Hasil untuk "Hydraulic engineering"

QASEM ABDELAL, MUHAMMAD RASOOL AL-KILANI, JAWAD AL-BAKRI et al.

Pavlo Kuznietsov, Olha Biedunkova, Ihor Statnik et al.

Abstract This study examines the process of environmental status determination based on heavy metal contamination levels in the Styr River, Ukraine, within a zone of anthropogenic load. Background The assessment of environmental pollution by heavy metals is crucial for understanding anthropogenic impacts on aquatic ecosystems. This study focuses on analyzing the process of environmental status determination in the Styr River based on heavy metal concentration levels in the zone of anthropogenic load. Methods Systematic monitoring was conducted from 2018 to 2022, with surface water samples collected downstream of the Rivne Nuclear Power Plant. The concentrations of Zn, Cd, Pb, Cu, Ni, Mn, As, and Cr were analyzed using inductively coupled plasma optical emission spectroscopy. Pollution Coefficients (PC) were calculated to assess contamination levels according to Ukrainian environmental standards. Results The findings indicate that the overall water quality for domestic, drinking, and fishery purposes predominantly falls within Class I (not polluted), with occasional instances of slight pollution in fishery use. Network analysis revealed significant correlations between PC levels and heavy metal concentrations, with Cd being the primary contributor to pollution in domestic and drinking water, while Cu had the highest influence on fishery water pollution. Conclusions This study highlights the effectiveness of PC calculations and network analysis in assessing environmental status, demonstrating their applicability for pollution monitoring and risk assessment. The results support the need for continued monitoring and targeted mitigation measures to prevent heavy metal contamination in aquatic systems. Graphical Abstract

José Peixoto, Alexis Gonzalez, Janki Bhimani et al.

Programmable caching engines like CacheLib are widely used in production systems to support diverse workloads in multi-tenant environments. CacheLib's design focuses on performance, portability, and configurability, allowing applications to inherit caching improvements with minimal implementation effort. However, its behavior under dynamic and evolving workloads remains largely unexplored. This paper presents an empirical study of CacheLib with multi-tenant settings under dynamic and volatile environments. Our evaluation across multiple CacheLib configurations reveals several limitations that hinder its effectiveness under such environments, including rigid configurations, limited runtime adaptability, lack of quality-of-service support and coordination, which lead to suboptimal performance, inefficient memory usage, and tenant starvation. Based on these findings, we outline future research directions to improve the adaptability, fairness, and programmability of future caching engines.

LIU Jia, LI Weihua, XUE Yi et al.

