The analysis of deformation, a critical aspect of high earth-rock dams, holds immense importance for ensuring the safety and stability of dam operations. The structural behavior of high earth-rock dams exhibits time-varying nonlinear characteristics influenced by materials and loads. Over time, the fitting and prediction abilities of static dam Structural Health Monitoring (SHM) models tend to diminish. To address this, a novel SHM model is proposed in this study. It leverages deep transfer learning to enhance prediction accuracy and generalization by incorporating a starting point timestamp and employing a transfer learning approach. The methodology begins with the construction of an encoder structure based on the graph convolutional network and long and short-term memory model. Additionally, the attention mechanism-based encoder structure is designed to include starting point time markers. Knowledge migration is then executed through transfer learning, thereby improving the model's generalization to the time-varying deformation challenge. The proposed model is applied to a horizontal displacement monitoring project for a 185.5 m-high panel rockfill dam. Ablation experiments demonstrate that the transfer learning method effectively enhances the model's handling of time-varying deformation by improving prediction accuracy, with a more pronounced effect observed for measurement points close to the top of the dam and the upstream dam face. Comparison with eight baseline models validates that the proposed model achieves optimal prediction, fitting performance, and generalization. Consequently, the model emerges as a more suitable choice for the deformation health monitoring of high earth-rock dam projects.

The inverse analysis of the deformation moduli of high arch dams based on displacement monitoring data is essential for structural safety assessment. In traditional inverse analysis methods, the deformation moduli are identified based on the single‐objective optimization and the hydrostatic component derived from the statistical model. This type of method has two main shortcomings: First, it treats the essential multi‐objective optimization problem as a single‐objective problem; second, the extracted hydrostatic component may be biased due to the multicollinearity of variables in the statistical model. This paper presents a methodology for the inverse analysis of the deformation moduli of high arch dams under a multi‐objective optimization strategy. The methodology employs empirical mode decomposition to extract the aging component from displacement monitoring data. Then, thermomechanical analysis is used to reconstruct the remaining hydrostatic and temperature components, thereby avoiding the biases encountered in solving the statistical model. The adaptive polynomial chaos expansion method is embedded in the NSGA‐III algorithm to establish and solve multi‐objective functions in the inverse analysis. Additionally, a composite decision index considering errors and test information is proposed to determine acceptable deformation moduli from the Pareto solution set. A high arch dam is selected to illustrate this methodology with static and dynamic monitoring data. The results show that the identified deformation moduli have errors of 3.8% and 7.2% in displacement and acceleration, respectively. The proposed methodology can yield deformation modulus values that are more consistent with the physical implications than those of the single‐objective optimization method.

Constructing an interpretable model for the long-term deformation Structural Health Monitoring (SHM) of earth-rock dams is of great significance for improving the safety state evaluation and monitoring effect. In this paper, a physics-data-driven model for the deformation SHM of earth-rock dams is proposed based on deep mechanism knowledge distillation. Firstly, the deterministic model is established based on the Finite Element Model (FEM) and outputs the hydraulic load component curve and aging component curve. Then a regression prediction model (HTSGAN) between influencing factors and deformation measurements at multiple measurement points is established based on the Graph Convolutional Network (GCN) and attention mechanism. Finally, the Teacher-Hydraulic-Time-Seepage Graph Attention Networks (T-HTSGAN) model is established based on the feature-based multi-teacher knowledge distillation using the knowledge of hydraulic loading physics and soil-rock creep physics of the FEM for mechanism constraints. The model effectively solves the problems of poor model interpretability and lack of physics knowledge constraints in previous earth-rock dam SHM models. The research results are applied to a project of a 185.5-meter-high concrete-faced rockfill dam, and the predictive performance of the model is more effective and stable through the comparison of six baseline models. The comparative analysis of the component curves proves the effectiveness of the proposed knowledge distillation method for mechanism constraints and improves the interpretability of the neural network model. Therefore, the model is more suitable for engineering applications.

Southwest China is the world's most densely populated area for high dams over 200 m and is also a region with high seismic activity. Earthquakes can significantly alter dam structures, resulting in substantial discrepancies between preearthquake and postearthquake deformation monitoring data. Deformation is a critical indicator of the structural response of dams to internal and external environmental factors. Establishing a dam deformation structural health monitoring (SHM) model promptly after an earthquake is crucial for postearthquake structural health analysis and preventing major accidents. In this paper, we propose a rapid modelling method for postearthquake deformation SHM of high arch dams that uses metalearning and graph attention techniques. First, we develop an SHM model tailored for postseismic small-sample data modelling, integrating a multihead attention mechanism with hydraulic-temporal graph feature fusion. On this basis, we introduce a metalearning framework to derive the initial model parameters from preearthquake data. The proposed model is applied to vertical radial deformation monitoring of the world's only 200-metre-high arch dam subjected to strong near-field earthquakes. The effectiveness of our metalearning framework for postearthquake data is validated by comparing it with the transfer learning framework. Through a comparison with nine baseline models across six postearthquake modelling scenarios, we demonstrate that the proposed model achieves the highest accuracy and exhibits unique engineering applicability for rapid postearthquake modelling tasks. Ablation experiments further confirm the effectiveness of the proposed modules.

