Search

For better stability, safety and water resource management in a dam, it is important to evaluate the amount of seepage from the dam body. This research is focused on machine learning approach to predict the amount of seepage from Pakistan’s Earth and rock fill Tarbela Dam during 2003 to 2015. The data of temperature, rainfall, water inflow, sediment inflow, reservoir level collected during 2003 to 2015 served as input while the seepage from dam during this period was the output. Artificial Neural Network (ANN), Random Forest (RF), Support Vector Machine (SVM), and CatBoost (CB), have been used to model the input-output relationship. The algorithms used to predict the dam seepage reported a high R2 scores between actual and predicted values of average seepage, suggesting their reliability in predicting the seepage in the Tarbela Dam. Moreover, the CatBoost algorithm outperformed, by achieving an R2 score of 0.978 in training, 0.805 in validation, and 0.773 in testing phase. Similarly, RMSE was 0.025 in training, 0.076 in validation, and 0.111 in testing phase. Furthermore, to understand the sensitivity of each parameter on the output (average seepage), Shapley Additive Explanations (SHAP), a model explanation algorithm, was used to understand the affect of each parameter on the output. A comparison of SHAP used for all the machine learning models is also presented. According to SHAP summary plots, reservoir level was reported as the most significant parameter, affecting the average seepage in Tarbela Dam. Moreover, a direct relationship was observed between reservoir level and average seepage. It was concluded that the machine learning models are reliable in predicting and understanding the dam seepage in the Tarbela Dam. These Machine Learning models address the limitations of humans in data collecting and analysis which is highly prone to errors, hence arriving at misleading information that can lead to dam failure.

Alternating direction implicit difference scheme (ADI-FDTD) divides a traditional time step into two time steps, with forward difference and backward difference, and integrates both the advantages of unconditional stability of implicit difference scheme and relatively simple calculation of explicit difference scheme, it breaks through the constraint of Courand–Friedrichs–Levy (CFL), and is characterized by unconditional stability. And the boundary condition of uniaxial anisotropic perfectly matched layer (UPML) is anisotropic medium PML applied in absorption edge of FDTD area, with the absorption of wide band, simple iterative formula, and easy programming, without the electric and magnetic field splitting. By discretizing the Maxwell equations in a two dimensional structure, numerical formulas of ADI-FDTD with UPML boundary condition for GPR wave are presented in detail. GPR simulations are carried out for two models. Based on the analysis of numerical results, insights of the radar wave spreading and changing in space are obtained, which can provide a better interpretation of real radar data. The results show that the ADI-FDTD algorithm based on the UPML boundary condition can deploy larger time step and eliminate the strong reflection on the truncated boundaries, lead to an efficient GPR modeling.

This paper describes a series of plugging agents which have been developed these years by the authors. The features of these agents can bestated as: (1) Stopping losses without stopping driling; (2) High ratio of success; (3) Improving fluids properties; (4) Suiting to many types of losses, even large cracks and caves; (5) Perfect efficiency in different conditions of pollution such as calcium, salt pollution and high temperature; (6) Sufficient resources of raw materials, no-poison, no-harm, no pollution to the environment; (7) Convenience in use; (8) Lower cost. The properties, mechanisms and applications of these agents are also described in this paper.

In order to improve the resolution and accuracy of the Ground-Penetrating Radar (GPR) forward modeling, this paper will focus on applying the mesh less algorithm-Element Free Method(EFM)which is widely used in the electromagnetism field into the radar numerical simulation. The EFM is dispensed with the concepts of elements, just needs a mass of node information and adopts the Moving Least Squares Method (MLS) to structure shape function. So the EFM has many advantages, such as simple preprocessing, high accuracy and high successive independent variable solutions. This paper, which is based on the radar wave equation, deduces the detailed solution of the element free method solving for GPR wave equation, and discusses the improved combination way of the improved sarma absorbing boundary and EFM using in the GPR forward modeling. And then, the paper compiles the matlab program of corresponding EFM GPR forward modeling, involving this program in a typical geo-electric model forward modeling. The result shows that EFM based on improved sarma absorbing boundary can gain complete and travel-time-precious GPR wave section with satisfaction in both precision and stability.