ObjectiveFor effective stimulation of energy exploitation from carbonatite geothermal reservoirs, this study aims at optimizing the acid fracturing process, an approach leveraging the chemical corrosion effect in tandem with the pressure effect during acid injection into formation. Although this synergy of impacts can enhance the permeability of geothermal reservoirs and thus augment the development potential and economic returns of such reservoirs, the optimization of the acidification fracturing scheme is quite challenging, given the pronounced heterogeneity within the reservoirs which brings disparities in permeability and storage capacity among different strata. Therefore, to boost the efficiency of acid pumping into the formation and corrosion of fractures, and to consequently improve the seepage capacity of reservoirs, the research targets the necessary aspects for the precise control of the fracturing technology, involving the site selection of injection wells, and the volume and rate of acid injection, meanwhile considering the actual conditions of reservoir fractures. Because of the complexity of these problems, enhancing computational efficiency is also a primary consideration.MethodsA coupled thermo-hydro-chemical model was developed for the analysis of hydrothermal carbonatite acid fracturing. Note that due to the usually low permeability and thermal conductivity of the bedrock in geothermal reservoirs, and the short fracturing duration, the seepage of acid fluid and heat transfer within the bedrock are often weak during the fracturing process. Therefore this context, to save calculation costs, the three-dimensional thermal reservoir model was simplified into a two-dimensional heterogeneous fracture model only taking into account the seepage, heat transfer, and acid concentration transport of the fracture surface, where the corrosion effect of acid was related to the physical and chemical parameters of the bedrock. Based on the turning band method, two-dimensional spatially-autocorrelated heterogeneous fracture aperture fields were generated with MATLAB and imported into COMSOL Multiphysics as an interpolation function. Four physical modules were adopted in the COMSOL software to establish the numerical model for the analyses of the fracture surface – Darcy's Law to solve the pressure field, Heat Transfer in Porous Media to compute the temperature field, Transport of Diluted Species in Porous Media to calculate the acid concentration field, and Domain Ordinary Differential Equations and Differential-Algebraic Equations to resolve the acidification fracture aperture. At the same time, because of the multiple physical fields involved, a segregated coupled solver was employed to separately solve each physical field in the process of transient solution. The reliability of the numerical model was demonstrated through the simulation of a solute diffusion problem accounting for groundwater flow. To optimize the acid fracturing process, five optimization parameters were studied, involving the injection point coordinates, injection pressure, injection temperature, acid concentration, and two objective functions, namely the average chemical aperture of the fracture surface and the acidification area. Considering the actual situation in engineering practices, the reasonable value ranges of each parameter were specified. The Design of Experiments sampling method was applied with a global arrangement to determine the data set used for the training of computationally efficient agent models, which were then constructed by employing the deep neural network algorithm, a machine learning technology. The rationality and accuracy of the agent models were verified by comparing the predicted values obtained by the agent models with the true values computed by the coupled numerical model. Finally, by combining the agent models with multiple optimization algorithms, i.e. the Nelder-Mead, BOBYQA, and COBYLA algorithms, the optimal value of the objective function could be determined when the maximum number of iterations of an optimization algorithm was reached.Results and Discussions　The results indicated that the homogeneity of the fracture decreased with the increase of correlation length, and away from the injection point, the acid concentration and fracture aperture distribution gradually smoothed after the acid fracturing. The concentration of fracturing fluid rose with proximity to the fracturing well, leading to a more efficient acid-rock reaction and consequently larger fracture aperture. Concerning surrogate models, the performance of prediction was improved as the scale of the training data enlarged, where the R² value increased while both the root mean square error and the absolute average error decreased, illustrating a steady enhancement in the agent models' predictive accuracy. Specifically, when the surrogate model was established using 500 training observations, the fit between the predicted and true values was optimized, demonstrating the importance of a sufficient amount of training data in fostering the precision of the agent models. Moreover, in the comparison of three optimization methods, the values of the objective function obtained by the BOBYQA and Nelder-Mead algorithms gradually stabilized with the rise of iteration number, while the COBYLA algorithm with a linear approximation scheme required more iterations to reach the maximum value of the objective function. Among them, the BOBYQA algorithm could maximize the objective function value and find the optimal solution in the fewest iterations.ConclusionsBased on machine learning techniques and trained with the data set obtained by coupled numerical simulation, the proposed method utilizing a surrogate model offers advantages for the optimization of geothermal reservoir modification. The analysis using a surrogate model precisely determines the optimal injection parameters for acidification fracturing. Meanwhile, this approach effectively reduces the computational costs and thus improves the efficiency in the modification process of acidification fracturing, which contributes to the development and utilization of geothermal energy from carbonatite reservoirs. The study provides scientific insights into the enhancement of the productivity and economic benefits of geothermal energy exploitation.

Pierre-Emmanuel Goffi, Raphaël Tremblay, Bentley Oakes

Successfully engineering interactive industrial DTs is a complex task, especially when implementing services beyond passive monitoring. We present here an experience report on engineering a safety-critical digital twin (DT) for beer fermentation monitoring, which provides continual sampling and reduces manual sampling time by 91%. We document our systematic methodology and practical solutions for implementing bidirectional DTs in industrial environments. This includes our three-phase engineering approach that transforms a passive monitoring system into an interactive Type 2 DT with real-time control capabilities for pressurized systems operating at seven bar. We contribute details of multi-layered safety protocols, hardware-software integration strategies across Arduino controllers and Unity visualization, and real-time synchronization solutions. We document specific engineering challenges and solutions spanning interdisciplinary integration, demonstrating how our use of the constellation reporting framework facilitates cross-domain collaboration. Key findings include the critical importance of safety-first design, simulation-driven development, and progressive implementation strategies. Our work thus provides actionable guidance for practitioners developing DTs requiring bidirectional control in safety-critical applications.