Constructing a long-term deformation monitoring model for earth–rock dams that integrates multisource monitoring information is highly important for enhancing the safety state evaluation and monitoring effectiveness of such dams. In this paper, we propose a new health monitoring model named the deformation–seepage–water level multimeasurement point health monitoring (DSW-MPHM) model for earth–rock dams based on deep graph feature fusion. This model fuses coupled seepage, deformation, and water level features from different monitoring sites of the dam body, base, and shoulder. To achieve this goal, we first establish a new module to fuse spatial and temporal features using graph convolutional networks and long short-term memory. Seepage features and water level features are then extracted using graph attention mechanisms. Subsequently, we employ the feature fusion technique, which incorporates principal component analysis and gated fusers, to construct the DSW-MPHM model, which effectively fuses information from multiple sources. This novel approach successfully addresses the issues of information redundancy and the limited reliability of monitoring models. To verify the validity of the model, it is applied to an endoscopic deformation monitoring program of a panel rockfill dam with a height of 185.5 m. The results demonstrate the superior stability and effectiveness of the proposed method compared to those of 10 baseline prediction models. Additionally, the characterization of the seepage and water level features extracted from the model is verified for its reasonableness. Thus, our proposed model is well suited for practical engineering applications.

Accurate prediction of global solar radiation (Rs) is vital for investment decisions and solar energy distribution. In this study, three hybrid models (ACO-SVM, CS-SVM, and GWO-SVM) based on ant colony optimization (ACO), cuckoo search (CS) and grey wolf optimization (GWO) algorithms were proposed to optimize support vector machine (SVM) for predicting Rs in four climate zones of China (temperate continental zone TCZ, mountain plateau zone MPZ, temperate monsoon zone TMZ, and subtropical monsoon zone SMZ). They were compared with the standalone backpropagation neural network model, decision tree, and support vector machines. The results demonstrated that among the standalone models, support vector machines performed best with the highest accuracy in Rs estimation in each climate zone of China, followed by the decision tree and backpropagation neural network models, with a coefficient of determination (R2) in 0.707–0.882, 0.694–0.881, and 0.681–0.850, respectively. In contrast, the hybrid models exhibited higher accuracy than standalone support vector machines in four climatic regions of China, with the coefficient of determination (R2) increasing by 5.361%, 5.476%, 7.382%, and 10.965%, respectively. Among hybrid models, GWO-SVM performed better than CS-SVM, and both had higher accuracy than ACO-SVM, with the coefficient of determination (R2) in 0.809–0.927, 0.804–0.926, and 0.793–0.930, respectively. Therefore, the hybrid models (ACO-SVM, CS-SVM, and GWO-SVM), especially GWO-SVM and CS-SVM, can significantly improve the accuracy for predicting Rs in various regions of China.

The evaluation of the structural safety risk for concrete-faced rockfill dams (CFRDs) is susceptible to the time-varying parameters and the environments of topography, geology, and operations. Traditional evaluation methods, due to their involvement in numerical simulation, cannot meet the current requirements for intelligent monitoring in terms of timeliness and practicality. Therefore, a dynamic evaluation method of reliability in CFRDs is proposed in this paper. Due to the nonlinear mapping relationship between the time-variant reliability and the monitoring characterizations of CFRDs, the long and short-term memory neural network (LSTM) and support vector machine (SVM) models are introduced to construct the potential functional relationships between the reliability sequences and the monitoring characterizations, and the matching between the accuracy of the two models and the frequency of monitoring characterizations has been fully studied, forming an adaptive evaluation model of safety risk for the CFRDs according to the different monitoring frequencies of different monitoring items. The application of Sanbanxi CFRD shows an average relative error of less than 1% and 5% for dam slope stability and slab cracking simulations, respectively. These results demonstrate that the LSTM and SVM models with adaptive monitoring frequency exhibit high fitting and prediction accuracy and applicability, and possess theoretical and engineering application value.

