Search

Data assimilation for unconfined seepage analysis has faced significant challenges due to hybrid causes, such as sparse measurements, heterogeneity of porous media, and computationally expensive forward models. To address these bottlenecks, this paper introduces a physics-informed neural network (PINN) model to resolve the data assimilation problem for seepage analysis of unsaturated earth–rockfill dams. This strategy offers a solution that decreases the reliance on numerical models and enables an accurate and efficient prediction of seepage parameters for complex models in the case of sparse observational data. For the first attempt in this study, the observed values are obtained by random sampling of numerical solutions, which are then contributed to the synchronous constraints in the loss function by informing both the seepage control equations and boundary conditions. To minimize the effects of sharp gradient shifts in seepage parameters within the research domain, a residual adaptive refinement (RAR) constraint is introduced to strategically allocate training points around positions with significant residuals in partial differential equations (PDEs), which could facilitate enhancing the prediction accuracy. The model's effectiveness and precision are evaluated by analyzing the proposed strategy against the numerical solutions. The results indicate that even with limited sparse data, the PINN model has great potential to predict seepage data and identify complex structures and anomalies inside the dam. By incorporating coupling constraints, the validity of our PINN model could lead to theoretically viable applications of hydrogeophysical inversion or multi-parameter seepage inversion. The results show that the proposed framework can predict the seepage parameters for the entire research domain with only a small amount of observation data. Furthermore, with a small amount of observation data, PINNs are able to obtain more accurate results than purely data-driven DNNs. [ABSTRACT FROM AUTHOR]

Climate change poses significant threats to the water reservoir, with potential repercussions for Pakistan's economy. This study investigates the impact of climatic variations on seepage in Tarbela dam, Pakistan, by analyzing hydro-climatological factors. The research employs statistical multivariant time series (SARIMA and ARDL) and machine learning (ANN and Catboost) modeling techniques to assess the influence of climatic variability on seepage dynamics. The analysis focuses on annual and seasonal climatic variables from a data set spanning from 2003 to 2015, aiming to explore the relationship between climatic variability and dam seepage. The SARIMA model is utilized to estimate water inflow, temperature, rainfall, and reservoir level, validating the integration of multiple data sets. Machine learning models demonstrate superior performance, achieving high R2 scores during the training, validation, and testing. The findings derived from the ARDL analysis reveal a significant increase in seepage with rising water inflow, rainfall, and reservoir level. Among the machine learning models, CatBoost performs the best. The study recommends implementing appropriate adaptive approaches to mitigate any negative impact on Tarbela dam seepage. Moreover, this research serves as a guide for enhancing the dam's power generation capacity and improving irrigation water management in the face of existing crises in Pakistan. [ABSTRACT FROM AUTHOR]

Estimating the quantity of seepage through the foundation and body of a dam using proper health and safety monitoring is critical to the effective management of disaster risk in a reservoir downstream of the dam. In this study, a deep learning model was constructed to predict the extent of seepage through Pakistan's Tarbela dam, the world's second largest clay and rock dam. The dataset included hydro-climatological, geophysical, and engineering characteristics for peak-to-peak water inflows into the dam from 2014 to 2020. In addition, the data are time series, recurring neural networks (RNN), and long short-term memory (LSTM) as time series algorithms. The RNN–LSTM model has an average mean square error of 0.12, and a model performance of 0.9451, with minimal losses and high accuracy, resulting in the best-predicted dam seepage result. Damage was projected using a deep learning system that addressed the limitations of the model, the difficulties of calculating human activity schedules, and the need for a different set of input data to make good predictions. [ABSTRACT FROM AUTHOR]