Miguel Goulão, Vasco Amaral, Marjan Mernik

Model-driven engineering (MDE) is believed to have a significant impact in software quality. However, researchers and practitioners may have a hard time locating consolidated evidence on this impact, as the available information is scattered in several different publications. Our goal is to aggregate consolidated findings on quality in MDE, facilitating the work of researchers and practitioners in learning about the coverage and main findings of existing work as well as identifying relatively unexplored niches of research that need further attention. We performed a tertiary study on quality in MDE, in order to gain a better understanding of its most prominent findings and existing challenges, as reported in the literature. We identified 22 systematic literature reviews and mapping studies and the most relevant quality attributes addressed by each of those studies, in the context of MDE. Maintainability is clearly the most often studied and reported quality attribute impacted by MDE. Eighty out of 83 research questions in the selected secondary studies have a structure that is more often associated with mapping existing research than with answering more concrete research questions (e.g., comparing two alternative MDE approaches with respect to their impact on a specific quality attribute). We briefly outline the main contributions of each of the selected literature reviews. In the collected studies, we observed a broad coverage of software product quality, although frequently accompanied by notes on how much more empirical research is needed to further validate existing claims. Relatively, little attention seems to be devoted to the impact of MDE on the quality in use of products developed using MDE.

Saba Sanami, Amir G. Aghdam

Reliable aero-engine anomaly detection is crucial for ensuring aircraft safety and operational efficiency. This research explores the application of the Fisher autoencoder as an unsupervised deep learning method for detecting anomalies in aero-engine multivariate sensor data, using a Gaussian mixture as the prior distribution of the latent space. The proposed method aims to minimize the Fisher divergence between the true and the modeled data distribution in order to train an autoencoder that can capture the normal patterns of aero-engine behavior. The Fisher divergence is robust to model uncertainty, meaning it can handle noisy or incomplete data. The Fisher autoencoder also has well-defined latent space regions, which makes it more generalizable and regularized for various types of aero-engines as well as facilitates diagnostic purposes. The proposed approach improves the accuracy of anomaly detection and reduces false alarms. Simulations using the CMAPSS dataset demonstrate the model's efficacy in achieving timely anomaly detection, even in the case of an unbalanced dataset.

ZHANG Jian 1, 2, YANG Ligong 1, 2, ZHANG Yuting 1, 2, ZHAO Yue 1, 2, LIU Yongjun 3, WU Wenhua 3

In view of the deformation of sheet piles and the pile-soil interaction mechanism of revetment slopes during channel excavation, the key physical parameters such as displacement of sheet piles, earth pressure and deformation of the adjacent soils are monitored by selecting typical sections for the field tests. Combined with the theoretical calculation, the stability of the sheet piles and the bank slope is analyzed, and the interaction mechanism between the soil deformation behind the piles and the active earth pressure with complex geological conditions is explored, and some useful conclusions are drawn. The horizontal displacement of the soils behind the sheet piles increases gradually with the excavation. When the slope reaches stability, the soil displacement at different depths is about 0.8~1 mm, and the horizontal displacement at the top of the sheet piles is about 1.2 mm. During channel excavation, the active earth pressure of the sheet piles gradually decreases with the excavation, and the reduction value of the earth pressure at pile side at different depths is about 1 kPa. Compared with the theoretical results, the field test results are small, indicating that the theoretical calculation is slightly conservative for the field tests. The conclusion can provide a useful reference for the revision of the existing related specifications for the sheet piles.

Raffael Theiler, Olga Fink

Pumped-storage hydropower plants (PSH) actively participate in grid power-frequency control and therefore often operate under dynamic conditions, which results in rapidly varying system states. Predicting these dynamically changing states is essential for comprehending the underlying sensor and machine conditions. This understanding aids in detecting anomalies and faults, ensuring the reliable operation of the connected power grid, and in identifying faulty and miscalibrated sensors. PSH are complex, highly interconnected systems encompassing electrical and hydraulic subsystems, each characterized by their respective underlying networks that can individually be represented as graphs. To take advantage of this relational inductive bias, graph neural networks (GNNs) have been separately applied to state forecasting tasks in the individual subsystems, but without considering their interdependencies. In PSH, however, these subsystems depend on the same control input, making their operations highly interdependent and interconnected. Consequently, hydraulic and electrical sensor data should be fused across PSH subsystems to improve state forecasting accuracy. This approach has not been explored in GNN literature yet because many available PSH graphs are limited to their respective subsystem boundaries, which makes the method unsuitable to be applied directly. In this work, we introduce the application of spectral-temporal graph neural networks, which leverage self-attention mechanisms to concurrently capture and learn meaningful subsystem interdependencies and the dynamic patterns observed in electric and hydraulic sensors. Our method effectively fuses data from the PSH's subsystems by operating on a unified, system-wide graph, learned directly from the data, This approach leads to demonstrably improved state forecasting performance and enhanced generalizability.

