A fast and accurate moisture content (MC) measurement of sand gravel is essential for hydraulic engineering project sites. Most existing measurement methods are unimodal, facing non-robust against external interference. To address this issue, a deep multimodal fusion (DMF) model for measuring the MC of sand gravel using images, near-infrared (NIR) spectra, and dielectric data, is proposed. A modified bottleneck transformer network (BoTNet) added with an extremely efficient spatial pyramid (EESP) block is first proposed to extract image features from different receptive fields. The improved convolutional neural network with attention blocks added (A-CNN) and gated recurrent unit with attention blocks added (A-GRU) networks are then adopted to extract local and sequential features from NIR spectra, respectively. The square root of dielectric data and above multimodal features are effectively fused according to their contribution to the target indicator in the Fusion module. Among other comparative models, the DMF model yielded the best performance (R2 = 0.962, RMSE = 0.645, RPD = 5.124) on the original sand gravel dataset, and still maintained the best accuracy (the average R2 and RPD mostly exceeded 0.85 and 2.5, respectively) when against general external noise.

Discrete fracture network (DFN) modeling is a popular method for studying reservoir characteristics. However, on one hand of the existing DFN studies, the correlations that may exist among the multi-dimensional parameters of the fracture (dip, dip direction, trace length, aperture, and roughness) were ignored and each property was estimated independently. On the other hand, the 3D DFN models were simplified as flats without roughness. Therefore, this study proposes a simulation method for the rough discrete fracture network while considering multi-parameter joint distribution to make up for the stated deficiencies: (1) An improved deep learning model for the joint distribution estimation and joint sampling of the multi-dimensional parameters of the fracture is proposed, which is the Neural Spline Flow improved by the Multimodal Distribution (NSF-MD). The initial distribution of the Neural Spline Flow (NSF) is improved from a unimodal Gaussian to a multimodal mixed Gaussian, which enhances the fitting ability of the NSF model for the multimodal joint distribution. (2) A modeling assumption of the 3D rough fracture surface is presented to improve the conventional flat ones. In this way, a 3D rough fracture surface is generated by the NURBS tensor product of the fractal traces, and the fractal trace is simulated parametrically based on the fractal dimension. Finally, an outcrop DFN modeling application from southwest China indicates that the NSF-MD can simulate the correlations among the fracture multi-dimensional parameters and that the 3D rough discrete fracture network model (RDFN) increases the representation of the fracture roughness. The Wasserstein distance, which is an indicator quantifying the accuracy of the joint multi-parameter estimation, of the NSF-MD model was 72.4% and 81.9% better than that of the NSF model and that of the conventional method with parameters estimated independently, respectively. The relative distance error (RDE) and the global angle error (GAE) of the fractal traces were 87.9% and 88.3% more accurate than those of the conventional linear traces, respectively.

Parametric 3D geological modeling facilitates the efficient updating of geological models. A comprehensive review of existing geological modeling methods reveals that two critical issues need to be addressed for realizing parametric 3D geological modeling: (1) complex surface topology caused by direct interpolation of geological data creates a significant obstacle to the parametric characterization of irregular geometry of geological interfaces, and (2) expert knowledge input essential during modeling the complex geological bodies have not been formalized quantitatively and objectively, which poses another barrier to the automation of parametric geological modeling. In response, this work presents a parametric 3D geological modeling method achieving the parametric expressions of geological interfaces and automatic modeling for different geological bodies without manual interaction. In this study, the NURBS Surface Dynamic Topology (NURBS-SDT) method is proposed to regularize the complex topological structure of the geological interfaces, thereby expressing them parametrically. Furthermore, 3D coordinates are converted into the control parameters of the geological interface geometry. On the other hand, subjective expert knowledge input is translated into objective modeling rules through the proposed Boolean Logic Sequence of Oriented Geological Interfaces (BLSOGI) method, which means different geological bodies can be automatically modeled. A practical application to a city in southern China demonstrates a 570-fold increase in efficiency of model updating by the proposed method compared to the conventional interactive approach. The comparison with four other representative automatic and semi-automatic methods also shows the higher applicability of the proposed method in modeling complex strata. Thus, this study can greatly improve the efficiency of the engineering geological investigation and design.

