and slip zone, as well as the density of the water body;

[0013] Step (2.4) calculates the landslide surge generation process, extracts the monitoring results of velocity monitoring points and initial surge monitoring points, and obtains the change process of initial surge and landslide velocity.

[0014] Step (3): Based on the topographic features of the target area, a shallow water equation model is used to establish a surge propagation calculation model. An initial surge input boundary is established at the intersection of the landslide body and the water surface in the surge propagation calculation model to input the initial surge.

[0015] Step (4): Input the initial surge wave into the surge wave propagation calculation model; input the initial surge wave waveform before the landslide stops moving into the initial surge wave input boundary, so that the total energy transferred from the landslide to the water body is consistent with the generalized landslide body; set the length of the initial surge wave input boundary to be the same as the width of the generalized landslide body, so that the energy release rate transferred from the landslide to the water body is consistent with the generalized landslide body, perform regional surge wave propagation calculation, and calculate the maximum surge wave amplitude of the river channel;

[0016] Step (5) compare the calculated maximum surge amplitude of the river channel with the maximum surge amplitude of the river channel in the physical model experiment to verify the accuracy of the calculation results of the surge propagation calculation model.

[0017] Step (6): Use the verified surge propagation calculation model to perform a seven-level two-factor orthogonal numerical simulation. Based on the changes in landslide surge characteristics caused by the change in landslide sliding velocity and the changes in landslide surge characteristics caused by the instability volume of the landslide, establish calculation samples with different sliding velocities and instability volumes.

[0018] Step (7) involves training an LSTM neural network using computational samples to establish an LSTM surrogate model; the process is as follows:

[0019] The computational samples are shuffled, and more than 70% of the shuffled computational samples are used as the training set, while the remaining computational samples are used as the test set. The LSTM is trained multiple times. The root mean square error (RMSE) and mean absolute percentage error (MAPE) of the multiple training sessions are used as evaluation metrics to perform hyperparameter tuning, obtain the optimal LSTM model parameters, and then establish the LSTM surrogate model.

[0020] Step (8) uses an LSTM surrogate model to perform Monte Carlo simulation to determine the probability distribution of landslide surge; the process is as follows:

[0021] Step (8.1): Based on the geological characteristics of the landslide, determine the distribution characteristics of the landslide velocity and the volume of instability of the landslide.

[0022] Step (8.2) involves sampling the distribution characteristics of the determined slip velocity and instability volume, and inputting the sampled slip velocity and instability volume into the LSTM surrogate model to obtain the surge characteristics; performing multiple calculations with different slip velocities and instability volumes to obtain the cumulative probability distribution of the landslide surge characteristics.

[0023] In step (2.1), a calculation model of the main profile of landslide surge is established based on the main engineering geological profile of the landslide body.

[0024] In step (2.4), the landslide surge generation process includes the process of the landslide body sliding down, impacting the water body, and generating surge.

[0025] In step (3), the center of the initial surge input boundary is located at the center of the intersection of the landslide body and the water surface.

[0026] In step (4), the generalized landslide width is calculated by dividing the three-dimensional landslide volume by the area of ​​the landslide in the main profile.

[0027] In step (8.2), at least 100,000 calculations with different sliding speeds and instability volumes are performed to obtain the cumulative probability distribution of landslide surge characteristics.

[0028] In step (4), the initial wave waveform before the landslide stops moving is input at the initial wave input boundary. After the landslide stops moving, the initial wave input boundary transforms into a land boundary, allowing the wave to move under its own energy.

[0029] In step (4), when the initial surge is input into the surge propagation calculation model, the total energy transferred from the landslide body to the water body and the energy release rate are consistent with the generalized landslide body.

[0030] The method to ensure that the energy release rate is consistent with the generalized landslide body is as follows: the length of the initial surge input boundary is set to be the same as the width of the generalized landslide body. After the initial surge is input to the boundary, the entire boundary produces the same surge change characteristics.

[0031] In step (6), the method for determining the changes in landslide surge characteristics caused by changes in landslide velocity is as follows:

[0032] Step (6.1): Adjust the friction coefficient of the sliding surface in the main profile calculation model, calculate the initial surge characteristics under different landslide sliding velocities, and extract the corresponding landslide sliding velocity and initial surge.

[0033] Step (6.2): ​​Keep the initial surge input boundary unchanged, input the initial surge characteristics of different sliding speeds into the initial surge input boundary, and calculate the landslide surge characteristics under different sliding speed conditions.