Dexian Ma, Yirong Liu, Wenbo Liu et al.

This work proposes a data-driven modeling and the corresponding hybrid motion control framework for unmanned and automated operation of industrial heavy-load hydraulic manipulator. Rather than the direct use of a neural network black box, we construct a reversible nonlinear model by using multilayer perceptron to approximate dynamics in the physical integrator chain system after reversible transformations. The reversible nonlinear model is trained offline using supervised learning techniques, and the data are obtained from simulations or experiments. Entire hybrid motion control framework consists of the model inversion controller that compensates for the nonlinear dynamics and proportional-derivative controller that enhances the robustness. The stability is proved with Lyapunov theory. Co-simulation and Experiments show the effectiveness of proposed modeling and hybrid control framework. With a commercial 39-ton class hydraulic excavator for motion control tasks, the root mean square error of trajectory tracking error decreases by at least 50\% compared to traditional control methods. In addition, by analyzing the system model, the proposed framework can be rapidly applied to different control plants.

Wenfei Liu, Xiaoling Su, Gengxi Zhang et al.

Abstract Projection of future dry‐wet evolution is essential for making long‐term regional climate adaptation strategies. In this study, the projection of regional dry‐wet evolution is conducted with a careful consideration on uncertainty attribution. The Evaporative Stress Index (ESI) is adopted due to its physical mechanisms for taking evaporative demand into account. A three‐dimensional framework is constructed for quantifying the range of uncertainty of the ESI in which six Global Climate Models (GCMs) in Coupled Model Intercomparison Project 6 (CMIP6), three latest Shared Socioeconomic Pathway (SSP) scenarios, and six Potential Evapotranspiration Models (PETMs) are used. The framework provides 108 different ESI simulations for two future periods: 2041–2070 (mid‐future) and 2071–2100 (far‐future). An agglomerative‐hierarchical clustering method and the Analysis of Variance methodology are employed to evaluate the relative contribution of each uncertainty source. The region of Northwest China is used as a case to illustrate the effectiveness of the proposed framework. The results indicate that most of the parts in Northwest China would experience dry mitigation in both mid‐future and far‐future periods. Projected ESI by PM[CO2] model and ACCESS‐ESM1‐5 suggestes a higher tendency for dry mitigation. Hierarchical clustering analysis of the 108 sets of ESI predictions indicate that most clusters are dominated by GCM forcing, and one cluster is dominated by the SSP1‐2.6 scenario. Furthermore, GCM‐related uncertainty′s relative contribution to the total projection uncertainty is the greatest, with an average value of 49.98% in the far future (i.e., 2071–2100 s). Although the contribution of SSP uncertainty is smaller (21.68%−28.43%), it increases in far‐future over mid‐future. The case study indicates that the large scale ensemble prediction of ESI and its uncertainty analysis provide a more comprehensive data set on climate change and help water managers to gain in‐depth understanding of future trends of drought projections.

DUAN Guoxiu, JIA Xiaoxu, BAI Xiao et al.

【Objective】 Crop growth and ecological functions in arid and semi-arid loess regions in northwestern China are limited not only by topsoil water directly available to crop but by deep soil water which functions as a reservoir. Slopes and lands formed by artificial retaining are two typical geographical units in small watersheds in the loess plateau. This paper presents a method to estimate soil water distribution in them up to 12.5 m deep. 【Method】 The method was based on electrical resistivity tomography (ERT). We measured electrical resistivity of the soil in typical slope and retained land in the small Liudaogou watershed in northern Shaanxi province. Using the measured soil volumetric water content, a power function relating the resistivity to soil water content was established, from which we calculated water distribution and water storage in the 0~12.5 m profile in both the slope and the retained land. 【Result】 The distribution of electrical resistivity over the slope was significantly affected by slope position, with the resistivity decreasing gradually from the slope top to the slope toe. Such changes were associated with vegetation consumption of the topsoil water and redistribution of the infiltrated precipitation over the slope. The vertical distribution of the resistivity in the retained land showed a high-low-high variation; this was also related to root water uptake from the topsoil and precipitation recharge to the subsoil. ERT inversion showed co-existence of a saturated zone and an unsaturated zone in the retained land, while the slope was partly saturated. In a 1.8×104 m2 of retained land, there was 1.49×104 m3 of water in the 0~6.5 m unsaturated layer, and 5.10×104 m3 of water in the saturated layer. In a same area but on the slope, there was only 2.59×104 m3 of water in the 0~12.5 m soil layer. 【Conclusion】 Retained land contains more water than slope, and the deep soil in it functions as a reservoir banking infiltration water in wet seasons. ERT is suitable for measuring spatiotemporal variation in soil moisture in both slopes and flatten plains in the loess plateau.