Common anomaly recognition methods are easy to misjudge and miss outliers for the online monitoring data. This is a bottleneck problem that needs to be overcome in dam safety management moving toward informatization. Based on the data of nine hydropower stations along Dadu River Basin, this paper analyzed existing problems of the common anomaly identification method and an algorithm was proposed based on improved M-robust regression recognition. In this algorithm, the AR factor was introduced to avoid the defect that the traditional model cannot simulate random variables. The extreme value method and robust estimation were utilized to avoid the leverage effect. The model collapse caused by maximum measured value was avoided through improving the residual calculation model of M-robust and optimizing the weight distribution function. The maximum of the three values, residual quartile difference, discrete quartile difference, and measurement accuracy, was used as an anomaly recognition criterion to improve the evaluation criteria. The algorithm compiled was used in the Dadu River Company since 2017. The statistics showed that for the 150,000 measured values per day, the evaluation time could be within 15 min, the missed judgment rate was 0%, and the misjudgment rate was less than 2%. The proposed algorithm achieved a great improvement and can meet the needs of online outlier recognition in dam safety management.

The online anomaly recognition of real-time dam safety monitoring data, such as deformation and seepage data from the automatic sensing instruments (e.g., the osmometer and the multi-point displacement meter), has the premise of ensuring data reliability, and it is also one of the core functional modules of online dam safety monitoring. To compensate for the limitation of a single method to identify outliers and further improve the reliability and the rapidity of the anomaly recognition of dam safety monitoring data, a self-matching model based on data-types for online anomaly recognition (SMM) was proposed in this paper. Based on a detailed classification of dam safety monitoring data sequences, this article describes a comparison and analysis of the applicability of a statistical regression model based on the least-squares regression (LSR) model and the online robust recognition and early warning (RREW) model for different datatype sequences. For the single-step-type sequences and normal-type sequences with low fitting accuracy, which could not be completely identified by the two models above, an improved cloud model recognition method based on the diurnal variation rate (ICM) was proposed to compensate for the limitations. Finally, the SMM was determined, that is, the LSR model was used for the multi-point-outlier-type and normal-type sequences with high fitting accuracy, the RREW model method was used for the double-step-type and oscillatory-type sequences, and the ICM method was used for the single-step-type sequences and normal-type sequences with low fitting accuracy. The engineering application of the Dadu River Basin showed that this method effectively solved the problems of low calculation efficiency and a 2% misjudgment rate when using the RREW model alone, and this method greatly improved the accuracy and timeliness of the anomaly recognition of dam safety monitoring data, so it had important theoretical significance and engineering application value.

The monitoring of data anomaly identification is an important basis for dam safety online monitoring and evaluation. In this research, a cluster of anomaly identification models for dam safety monitoring data was constructed, and a three-stage online anomaly identification method was proposed to discriminate outliers. The proposed method combined anomaly detection for measured values based on a single-point time series simulation, measurement error reduction based on remote retesting and spatio-temporal analysis, and environmental response mutation recognition. It brought about efficient and accurate detection for data mutation and online classified identification for its inducement. Additionally, problems such as missing outliers, misjudging normal values induced by the environmental response, and difficulty in online identification for measurement errors were effectively solved. The research productions were applied to the online monitoring system for the safety risk of reservoirs and dams in the Dadu River Basin. The results showed that the proposed method could effectively improve the accuracy of anomaly identification and reduce the misjudgment and omission rate to less than 2%. It could also successfully recognize and subtract nonstructural anomalies such as accidental errors, instrument faults, and environmental responses online, which provided reliable data for online dam safety monitoring.

The safe operation of dams is related to the lifeline of the national economy, the safety of the people, and social stability, and dam safety monitoring plays an essential role in scientifically controlling the safety of dams. Since the effects of environmental variables were not considered in conventional monitoring data repairing methods (such as the single time series model and spatial interpolation model), a spatial model for repairing monitoring data combining the variable importance for projection (VIP) method and cokriging was put forward in this paper. In order to improve the accuracy of the model, the influence of different combinations of covariates on it was discussed, and the VIPj value greater than 0.8 was proposed as the threshold of covariates. The engineering verification shows that the VIP-cokriging spatial model had the advantages of high precision and strong applicability compared with the inverse distance weighting (IDW) model, the ordinary kriging model, and the universal kriging model, and the overall error can be reduced by more than 60%, which could better realize the expansion of the monitoring effect variable to the whole area of the dam space. The engineering application of the PBG dam showed that the model scientifically correlated the existing monitoring points with the spatial location of the dam, and reasonably repaired the measured values of the stopping and abnormal measured points, effectively ensuring that the spatial regular of the monitoring data could truly reflect the actual safety and operational status of the dam.