[0034] In step (6), the method for determining the changes in landslide surge characteristics caused by the instability volume of the landslide body is as follows:

[0035] Step (a) ensures that the main cross-sectional area of ​​the landslide body remains constant. Based on different landslide body instability volumes, calculate the initial surge input boundary length under different instability volumes, and establish a surge propagation calculation model for different initial surge input boundary lengths.

[0036] Step (b) involves inputting the initial waveform of the surge to initial surge input boundaries of different lengths and calculating the characteristics of landslide surge under different instability conditions.

[0037] Beneficial effects: Compared with the prior art, the present invention has the following advantages:

[0038] (1) The reservoir landslide surge probability calculation method based on energy equivalence proposed in this invention obtains landslide surge characteristics under different sliding velocities by changing the friction coefficient of the sliding surface; and quickly obtains landslide surge characteristics under different instability quantities by changing the length of the initial surge input boundary, thus providing conditions for quickly and accurately constructing a large number of landslide surge samples.

[0039] (2) This invention uses an energy-equivalent reservoir landslide surge probability calculation method combined with an LSTM neural network to train samples, establish a landslide surge LSTM surrogate model, and use the LSTM surrogate model to perform Monte Carlo sampling simulation, thereby quickly and accurately obtaining the probability distribution characteristics of landslide surges. This invention provides a useful approach for probabilistic assessment of landslide surges. Attached Figure Description

[0040] Figure 1 This is a flowchart of the reservoir landslide surge probability calculation method based on energy equivalence according to the present invention;

[0041] Figure 2 This is a calculation model diagram of the main profile of landslide surge established in an embodiment of the present invention;

[0042] Figure 3 This is a diagram of the landslide motion process obtained by the landslide surge main profile calculation model in an embodiment of the present invention;

[0043] Figure 4 The figures show the landslide velocity variation curve and the initial surge water level process line obtained by calculating the main profile model of the landslide surge in this embodiment of the invention.

[0044] Figure 5 This is a diagram of the surge propagation calculation model established in an embodiment of the present invention;

[0045] Figure 6 This is a comparison chart of the calculation model results of surge propagation and the experimental results of the physical model in an embodiment of the present invention;

[0046] Figure 7 The graph shows the calculation results for different sliding speeds and instability quantities in the embodiments of the present invention;

[0047] Figure 8 This is a training result diagram showing different sliding speeds and instability quantities in an embodiment of the present invention;

[0048] in, Figure 8 (a) is a diagram of the training set results in an embodiment of the present invention;

[0049] Figure 8 (b) is a test set result diagram in an embodiment of the present invention;

[0050] Figure 9 This is a diagram showing the distribution characteristics of landslide sliding velocity and instability volume in an embodiment of the present invention;

[0051] Figure 10 This is a cumulative probability distribution curve of the surge climbing up the dam surface obtained by Monte Carlo simulation in an embodiment of the present invention. Detailed Implementation

[0052] like Figure 1 As shown, the method for calculating the probability of reservoir landslide surge based on energy equivalence in this invention includes the following steps:

[0053] Step (1) generalizes the complex-shaped landslide into a landslide with the same volume formed by stretching the main profile, and sets the surge characteristics at the water entry point of the landslide to be consistent.

[0054] Step (2) involves establishing a main profile calculation model for the landslide surge, and calculating the sliding velocity of the landslide body and the initial surge height. This process includes steps (2.1)-(2.4):

[0055] Step (2.1) considers the lithological distribution characteristics of the landslide body, uses the Mohr-Coulomb constitutive model to characterize the mechanical characteristics of the landslide body, considers the weakening effect of the mechanical parameters of the slip zone, and establishes the water body according to the water level conditions.

[0056] Step (2.2): Set up velocity monitoring points inside the landslide body and set up initial surge monitoring points in front of the landslide body.

[0057] Step (2.3) sets the calculation parameters, which include the cohesion, internal friction angle and density of the landslide body and slip zone, as well as the density of the water body.

[0058] Step (2.4) calculates the landslide surge generation process, extracts the monitoring results of velocity monitoring points and initial surge measurement points, and obtains the change process of initial surge and landslide velocity.

[0059] Step (3): Based on the topographic features of the target area, a shallow water equation model is used to establish a surge propagation calculation model. An initial surge input boundary is established at the intersection of the landslide body and the water surface in the surge propagation calculation model to input the initial surge.

[0060] The center of the surge input boundary is located at the intersection of the landslide body and the water surface.