Sy Hung Mai, Huy-Khanh Dang, Van Thuan Nguyen et al.

This work comprehensively investigates the realistic ultimate resistance of steel frame structures based on the probability study model of input random variables due to uncertainty of material properties and geometric parameters. Four independent parameters and thirteen of their dependent components vary randomly in a variational range and seven scenarios of random combination are investigated by the probabilistic modeling, which incorporates the advanced inelastic analysis with direct Monte Carlo simulation. The analysis results for a typical example shows that the stochastic analysis result of steel frame's realistic ultimate resistance is lower than its deterministic analysis result, indicating that a stochastic analysis is needed to comprehensively qualify the safety of the structure. Moreover, a spectrum response, which is produced based on stochastic analysis, provides a better scientific solution.

Bradley P. Allen

We describe a method, based on Jennifer Nado's proposal for classification procedures as targets of conceptual engineering, that implements such procedures by prompting a large language model. We apply this method, using data from the Wikidata knowledge graph, to evaluate stipulative definitions related to two paradigmatic conceptual engineering projects: the International Astronomical Union's redefinition of PLANET and Haslanger's ameliorative analysis of WOMAN. Our results show that classification procedures built using our approach can exhibit good classification performance and, through the generation of rationales for their classifications, can contribute to the identification of issues in either the definitions or the data against which they are being evaluated. We consider objections to this method, and discuss implications of this work for three aspects of theory and practice of conceptual engineering: the definition of its targets, empirical methods for their investigation, and their practical roles. The data and code used for our experiments, together with the experimental results, are available in a Github repository.

Andreas Möri, Brice Lecampion

In impermeable media, a hydraulic fracture can continue to expand even without additional fluid injection if its volume exceeds the limiting volume of a hydrostatically loaded radial fracture. This limit depends on the mechanical properties of the surrounding solid and the density contrast between the fluid and the solid. Self-sustained fracture growth is characterized by two dimensionless numbers. The first parameter is a buoyancy factor that compares the total released volume to the limiting volume to determine whether buoyant growth occurs. The second parameter is the dimensionless viscosity of a radial fracture at the time when buoyant effects become of order 1. This dimensionless viscosity notably depends on the rate at which the fluid volume is released, indicating that both the total volume and release history impact self-sustained buoyant growth. Six well-defined propagation histories can be identified based on these two dimensionless numbers. Their growth evolves between distinct limiting regimes of radial and buoyant propagation, resulting in different fracture shapes. We can identify two growth rates depending on the dominant energy dissipation mechanism (viscous flow vs fracture creation) in the fracture head. For finite values of material toughness, the toughness-dominated limit represents a late-time solution for all fractures in growth rate and head shape (possibly reached only at a very late time). The viscosity-dominated limit can appear at intermediate times. Our three-dimensional simulations confirm the predicted scalings and highlight the importance of considering the entire propagation and release history for accurate analysis of buoyant hydraulic fractures.

Tiina Nõges, Sirje Vilbaste, Mark J. McCarthy et al.

We analysed long-term (1992–2020) changes in fertiliser use, wastewater treatment, and river water nutrient status in Estonia (N-E Europe) in the context of changing socio-economic situations and legislation. We hypothesised that improved regulation of fertiliser usage and wastewater treatment are reflected as declining riverine nutrient concentrations, with the largest relative improvements occurring in catchments with initially high proportions of point source loading. We analysed nutrient dynamics in 16 rivers differing by catchment land use, population and livestock densities. Data on fertiliser use and wastewater treatment originated from the Statistics Estonia database, and riverine nutrient concentrations from the State Environmental Monitoring Database. We clustered the rivers by their catchment properties and analysed trends in their nutrient status. Point source nutrient loading reductions explained most of the decline in riverine nutrient concentrations, whereas application of mineral fertilisers has increased, hindering efforts to reach water quality and nutrient load targets set by the EU Water Framework Directive and the Baltic Sea Action Plan. Highest nitrogen concentrations and strongest increasing trends were found in rivers within the Nitrate Vulnerable Zone, indicating violation of the EU Nitrates Directive. To comply with these directives, resource managers must address non-point source nutrient loading from river watersheds. HIGHLIGHTS Drop in point source loading explained the decline in riverine nitrogen (N) and phosphorus (P) since 1994.; Fertiliser and wastewater management measures failed short to meet the water quality and nutrient load targets set by the EU Water Framework Directive.; Highest N concentrations and strongest increasing trends were found in rivers within the nitrate vulnerable zone violating the EU Nitrates Directive.;

Hassan Hage, Emmanouil Seferis, Vahid Hashemi et al.