The hydropower station real-time safety monitoring is not only related to the safety, but also to the generation benefit of the power station. However, as the amount of monitoring data is very huge, and the dam structure is very complex, real-time monitoring cannot be carried out in most of the dam. It is useful to develop a high-performance safety monitoring system to assist the dam safety management work. As such, the paper introduces an expert system for dam safety management with a mode which can carry out the entire process of real-time dam safety monitoring, covering abnormal importing data, abnormal data changes, monitoring work, dam safety. The system has abundant functions such as dam safety analysis, visual query and remote consultation, etc. The system has been successfully applied to more than ten dam projects in China. The application shows that reliable data evaluation can identify data anomalies such as single point jump, multi point outlier and step change. Trend tracking can find possible structural changes. And clustering comprehensive evaluation can give a reasonable evaluation of the safety state of different parts of the dam. Complete online system can provide a supporting decision-making platform for the safety monitoring of the dams and improve the efficiency of power station management.

The hydropower project had comprehensively benefited people from aspects of the economy, society, ecology, etc. The comprehensive benefit is a key indicator for evaluating a project's performance. However, the existing studies only evaluated the comprehensive benefit and ignored the relationship among different benefits, which is of great significance for the sustainable development of a project. Therefore, in the framework of the complex system composed of economic, social, and ecological benefit subsystems, a synergy degree evaluation method is constructed based on the evaluation index system of the comprehensive benefits, and the compound weight is determined by using the non-linear model and the Lagrange function. Thus, the changing rules of the order degree and the synergy degree for the subsystems in different years can be obtained. The proposed method is applied to a gate dam (named SG) to appraise the relationship among benefits. The results show that the economic, social, and ecological benefits of the SG dam from 2011 to 2018 are gradually to be a better state, but the synergy degrees of the complex system belong to "bottom synergy " and "moderate synergy " level, which indicates that there is no close cooperation among the three benefit subsystems.

Anomaly recognition and early warning of monitoring data are of great significance in the field of modern dam safety management. Multidimensional least-squares regression model with the Pauta criterion is a well-known traditional method, but it is easy to misjudge the normal value and miss the outliers. Thereby, an online robust recognition and early warning model combining robust statistics and confidence interval is proposed to detect outliers. The threshold 3 S T + D is set based on the derived confidence interval D and the scale estimator S T (derived from the location M-estimator). Monitoring data obtained from a gravity dam and a rockfill dam were taken as examples to demonstrate the robust recognition and early warning model. The results show that the proposed method can effectively improve the reliability of anomaly recognition and early warnings, which is valuable in engineering applications.

Deformation monitoring is one of the most important means of providing feedback to ensure the safety of projects. Problems plague the existing automatic monitoring system, such as the small monitoring range of monitoring devices, the inadequate field safety protection, and the low accuracy under extreme weather conditions. These problems greatly reduce the real time and reliability of deformation monitoring data and restrict the real-time intelligent control of engineering safety risk. In this paper, a multitype instrument-integrated monitoring system based mainly on the total positioning station (TPS) and supplemented by the Global Navigation Satellite System (GNSS) was promoted with the methods of large field angle, data complementation, environmental perception and judgment, automatic status control, and baseline calibration-meteorological fusion correction. The application results of Pubugou Station show that the averages of mean square error of points (APMSE) for the dam are 0.41∼1.65 mm and the averages of mean square error of height (AHMSE) are 0.42∼0.89 mm. Moreover, the APMSE and AHMSE for the slope are less than 3 mm. The maximum relative error of the TPS and GNSS data compared with the artificial monitoring data is less than 10%. Besides, the system has good overall performance and is of significant comprehensive benefits. The proposed system realizes the all-weather real-time monitoring of deformation and enhances the emergency response capability of special conditions in dams during the operation period.

With the help of GIS, Terra Vista and MultiGen Creator platform, this paper achieves the digitalization of high precision dam, reservoir area and flood. Using DEM technology to simulate flood evolution, it analyzes flood inundation process and possible disaster area, and combines material reserve, traffic situation, and possible refuge path, and uses GIS geographic information to seek the best withdrawal. The path provides a scientific analysis way for the scientific decision of dam disaster response.

. It is a fundamental issue to find a small subset of influential individuals in a complex network such that they can spread information to the largest scope of nodes in the network. Informative functions in complex software network can lead maximum information propagation scope, and mining such function nodes is important to understand the system’s topology structure and information transfer flows. In this paper, a novel top-k informative nodes mining approach based on global information in directed-weighted complex software network is proposed. Firstly, we map functions and call relationships between them as a Directed-Weighted Function Call Network (DWFCN). Secondly, an algorithm path-compute based on Software Executing Path Sequence (SEPS) generated by Depth-First-Search strategy is used to compute the software executing paths and get nodes’ information Accessible Set (AS). Thirdly, algorithm Entropy-Compute is proposed to calculate each source node’s information entropy in each information transfer flow process. Finally, an Informative Node Rank (IN-Rank) algorithm is put forward to mine most informative nodes. Experimental results show that our new method is accurate and effective.