[0061] Step (4) involves inputting the initial surge into the surge propagation calculation model to calculate the regional surge propagation. When inputting the initial surge into the surge propagation calculation model, it is ensured that the total energy transferred from the landslide body to the water body and the energy release rate are consistent with the generalized landslide body.

[0062] One method to ensure that the total energy transferred from the landslide to the water body is consistent with that of the generalized landslide is to input the initial surge characteristics of the landslide before it stops moving only at the initial surge input boundary. After the landslide stops moving, the initial surge input boundary becomes a land boundary, allowing the surge to move under its own energy.

[0063] The method to ensure that the energy release rate is consistent with the generalized landslide body is as follows: the length of the initial surge input boundary is set to be the same as the width of the generalized landslide body. After the initial surge is input to the boundary, the entire boundary produces the same surge change characteristics.

[0064] Step (5) Verify the accuracy of the calculation results of the surge propagation calculation model by combining the physical model experiment. The verification method is to compare the maximum surge amplitude of the river channel obtained by calculation with the maximum surge amplitude of the river channel in the physical model experiment.

[0065] Step (6) uses the verified surge propagation calculation model to perform a seven-level two-factor orthogonal numerical simulation. Based on the changes in landslide surge characteristics caused by the change in landslide velocity and the changes in landslide surge characteristics caused by the instability volume of the landslide, calculation samples with different sliding velocities and instability volumes are established.

[0066] The method for determining the changes in landslide surge characteristics caused by changes in landslide velocity is as follows:

[0067] Step (6.1): Adjust the friction coefficient of the sliding surface in the main profile calculation model, calculate the initial surge characteristics under different landslide sliding velocities, and extract the corresponding landslide sliding velocity and initial surge characteristics.

[0068] Step (6.2): ​​Keep the initial surge input boundary unchanged, input the initial surge characteristics of different sliding speeds into the initial surge input boundary, and calculate the landslide surge characteristics under different sliding speed conditions.

[0069] The method for determining the changes in landslide surge characteristics caused by the instability volume of the landslide body is as follows:

[0070] Step (a) ensures that the main cross-sectional area of ​​the landslide body remains constant. Based on different landslide body instability volumes, calculate the length of the initial surge input boundary under different instability volumes, and establish a surge propagation calculation model for different initial surge input boundary lengths.

[0071] Step (b) involves inputting the initial waveform of the surge into the initial surge input boundary of different lengths and calculating the landslide surge characteristics under different instability volume conditions.

[0072] Step (7) uses computational samples to train the LSTM neural network and establishes an LSTM surrogate model.

[0073] The method for establishing the LSTM surrogate model is as follows: Randomly shuffle the computational samples, using 70% of the shuffled samples as the training set and 30% as the test set to train the LSTM. Repeat this step 5 times, using the average RMSE and MAPE from the 5 iterations as evaluation metrics to fine-tune the hyperparameters and obtain the optimal LSTM model parameters, thus establishing the LSTM surrogate model.

[0074] Step (8) uses an LSTM surrogate model to perform Monte Carlo simulation to determine the probability distribution of landslide surge.

[0075] The method for Monte Carlo simulation using an LSTM surrogate model is as follows:

[0076] Step (8.1): Based on the geological characteristics of the landslide body and the statistics of landslide disasters in the region, determine the distribution characteristics of the landslide body sliding velocity and instability volume.

[0077] Step (8.2) involves sampling the distribution characteristics of the determined slip velocity and instability volume, and inputting the sampled slip velocity and instability volume into the LSTM surrogate model to obtain the surge characteristics. At least 100,000 calculations with different slip velocities and instability volumes are performed to obtain the cumulative probability distribution of the landslide surge characteristics.

[0078] Example

[0079] Taking a landslide surge in a river as an example, this invention provides an example of the reservoir landslide surge probability calculation method based on energy equivalence.

[0080] (1) A main profile calculation model was established using the main engineering geological profile. Velocity monitoring points and initial surge measurement points were set up in the main profile calculation model, such as... Figure 2 .

[0081] (2) Calculate the landslide surge evolution process of the main profile calculation model, such as Figure 3 Extract the initial surge characteristics and landslide velocity of the main profile calculation model, such as... Figure 4 .

[0082] (3) Based on the topographic features of the target area, a surge propagation calculation model was established. The landslide volume is 2661 × 10⁴ m³. 3 The area of ​​the landslide body in the main profile model is 23962 m². 2 The initial surge input boundary length was determined to be 1110.5m, and it was set at the intersection of the landslide and the water surface. The boundary center was located at the center of the intersection of the landslide and the water surface. Figure 5 .