Product Engineering Processes (PEPs) are used for describing complex product developments in big enterprises such as automotive and avionics industries. The Business Process Model Notation (BPMN) is a widely used language to encode interactions among several participants in such PEPs. In this paper, we present SMC4PEP as a tool to convert graphical representations of a business process using the BPMN standard to an equivalent discrete-time stochastic control process called Markov Decision Process (MDP). To this aim, we first follow the approach described in an earlier investigation to generate a semantically equivalent business process which is more capable of handling the PEP complexity. In particular, the interaction between different levels of abstraction is realized by events rather than direct message flows. Afterwards, SMC4PEP converts the generated process to an MDP model described by the syntax of the probabilistic model checking tool PRISM. As such, SMC4PEP provides a framework for automatic verification and validation of business processes in particular with respect to requirements from legal standards such as Automotive SPICE. Moreover, our experimental results confirm a faster verification routine due to smaller MDP models generated from the alternative event-based BPMN models.

Alex V. Lukyanov, Tristan Pryer, Edward Calver

The formation of a planar hydraulic jump has been analysed in the framework of a full depth-averaged thin film model (DAM) with surface tension effects included. We have demonstrated regular weak solutions of the full DAM and analysed surface tension effects. It has been shown that surface tension effects within the parameter range relevant to the recent experiments are expected to be very weak and practically negligible. The developed methodology can be used in the analysis of laminar flow regimes and as a benchmark in developing full scale hydrodynamic models.

Sabah Al-Fedaghi, Mahdi Modhaffar

Models are fundamentally crucial to many scientific fields, including software engineering, systems engineering, enterprise modeling, and business modeling. This paper focuses on diagrammatic conceptual modeling, as opposed to mathematical or computational models, wherein a conceptual model is a translation of reality processes into an abstract mechanism that has similar structure and parallel events of the external processes. Although various modeling approaches exist, including UML (Unified Modeling Language) in software engineering and its dialect, SysML (System Modeling Language), in systems engineering, several difficulties arise in such models, including the problem of model multiplicity that is related to the lack an integrated view of structure and behavior. This paper generalizes conceptual modeling to be applied in organizations at large. According to authorities, the so-called organization theory portrays organizations as machine-like systems. As a machine, an organization coordinates its parts to transform inputs into outputs. Therefore, we synthesize the notion of an organization as a machine and apply a new modeling methodology called thinging machine (TM) to real engineering operations. The results show the viability of the TM methodology serving as a foundation for high-level modelling of systems.

Dashan Yang, Chongshi Gu, Yantao Zhu et al.

Dams are the main water retaining structures in the hydraulic engineering field. Safe operations of dams are important foundations to ensure the hydraulic functionalities of these engineering structures. Deformation, as the most intuitive feature of the dams’ operation behaviors, can comprehensively reflect the dam structural states. In this case, the analysis of the dam prototype deformation data and the establishment of a real-time prediction model become frontier research contents in the field of dam safety monitoring. Considering the multi-nonlinear relationships between dam deformation and relative influential factors as well as the time lag effect of these influential factors, this article adopts long-short-term memory (LSTM) network algorithm in deep learning to deal with the long-term dependence existing in dam deformation and explore the deformation law. The method proposed in this work can effectively avoid the gradient disappearance and gradient explosion problems by using the recurrent neural network (RNN). In addition, this work adopts the Attention mechanism to screen the information that has significant influence on deformation, combining the Adam optimization algorithm that has high calculation efficiency and low memory requirement to improves the learning accuracy and speed of the LSTM. The model overfitting is avoided by applying the Dropout mechanism. The effectiveness of this proposed model in studing the long time series deformation prediction of concrete dams is confirmed by case studies, whose MSE (mean square error) and other 4 error indexes can be reduced.