Grouting power is a vital parameter that can be used as an indicator for simultaneously controlling grouting pressure and injection rate. Accurate grouting power prediction contributes to the real-time optimization of the grouting process, guaranteeing grouting safety and quality. However, the strong nonlinearity of the grouting power time series makes the forecasting task challenging. Hence, this paper proposes a novel hybrid model for accurate grouting power forecasting. First, empirical wavelet transform (EWT) is employed to decompose the original grouting series into several subseries and one residual adaptively. Second, partial autocorrelation function (PACF) is applied to identify the optimal input variables objectively. Then, support vector regression (SVR) is adopted to obtain prediction outcomes of each subseries, while an improved Jaya (IJaya) algorithm by coupling chaos theory and Lévy flights to improve the algorithm’s accuracy performance is proposed to optimize the SVR hyperparameters. Finally, the prediction results of decomposed subseries are superimposed to produce the final results. A consolidation grouting project is taken as a case study and the computation results with the RMSE = 0.2672 MPa·L/min, MAE = 0.2165 MPa·L/min, MAPE = 3.85% and EC = 0.9815 demonstrate that the proposed model exhibits superior forecasting ability and can provide a viable reference for grouting construction.

Construction simulation has been widely applied in schedule analysis. However, traditional simulation is based on static models built in the planning or design phase, which focuses on overall project-level schedule analysis. To provide activity-level simulation for on-site schedule management, a construction phase oriented dynamic simulation method is proposed, which takes roller compacted concrete (RCC) dam placement process as an example. Considering various inner-layer and inter-layer activities and different construction organization modes, a detailed placement process simulation model is built. Based on construction data collected by real-time monitoring, a construction activity modeling method is given. Additionally, Dirichlet process mixture (DPM) models are applied for simulation parameter updates, which endows density estimation with considerable flexibility and robustness. A fast inference algorithm is also proposed to realize the fast posterior computation of DPM models. The proposed method is tested by an RCC dam project in southwest China. The results show that the proposed method can reflect the dynamic features of the actual placement process in the construction phase and provide accurate schedule predictions for on-site construction management.

During the storehouse surface rolling construction of a core rockfill dam, the spreading thickness of dam face is an important factor that affects the construction quality of the dam storehouse’ rolling surface and the overall quality of the entire dam. Currently, the method used to monitor and control spreading thickness during the dam construction process is artificial sampling check after spreading, which makes it difficult to monitor the entire dam storehouse surface. In this paper, we present an in-depth study based on real-time monitoring and control theory of storehouse surface rolling construction and obtain the rolling compaction thickness by analyzing the construction track of the rolling machine. Comparatively, the traditional method can only analyze the rolling thickness of the dam storehouse surface after it has been compacted and cannot determine the thickness of the dam storehouse surface in real time. To solve these problems, our system monitors the construction progress of the leveling machine and employs a real-time spreading thickness monitoring model based on the K-nearest neighbor algorithm. Taking the LHK core rockfill dam in Southwest China as an example, we performed real-time monitoring for the spreading thickness and conducted real-time interactive queries regarding the spreading thickness. This approach provides a new method for controlling the spreading thickness of the core rockfill dam storehouse surface.

Effective construction scheme planning is critical for schedule management, but heavy rain can affect construction processes. In previous studies, stochastic rainfall characteristics are often ignored, and their impact on macro‐ and microconstruction states are not depicted comprehensively. This research presents a construction simulation model to design reasonable construction schemes considering impact of stochastic rainfall. First, a rainfall model suitable for areas with heavy rainfall and uneven seasonal rainfall distribution is built. Then, multiaspect indicators are defined to intuitively quantify rainfall impact. Two case studies are conducted to evaluate applicability of the proposed method. Results demonstrate that the developed rainfall model aligns closely with observed data. Simulation findings reveal that if stochastic rainfall characteristics are ignored, the schedule and queuing probability of trucks will be underestimated, while machinery utilization will be overestimated. This research provides an effective simulation tool for determining adaptive measures to mitigate impacts of rainfall events.