[0083] (4) Input the initial surge characteristics obtained from the main profile calculation model into the initial surge input boundary. Only input the initial surge characteristics before the landslide stopped moving into the initial surge input boundary, so that the entire boundary produces the same surge change characteristics. After the landslide stops moving, the initial surge input boundary becomes a land boundary, and then the surge propagation process is calculated. Extract the maximum wave amplitude of the river channel in the surge propagation calculation model and compare it with the physical model experiment to verify the effectiveness of the surge propagation calculation model, such as... Figure 6 .

[0084] (5) Considering the uncertainties in sliding velocity and instability volume, a seven-level, two-factor orthogonal numerical simulation was conducted. The calculation samples are shown in Table 1. The sliding velocity of the landslide body ranges from 13.40 to 25.59 m / s, and the instability volume ranges from 5.32 × 10⁻⁶ m / s. 6 m 3 ~37.25×10 6 m 3A total of 49 different sliding velocities and landslide instability volumes were calculated. The sliding velocity was changed by adjusting the friction coefficient of the sliding surface in the main profile calculation model, and the instability volume was changed by altering the initial surge input boundary length of the surge propagation calculation model. The maximum surge run-up at the dam was extracted as the calculation result for different working conditions.

[0085] (6) Calculation results of surge run-up for different samples are as follows: Figure 7 As shown, with the increase of landslide velocity and landslide volume, the rise of the surge wave on the dam face shows a significant increase.

[0086] (7) An LSTM neural network model was used to train the samples to establish a landslide surge LSTM surrogate model. 70% of the data was used as the training set, and 30% as the test set. The calculation results are as follows: Figure 8 As shown, the RMSE of the training set is 0.365, and the RMSE of the test set is 0.398. The MAPE of the training set is 1.33%, and the MAPE of the test set is 1.60%. This demonstrates that the established LSTM model has good generalization and accuracy in altitude prediction.

[0087] (8) Distribution characteristics of slip velocity and instability volume, as shown in Figure 9 As shown, the instability volume of the landslide follows a log-normal distribution with a mean of 26.61 × 10⁻⁶. 6 m 3 The variance is 0.07. It is assumed that the sliding velocity of the landslide body follows a normal distribution with a mean of 17.82 m / s and a variance of 2.0. 100,000 Monte Carlo calculations are performed using an LSTM surrogate model to calculate the probability distribution of the surge dam's run-up. Figure 10 As shown, when the surge needs to climb 12m to cross the dam, the probability of the surge crossing the dam is 0.37. When the surge needs to climb 13.2m to cross the dam, the probability of the surge crossing the dam drops sharply to 0.02.

[0088] Table 1

[0089]

