An empirical comparison of machine learning techniques for dam behaviour modelling

Abstract

Predictive models are essential in dam safety assessment. Both deterministic and statistical models applied in the day-to-day practice have demonstrated to be useful, although they show relevant limitations at the same time. On another note, powerful learning algorithms have been developed in the field of machine learning (ML), which have been applied to solve practical problems. The work aims at testing the prediction capability of some state-of-the-art algorithms to model dam behaviour, in terms of displacements and leakage. Models based on random forests (RF), boosted regression trees (BRT), neural networks (NN), support vector machines (SVM) and multivariate adaptive regression splines (MARS) are fitted to predict 14 target variables. Prediction accuracy is compared with the conventional statistical model, which shows poorer performance on average. BRT models stand out as the most accurate overall, followed by NN and RF. It was also verified that the model fit can be improved by removing the records of the first years of dam functioning from the training set.

Data-Based Models for the Prediction of Dam Behaviour: A Review and Some Methodological Considerations

“…The authors found that models with relatively high ARV corresponded with very low M AE, close to the measurement error [58].…”

“…The authors [58] performed a comparative study among various statistical and ML methods, including HST, NN, and others which had been never used before in dam monitoring, such as random forests (RF) or boosted regression trees (BRT). It was reported that innovative ML algorithms offered the most accurate results, although no one performed better for all 14 outputs analysed, which corresponded to radial and tangential displacements and leakage flow in an arch dam.…”

“…The authors performed a similar analysis for 14 instruments in an arch dam [58], and reported that the prediction accuracy was higher in some cases for models trained over the most recent 5 years of data (the maximum training set length was 18 years).…”

“…They have the advantage over the correlation coefficient of being sensitive to differences in the means and variances of observations and predictions, while maintaining the ability to compare models fitted to different data [58].…”

Predicting underground tunnel hazards using machine learning techniques

“…Additionally, in actual practice, approaches like Scheme 3 might be also preferable if a conservative prediction is of interest. (Mirzavand et al 2015), dam behavior (Salazar et al 2015), coal pillar stability (Zhou et al 2015), slope stability (Suman et al 2016), and rock burst hazards (Zhou et al 2016a). 6.2 Selection of parameters and model construction (Goel and Singh 2011) and it is therefore often unavailable at early stages of one project; (ii) actual measurements of tunnel support pressure are unavailable until the support is constructed, and their prior estimations are often unfeasible, as suggested by the large discrepancies found between estimations and measurements in real cases (Bhasin and Grimstad 1996); and (iii) the ranges of variability of rock mass unit weights are often limited, so that they can often be assumed constant in this type of practical empirical prediction (Dwivedi et al 2013). …”