The parameters of existing roller-compacted concrete (RCC) dam construction simulation are usually fixed based on experience while the actual construction conditions of an RCC dam change during the process of the project. The simulation accuracy of an RCC dam is therefore reduced because the change has not been considered. A new method for RCC dam construction simulations based on real-time monitoring is presented in this paper. First, real-time monitoring technology is used to collect and analyze the actual construction information. Second, meteorological data obtained from the real-time monitoring system are analyzed using the fuzzy average function method, and the weather conditions of the next stage are forecasted. Then the construction schedule simulation model is updated via the Bayesian update method. Results of the analysis are used as the input to the construction simulation parameters, and the construction simulation is performed. A real-world engineering example is presented to compare the simulation results with the actual construction schedule. The results demonstrate that the method can effectively improve the accuracy and real-time performance of construction simulations.

The quality of compaction is key to the safety of dam construction and operation. However, because of incomplete information about the construction process and the unknown relationship between compaction quality and the factors that influence it, traditional evaluation methods such as neural networks and multivariate linear regression models fail to take uncertainty fully into account. This paper proposes a cloud-fuzzy method for assessing compaction quality by considering randomness, fuzziness, and incomplete information. The compaction parameters and material source parameters are the key parameters in the assessment of compaction quality. A five-layer neural-network model of compaction quality assessment is established that considers compacted dry density and its classification membership and probability as the criteria, and the rolling speed, rolling passes, and compacted layer thickness as alternatives. Because of uncertainties in the criteria and alternatives, the cloud-fuzzy method, in which a fuzzy neural network is extended with a cloud model to handle uncertain and fuzzy problems more effectively, is introduced to determine the compaction quality. A case study is presented to evaluate the compaction quality of a hydropower project in China. The results indicate that the cloud-fuzzy model is feasible in relation to precision and makes up for the sole focus on precision by traditional methods. The proposed method provides a triple index for understanding compaction quality, which facilitates assessment of the compaction quality of an entire dam surface.

The compaction quality of earth-rock dam materials is a major concern in the evaluation of earth-rock dams. Current compaction quality assessment methods, such as graphical reports or simple prediction models, are imprecise and can cause unobserved quality assessment defects. These methods do not comprehensively consider factors that affect the compaction quality because they do not integrate heterogeneous construction data sets collected by different data acquisition systems. In this research, a method of assessing compaction quality on the basis of support vector regression (SVR), the chaos-based firefly algorithm, is presented. The assessment method has three stages. In the first stage, a chaotic firefly algorithm (CFA) is proposed to optimize the SVR hyperparameters. In the second stage, a multisource heterogeneous data integration subsystem based on the compaction monitoring system is designed, in which compaction monitoring data, material source statistical data, and detected data from test pits are integrated. Finally, the optimized SVR is used to evaluate the compaction quality of the storehouse surface. The significance of the proposed method is threefold: first, it integrates both chaos theory and the firefly algorithm to optimize the SVR hyperparameters; second, it integrates heterogeneous construction data, allowing comprehensive consideration of factors that affect the compaction quality; and third, it has high prediction accuracy because it implements structural risk minimization. Compared with current models based on empirical risk minimization, the proposed method performs the best according to several error measures. (C) 2018 American Society of Civil Engineers.