Claims

1. A method for calculating the probability of reservoir landslide surge based on energy equivalence, characterized in that: By changing the friction coefficient of the sliding surface, the characteristics of landslide surge at different sliding velocities were obtained; By changing the length of the initial surge input boundary, landslide surge characteristics under different instability quantities can be quickly obtained; The method includes the following steps: Step (1) generalizes the complex-shaped landslide into a landslide with the same volume formed by stretching the main profile, and sets the surge characteristics at the water entry point of the landslide to be consistent. Step (2) establishes a main profile calculation model of the landslide surge, including the sliding surface, landslide body, and water body, and calculates the sliding velocity of the landslide body and the initial surge height; the process is as follows: Step (2.1) uses the Mohr-Coulomb constitutive model to characterize the mechanical features of the landslide and establishes the water body based on the water level conditions; Step (2.2): Set up velocity monitoring points inside the landslide body and set up initial surge monitoring points in front of the landslide body; Step (2.3) sets the calculation parameters, which include the cohesion, internal friction angle and density of the landslide body and slip zone, and the density of the water body; Step (2.4) calculates the landslide surge generation process, extracts the monitoring results of velocity monitoring points and initial surge monitoring points, and obtains the change process of initial surge and landslide velocity. Step (3): A shallow water equation model is used to establish a wave propagation calculation model. An initial wave input boundary is established at the intersection of the landslide body and the water surface in the wave propagation calculation model to input the initial wave. Step (4): Input the initial surge wave into the surge wave propagation calculation model; input the initial surge wave waveform before the landslide stops moving into the initial surge wave input boundary, so that the total energy transferred from the landslide to the water body is consistent with the generalized landslide body; set the length of the initial surge wave input boundary to be the same as the width of the generalized landslide body, so that the energy release rate transferred from the landslide to the water body is consistent with the generalized landslide body; perform regional surge wave propagation calculation, and calculate the maximum surge wave amplitude of the river channel; Step (5) compare the calculated maximum surge amplitude of the river channel with the maximum surge amplitude of the river channel in the physical model experiment to verify the accuracy of the calculation results of the surge propagation calculation model. Step (6): A seven-level, two-factor orthogonal numerical simulation is performed using the validated surge propagation calculation model; calculation samples with different sliding velocities and instability volumes are established based on the changes in landslide surge characteristics caused by changes in landslide sliding velocity and the changes in landslide surge characteristics caused by changes in landslide instability volume; the method for determining the changes in landslide surge characteristics caused by changes in landslide sliding velocity is as follows: Step (6.1): Adjust the friction coefficient of the sliding surface in the main profile calculation model, calculate the initial surge characteristics under different landslide sliding velocities, and extract the corresponding landslide sliding velocity and initial surge. Step (6.2): ​​Keep the initial surge input boundary unchanged, input the initial surge characteristics of different sliding velocities into the initial surge input boundary, and calculate the landslide surge characteristics under different sliding velocity conditions; The method for determining the changes in landslide surge characteristics caused by the instability volume of the landslide body is as follows: Step (a): The main cross-sectional area of ​​the landslide body is constant. Based on different landslide body instability volumes, the length of the initial surge input boundary under different instability volumes is calculated, and a surge propagation calculation model for different initial surge input boundary lengths is established. Step (b) involves inputting the initial waveform of the surge to initial surge input boundaries of different lengths and calculating the characteristics of landslide surge under different instability volume conditions. Step (7) involves training an LSTM neural network using computational samples to establish an LSTM surrogate model; the process is as follows: The computational samples are shuffled, and more than 70% of the shuffled computational samples are used as the training set, while the remaining computational samples are used as the test set. The LSTM is trained multiple times. The average root mean square error and average absolute percentage error of the multiple training sessions are used as evaluation metrics to perform hyperparameter tuning, obtain the optimal LSTM model parameters, and then establish the LSTM surrogate model. Step (8): Monte Carlo simulation is performed using an LSTM surrogate model to determine the probability distribution of landslide surge waves; the process is as follows: Step (8.1): Based on the geological characteristics of the landslide, determine the distribution characteristics of the landslide velocity and the volume of instability of the landslide. Step (8.2) involves sampling the distribution characteristics of the determined slip velocity and instability volume, and inputting the sampled slip velocity and instability volume into the LSTM surrogate model to obtain the surge characteristics; performing multiple calculations with different slip velocities and instability volumes to obtain the cumulative probability distribution of the landslide surge characteristics.

2. The method for calculating the probability of reservoir landslide surge based on energy equivalence as described in claim 1, characterized in that: In step (2.1), a calculation model of the main profile of landslide surge is established based on the main engineering geological profile of the landslide body.

3. The method for calculating the probability of reservoir landslide surge based on energy equivalence as described in claim 1, characterized in that: In step (2.1), the mechanical characteristics of the landslide body are characterized by the Mohr-Coulomb constitutive model based on the lithological distribution characteristics of the landslide body; and a water body is established based on the water level conditions.

4. The method for calculating the probability of reservoir landslide surge based on energy equivalence as described in claim 1, characterized in that: In step (2.4), the landslide surge generation process includes the process of the landslide body sliding down, impacting the water body, and generating surge.

5. The method for calculating the probability of reservoir landslide surge based on energy equivalence as described in claim 1, characterized in that: In step (3), a wave propagation calculation model is established using the shallow water equation model based on the topographic features of the target area.

6. The method for calculating the probability of reservoir landslide surge based on energy equivalence as described in claim 1, characterized in that: In step (3), the center of the initial surge input boundary is located at the center of the intersection of the landslide body and the water surface.

7. The method for calculating the probability of reservoir landslide surge based on energy equivalence as described in claim 1, characterized in that: In step (4), the generalized landslide width is calculated by dividing the three-dimensional landslide volume by the area of ​​the landslide in the main profile.

8. The method for calculating the probability of reservoir landslide surge based on energy equivalence as described in claim 1, characterized in that: In step (8.2), at least 100,000 calculations with different sliding speeds and instability volumes are performed to obtain the cumulative probability distribution of landslide surge characteristics.

Images