Closure to “Discussion of ‘Development of an Internal Camera-Based Volume Determination System for Triaxial Testing’ by S. E. Salazar, A. Barnes, and R. A. Coffman” by A. Mehdizadeh, M. M. Disfani, R. Evans, A. Arulrajah, and D. E. L. Ong

145

Summary The unique structures and foundations of a dam make its safety monitoring a complex task. As the most intuitive effect of dams, deformation contains important information on dam evolution. Actual response has the purpose of diagnosis and early warning compared with model prediction. Given the poor generalization ability of the conventional statistical model, establishing a dam deformation monitoring model is thus essential. The prediction of concrete dam deformation using statistical model and random forest regression (RFR) model is studied. To build an optimized RFR model, the statistical model is used to establish input variables, select the appropriate parameters Mtry and Ntree according to out‐of‐bag error, and extract strong explanatory variables. The model's advantage is that the influence factors can describe concrete dam deformation, and RF can serve as a sensible new data mining tool. The importance of variables for deformation prediction is measured by RF. The RFR method can extract representative influencing factors based on variable importance. The methods are applied to an actual concrete dam. Results indicate that the RFR model can be applied for analysis and prediction of other structural behavior.

Section: Introductionmentioning

confidence: 99%

Statistical model optimized random forest regression model for concrete dam deformation monitoring

Dai

Zhao

et al. 2018

145

“…Therefore, traditional qualitative analysis is a common method useful for dam safety identification. [4,5] Its advantages consist in simplicity and intuitive grasp, whereas the main challenge is that it needs priori information and rich engineering experiences. This prevents engineers from finding unexpected structural response causes.…”

Section: Introductionmentioning

confidence: 99%

“…[7][8][9] So it seems of interest to further investigate the collected time series using data mining technique, in order to detecting exceptions and outliers in time. [5] Actually, uplift pressure is a variable depended on external independent variables (reservoir level, sediment pressure, temperature, etc.). Comparison among multivariate time series can provide valuable information for dam safety monitoring on various time scales.…”

Section: Introductionmentioning

confidence: 99%

Anomaly identification of foundation uplift pressures of gravity dams based on DTW and LOF

2018

Anomaly can provide valuable information for dam safety monitoring. In this paper, a methodology integrating dynamic time warping and local outlier factor to identify anomalies of time series on various time scales is proposed. The main steps of the methodology are introduced in detail. First, measured time series are preprocessed using moving average and normalization, respectively, to eliminate influence of random and amplitude variations. Following this, time series of independent variables (predictors) are selected as templates to calculate dynamic time warping distances using the global constraint to measure similarities with the dependent variable (a predicant) on a large time scale. Then, traditional regression models are built to find out contributions of independent variables to structural behavior, as well as to highlight unwanted behaviors at relatively on a small time scale. Furthermore, local outlier factor values of multivariate are computed, which helps finding anomalies on a small time scale. Finally, causes of anomalies are analyzed comprehensively. A case study is presented focus on the measured foundation uplift beneath the local riverbed blocks of Xixi reservoir dam. Results show that point anomalies occur with sudden changes of independent variables, but contextual anomalies boil down to coupled effect of high reservoir level and low ambient temperature. The natural condition of the rock foundation underlying blocks, which is characterized as intensively open tension joints, is responsible for uplift anomalies.

“…[1] This increase in the amount of available data led to the use of more powerful tools for its analysis, from enhanced versions of the multiple linear regression (e.g., Mata et al [2] ) up to algorithms developed in the field of machine learning, such as neural networks (NN), [3] support vector machines, [4,5] or adaptive neuro-fuzzy inference systems, [6] among others. [7,8] However, these methods are still not widely applied by practitioners, who mostly limit the data analysis to graphical exploration of the time series of data, [9] along with simple statistical models. [1,10] The vast majority of examples of application of advanced tools focus on the development of behaviour models to predict the value of a given response variable of the dam (e.g., radial displacement) as a function of the loads.…”

Section: Introductionmentioning

confidence: 99%

Early detection of anomalies in dam performance: A methodology based on boosted regression trees

Salazar

Toledo

González

et al. 2017

Summary The advances in information and communication technologies led to a general trend towards the availability of more detailed information on dam behaviour. This allows applying advanced data‐based algorithms in its analysis, which has been reflected in an increasing interest in the field. However, most of the related literature is limited to the evaluation of model prediction accuracy, whereas the ulterior objective of data analysis is dam safety assessment. In this work, a machine‐learning algorithm (boosted regression trees) is the core of a methodology for early detection of anomalies. It also includes a criterion to determine whether certain discrepancy between predictions and observations is normal, a procedure to compute a realistic estimate of the model accuracy, and an original approach to identify extraordinary load combinations. The performance of causal and noncausal models is assessed in terms of their ability to detect different types of anomalies, which were artificially introduced on reference time series generated with a numerical model of a 100‐m‐high arch dam. The final approach was implemented in an online application to visualise the results in an intuitive way to support decision making.