Compaction quality is one of the most important aspects of the construction quality control of earth-rockfill dams. In recent years, real-time compaction quality monitoring technology based on a global navigation satellite system (GNSS) for earth-rockfill dams has realized effective control for earth-rockfill dam construction quality. However, many high earth-rockfill dams have been built in deep, narrow valleys, where the satellite signal is blocked by the tall, steep slopes such that the accuracy of the location is not satisfactory or that determining the location is fully impossible, thus seriously affecting the continuity and accuracy of real-time monitoring. The existing monitoring technology cannot meet the monitoring requirements in deep, narrow valleys. This paper establishes the theory and a mathematical model of real-time compaction quality monitoring in deep, narrow valleys and proposes a new method for real-time compaction quality monitoring based on positioning compensation technology (PCT), which combines GNSS and Robotic Total Station (RTS). An all-terrain and whole-process compaction quality monitoring of earth-rockfill dam construction is realized through this new method, making up for the shortcomings of the monitoring of the compaction process that relies solely on GNSS. Practical application shows that the method guarantees the objectivity and integrity of the real-time monitoring results, which ensures the compaction quality of earth-rockfill dam construction in deep, narrow valleys.

Accurate 3-D fracture network model for rock mass in dam foundation is of vital importance for stability, grouting and seepage analysis of dam foundation. With the aim of reducing deviation between fracture network model and measured data, a 3-D fracture network dynamic modeling method based on error analysis was proposed. Firstly, errors of four fracture volume density estimation methods (proposed by ODA, KULATILAKE, MAULDON, and SONG) and that of four fracture size estimation methods (proposed by EINSTEIN, SONG and TONON) were respectively compared, and the optimal methods were determined. Additionally, error index representing the deviation between fracture network model and measured data was established with integrated use of fractal dimension and relative absolute error (RAE). On this basis, the downhill simplex method was used to build the dynamic modeling method, which takes the minimum of error index as objective function and dynamically adjusts the fracture density and size parameters to correct the error index. Finally, the 3-D fracture network model could be obtained which meets the requirements. The proposed method was applied for 3-D fractures simulation in Miao Wei hydropower project in China for feasibility verification and the error index reduced from 2.618 to 0.337.

We propose a novel geological modeling method based on T‐splines for computer‐aided design (CAD) and building information modeling (BIM) systems of geotechnical engineering and perform original research on special T‐splines modeling strategies and the algorithm aiming at depicting structural complexity to develop and introduce the T‐splines technology into engineering geological modeling. A methodology of parametric geological modeling with T‐splines is established, where a topology‐geometry modeling strategy is adopted, the inhomogeneity, arbitrary connectivity and discontinuity of geological structures are quantified and associated with T‐spline surface elements in the topology phase. A representative parametric algorithm called IBALR is presented in the context of dam foundation geological modeling to generate an inhomogeneous local refined T‐mesh with high genus topology which can fit a complex geological layer. The proposed method is flexible and effective in improving geological representation within geotechnical engineering.

Time, cost, and quality are three key control factors in rockfill dam construction, and the tradeoff among them is important. Research has focused on the construction time-cost-quality tradeoff for the planning or design phase, built on static empirical data. However, due to its intrinsic uncertainties, rockfill dam construction is a dynamic process which requires the tradeoff to adjust dynamically to changes in construction conditions. In this study, a dynamic time-cost-quality tradeoff (DTCQT) method is proposed to balance time, cost, and quality at any stage of the construction process. A time-cost-quality tradeoff model is established that considers time cost and quality cost. Time, cost, and quality are dynamically estimated based on real-time monitoring. The analytic hierarchy process (AHP) method is applied to quantify the decision preferences among time, cost, and quality as objective weights. In addition, an improved non-dominated sorting genetic algorithm (NSGA-II) coupled with the technique for order preference by similarity to ideal solution (TOPSIS) method is used to search for the optimal compromise solution. A case study project is analyzed to demonstrate the applicability of the method, and the efficiency of the proposed optimization method is compared with that of the linear weighted sum (LWS) and NSGA-II.

Rock-fill dam compaction quality depends on compaction parameters and material parameters. Although it is difficult to gain material sources parameters at any dam surface location, the real-time quality monitoring technology of rock-fill dam construction provides for gaining compaction parameters of any surface location. Based on test pit data and real-time data, establish the combination model connecting adaptive network-based fuzzy inference system and improved back propagation neural network, and realize the nonlinear mapping relationship between compaction parameters and construction quality.-The predicting accuracy of the combination model is higher than single adaptive network-based fuzzy inference system or improved back propagation neural network through model testing. Finally combination model is applied to forecast some dam surface dry density.

