Development of support vector regression identification model for prediction of dam structural behaviour

Ranković, Vesna; Grujović, Nenad; Divac, Dejan; Milivojević, Nikola

doi:10.1016/j.strusafe.2014.02.004

Cited by 170 publications

(74 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The mathematical development of the method is complex and beyond the scope of the paper. Detailed 280 and rigorous descriptions can be found in [34] and [35], and a recent application in predicting dam behaviour is reported in [36]. The method uses an ε-insensitive error function that neglects errors below the threshold ε. , [36], [37].…”

Section: Support Vector Machines (Svm)mentioning

confidence: 99%

“…Detailed 280 and rigorous descriptions can be found in [34] and [35], and a recent application in predicting dam behaviour is reported in [36]. The method uses an ε-insensitive error function that neglects errors below the threshold ε. , [36], [37]. A similar criterion was followed in this work, where the library "e1071" within the R environment [38] was used to tune the most important parameters [37] of an SVM model:…”

Section: Support Vector Machines (Svm)mentioning

confidence: 99%

See 1 more Smart Citation

An empirical comparison of machine learning techniques for dam behaviour modelling

et al. 2015

View full text Add to dashboard Cite

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.

show abstract

Section: Support Vector Machines (Svm)mentioning

confidence: 99%

Section: Support Vector Machines (Svm)mentioning

confidence: 99%

An empirical comparison of machine learning techniques for dam behaviour modelling

et al. 2015

View full text Add to dashboard Cite

show abstract

“…Although the results were highly accurate, the computational time was high. Rankovic et al [55] built a behaviour model based on SVM for predicting tangential displacements.…”

Section: Other ML Techniquesmentioning

confidence: 99%

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

Salazar

Morán

Toledo

et al. 2015

Arch Computat Methods Eng

255

166

View full text Add to dashboard Cite

“…Recently, SVM has been utilized to model complex civil engineering problems to provide an extensive understanding of the variables involved in the model [9][10][11][12][13][14][15][16].In the majority of the conducted researches, SVM is utilized to predict specific responses depending on the input variables or to simulate performance of a particular engineering system with special conditions. For example, in hydraulic structure design, SVM has been used to predict the forecasting of the tangential shift of a concrete dam [14], and to predict future dam responses with environmental variables [17].…”

Section: Construction Of Hydraulic Water Retainingmentioning

confidence: 99%

Linked Simulation-Optimization Model for Optimum Hydraulic Design of Water Retaining Structures Constructed on Permeable Soils

Al-Juboori

Datta²

2018

GEOMATE

View full text Add to dashboard Cite

Hydraulic Water Retaining Structures (HWRS), such as dams, weirs and regulators are important projects and necessary for water management. Seepage analysis results under HWRS substantially influences the design of HWRS. One of the biggest challenges in design of HWRS is to determine the accurate seepage characteristics with complex flow conditions, and simultaneously to find the optimum design considering safety and cost. Therefore, this study concentrates on developing a linked simulation-optimization (S-O) model for complex flow conditions. This is achieved via linking the numerical seepage simulation (Geo-Studio/SEEPW) with the Genetic Algorithm (GA) evolutionary optimization solver. Since, a direct linking of numerical model with optimization model is computationally expensive and time consuming, well-trained Support vector machine (SVM) surrogate models are linked to the optimization model instead of a numerical model within the S-O model. The seepage characteristics of optimum design obtained by S-O are evaluated for accuracy by comparing these with the numerical seepage modelling (SEEPW) solutions. The comparison, in general, shows good agreements. Accordingly, the S-O methodology is potentially applicable for providing safe, efficient and economical design of HWRS constructed on a complex seepage flow domain.

show abstract

Development of support vector regression identification model for prediction of dam structural behaviour

Cited by 170 publications

References 30 publications

An empirical comparison of machine learning techniques for dam behaviour modelling

An empirical comparison of machine learning techniques for dam behaviour modelling

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

Linked Simulation-Optimization Model for Optimum Hydraulic Design of Water Retaining Structures Constructed on Permeable Soils

Contact Info

Product

Resources

About