This paper introduces a novel stochastic inverse method that utilizes perturbation theory and advanced intelligence techniques to solve the multi-parameter identification problem of concrete dams using displacement field monitoring data. The proposed method considers the uncertainties associated with the dam displacement monitoring data, which are comprised of two distinct sources: the first is related to stochastic mechanical properties of the dam, and the second is due to observation errors. The displacements at different measuring points generated by dam mechanical properties exhibit spatial correlation, while the observation errors at different points can be considered statistically random. In this context, the inversion formulas are derived for unknown stochastic parameters of the dam by combining perturbation equations and Taylor expansion methods. An improved meta-heuristic optimization method is employed to identify the mean of stochastic parameters, while mathematical and statistical methods are used to determine the variance of stochastic parameters. The feasibility of the proposed method is verified through numerical examples of a typical dam section under different conditions. Additionally, the paper discusses and demonstrates the applicability of this method in a practical dam project. Results indicate that this method can effectively capture the uncertainty of dam's mechanical properties and separates them from observation errors.

@en

Dam safety monitoring has become an important topic and is critical for evaluating a dam's safety status. This study focuses on identifying the mechanical properties of a concrete dam from long-term viscoelastic deformation monitoring data. A novel inversion framework is proposed in which a surrogate model, instead of the finite element model, is placed inside the optimization loop. First, a multi-output surrogate model based on Gaussian process is trained by using data from a finite element simulation in the creep regime. In order to efficiently create a high-precision and reliable surrogate model, three test instances are conducted to investigate the impact of sample size, parameter range and output quantity on the performance of the surrogate model. Subsequently, a meta-heuristic optimization, multi-verse optimizer, is employed to identify the unknown viscoelastic parameters. The results illustrate that the identified properties allow predictions on dam displacement which are consistent with the monitoring data. Compared with the traditional inversion method based on finite element modelling, the proposed inversion method based on the multi-output surrogate model not only achieves accurate estimation of mechanical parameters but also greatly improves computational efficiency.@en

A roller compacted concrete dam (RCCD) located in Cambodia has been gradually deformed over the operation period (2011–2019), and the creep effect of the dam foundation is significant. In order to make integrity and safety assessments of the dam, it is necessary to know the actual mechanical properties of the foundation. This research proposes an intelligent computational framework for analysing the time-dependent working behaviour of the RCCD combined with viscoelastic finite element methods and advanced software techniques. According to the long-term deformation characteristics of the foundation, the Burgers model is employed to describe the constitutive relation of the bedrock. A finite element formulation describes the relationship between dam deformation and mechanical properties in the creep regime. A structural inverse methodology based on improved parallel grey wolf optimization (IGWO) is developed in order to search and identify viscoelastic parameters of the dam foundation. The nonlinear convergence factor strategy and multi-core parallel computing are introduced to enhance global search capability and improve the accuracy of the optimization algorithm. An example of analysis is performed on a dam section, and the results, which are compared with actual measurements for discussion, demonstrate that the selected constitutive model is reasonable and the designed inverse methodology is feasible. Moreover, the proposed IGWO algorithm is very competitive with other state-of-the-art optimization methods such as basic grey wolf optimization (GWO), particle swarm optimization (PSO) and whale optimization algorithm (WOA) for parameter inversion and real-time problems.

@en

High and steep slopes which have developed fractures and intercalations are a great threat to the operation of dams and reservoirs. In this work, the geological conditions and potential modes of failure of the slope found in the right bank of Suofengying hydropower station are investigated for the slope stability and the results are presented. In order to strengthen the slope, an innovative stabilization scheme is employed. The stabilization techniques include development of anti-shear tunnels, anti-slide piles, anchor cables, concrete support structure, etc. Further, the slope stability and reinforcement effects using various stabilization techniques are studied by using finite element strength reduction method. Moreover, in situ monitoring is carried out and the data obtained is analyzed. From the results, it is observed that the deformations that are detected using multipoint extensometers have decreased after the installation of remedial reinforcements. From the analysis of remedial reinforcement methods, it is found that the most critical reinforcement is the development of anti-shear tunnels. In order to monitor the stresses in stirrups and the propagation of cracks in the anti-shear tunnels, three levels of safety monitoring index are proposed. The safety monitoring index is developed based on the results obtained by the simulation of the process of failure of the reinforced slope. The developed safety monitoring index is further applied to the Suofengying project in order to evaluate the overall stability of the slope. The results obtained by monitoring indicate that the performance of the reinforcement structures is satisfactory and the slope has better stability. The methodology proposed in this work shall be useful for similar projects to obtain stability of slopes.

@en

During the long-term operating period, the mechanical parameters of hydraulic structures and foundation deteriorated gradually because of the environmental factors. In order to evaluate the overall safety and durability, these parameters should be calculated by some accurate analysis methods, which are hindered by slow computational efficiency and optimization performance. The improved deep Q-network (DQN) algorithm combined with the deep neural network (DNN) surrogate model was proposed in this paper to ameliorate the above problems. Through the study cases of different zoning in the dam body and the actual engineering foundation, it is shown that the improved DQN algorithm has a good application effect on inversion analysis of material mechanical parameters in this paper.

@en