Construction simulation is an effective means to describe the dam filling process of a core rockfill dam. However, present research studies adopt a simplified method to analyze the placement of a rockfill subsystem because of various uncertain working constraints and diverse flow shop construction processes, which include preparing, discharging, paving, compacting, and checking. In addition, because the time and cost minimization and equilibrium of the filling intensity are important matters in the design of core rockfill dam construction stages and zones, full consideration of these matters is required during the construction process. In this work, a construction simulation model for a rockfill dam based on flow shop construction is presented. The model can provide different reasonable construction schemes of each filling layer. In addition, to obtain an optimal plan for the construction stages and zones, entropy weight method and the improved genetic algorithm (GA) are used to address the time-cost trade-off problem and to minimize to disequilibrium degree of filling intensity simultaneously. A core rockfill dam in southwest China is taken as a case study. The result demonstrates that the plan obtained by this approach can not only guarantee a low disequilibrium degree of filling intensity (5.66) but also shorten the construction period with a reasonable mechanical investment plan. The comparison between the optimal plan and the plan made in accordance with experience, of which the disequilibrium degree of the filling intensity is 7.96, demonstrates the superiority of the optimal plan and the feasibility of the proposed approach. (C) 2015 American Society of Civil Engineers.

In the light of the shortcomings of the traditionalIn the light of the shortcomings of the traditional methods on archiving monitoring information of dam safety, the principle of web-based 3D visualization based on Unity3D engine was explored for construction managers to grasp dam operation conditions immediately and intuitively, and the web-based 3D visualization Dam Safety Monitoring System (DSMS) was developed in B/S structure. The networked, digital and 3D visualized system was developed to provide an effective analysis platform for the construction managers to make decisions. The research is of great practical significance.

Reservoir Construction Layout, which needs collection and interpretation of generous geographic information and other construction data, is very important and complicated. In this paper, the 3D digital model of Reservoir Construction Layout is established using some software such as 3ds Max. Depending on the established digital model, 3D cruising of the whole scene and local observing of the reservoir are realized. QingLinJing Reservoir, which is located in northeast of Shenzhen City of China, is taken as a case study. The 3D visualization of the whole scene of QingLinJing Reservoir Construction Layout is realized with 3ds Max. The 3D visualization scene is visualized in 3D animation. 3D visual simulation is important to design and optimization of reservoir construction and it provides an effective way for the construction management and decision-making for reservoir construction layout project. © 2010 IEEE.

A complete scheme for solving the key scientific problems associated with high-standard, high-intensity continuous construction of high arch dams was presented. First, based on a coupling analysis of construction system decomposition and coordination for a high arc dam, a mathematical model for real-time control of construction quality and progress that considers complex constraints was developed. Second, a method of progress control was proposed based on a dynamic simulation. Third, a dynamic quality control mechanism was established based on construction information collected using a PDA. Fourth, a system for integrating collected information, progress simulation and quality control analyses under a network environment was developed. Finally, these methods were applied to a practical project to show that each aspect of a construction process can be managed effectively and that real-time monitoring and feedback control can be realized. Our methods provide new theoretical principles and technical measures for quality and progress control in the high arc dam construction process.

Any reservoir construction layout involves the collection and interpretation of large amounts of geographic and other data. Because the traditional methods to analyze reservoir construction layout are not efficient and intuitive, visual simulation is the key to improve the design efficiency and management of the project. In this paper, triangulated irregular network (TIN) is used to help to establish the digital terrain model. The different entity models were established with different methods. The models were matched to show the actual terrain based on 3DS MAX mapping. Gongming reservoir in South China is taken as a case study. The finished scene was visualized in three-dimensional animation, providing a scientific, effective and visual analyzing method for the management and decision-making associated with the project.