Search

• Seepage failure probability of high core rockfill dams during RDD is investigated. • The randomness of the parameters is estimated by Bayesian inference. • Seepage failure is most likely to occur in the mid-term of RDD. Rapid drawdown (RDD) is an abnormal reservoir flood control operation that may cause seepage failure of high core rockfill dams. In this paper, a novel framework combining Bayesian inference and transient unsaturated seepage simulation is proposed to evaluate the seepage failure probability of the core during RDD. The dynamic hydraulic boundaries utilized in the simulation are first determined by the performance of water release structures. The extreme learning machine (ELM) is then employed as the surrogate model to establish the mapping relationship between material parameters with simulated seepage pressure and hydraulic gradient. The Bayesian inference method is applied to characterize the uncertainty of material parameters based on seepage monitoring data. Finally, the posterior distribution samples of random parameters are utilized to calculate the seepage failure probability of the core. The seepage safety of the core under different drawdown rates is studied with a real 186 m high core rockfill dam. The results reveal that a faster RDD rate can lead to an increased probability of seepage failure. The maximum failure probability is 35.56 % for the four release structures working together (about 0.316 m/h), higher than 23.16 % for the three release structures (about 0.165 m/h). It is also found that the peak of seepage failure probability often occurs in the middle stage of RDD. Therefore, to ensure the safety of the project, it is important to control the RDD rate and carefully monitor the seepage pressure. From the point of view of seepage failure, the proposed method for evaluating the safety of core rockfill dams under RDD is of high guiding significance to the risk assessment of similar projects under extreme conditions. [ABSTRACT FROM AUTHOR]

• An method of multi-source data-driven unsaturated seepage parameter inversion. • NSGA-II is used to generate the pareto-optimal solutions of parameters. • Parameter decision based on a comprehensive index composed of multi-source data. • Inversion results are more in line with engineering practice and have higher accuracy. The inverse parameter's estimation of unsaturated seepage models is important to ensure the safety of earth dams, while most existing studies determine the hydraulic characteristics only based on seepage pressure monitoring data, which may lead to inversion results deviating from the intrinsic properties of dam materials. This study proposed an advanced methodology for unsaturated seepage parameter inversion of earth dams driven by multi-source data. The transient unsaturated seepage behavior of earth dams is simulated using the Van Genuchten (VG) model, and the extreme learning machine (ELM) is adopted as the surrogate model to estimate the approximate simulation results. Then the multi-objective function for parameter inversion is constructed by combining the observed transient pressure data and ELM models, and the Pareto-optimal solutions are generated using the elitist non-dominated sorting genetic algorithm (NSGA-II). Multi-source information, including parameter design values, in-situ testing results, and monitoring data, are fused to construct a comprehensive index to drive the parameter decision. The proposed method is applied to the PB core rockfill dam, which has a well-established seepage pressure monitoring system. In addition, the design values and in-situ testing results of the hydraulic conductivity of the dam material are also available. The results showed that the inversion parameters of the proposed method are closer to the physical characteristics of materials and have advantages in prediction accuracy. [ABSTRACT FROM AUTHOR]