Nathalia Silva-Cancino
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Nathalia Silva-Cancino has a total of 12 co-authors. Here are the top co-authors by number of shared publications:
- Fernando Salazar
Centre Internacional de Mètodes Numèrics en Enginyeria (CIMNE), Barcelona, Spain
6 shared publications - Ernest Bladé
Polytechnic University of Catalonia
4 shared publications - Marcos Sanz-Ramos
Universitat Politecnica de Catalunya
4 shared publications - David López-Gómez
Centro de Estudios y Experimentación de Obras Públicas
2 shared publications - Joaquín Irazábal
Centre Internacional de Mètodes Numèrics en Enginyeria (CIMNE), Flumen Research Institute, Universitat Politècnica de Catalunya-BarcelonaTech (UPC), 08034 Barcelona, Spain
2 shared publications - Danial Dehghan-Souraki
Flumen Institute, International Center for Numerical Methods in Engineering (CIMNE)—Universitat Politècnica de Catalunya (UPC), 08034 Barcelona, Spain
1 shared publication
Latest Publications
- Open Access
- 26 June 2025
- Published by MDPI AG in Infrastructures
- Vol. 10 (7) , 158
Adaptive Warning Thresholds for Dam Safety: A KDE-Based Approach
Dams are critical infrastructures that provide essential services such as water supply, hydroelectric power generation, and flood control. As many dams age, the risk of structural failure increases, making safety assurance more urgent than ever. Traditional monitoring systems typically employ predictive models—based on techniques such as the finite element method (FEM) or machine learning (ML)—to compare real-time data against expected performance. However, these models often rely on static warning thresholds, which fail to reflect the dynamic conditions affecting dam behavior, including fluctuating water levels, temperature variations, and extreme weather events. This study introduces an adaptive warning threshold methodology for dam safety based on kernel density estimation (KDE). The approach incorporates a boosted regression tree (BRT) model for predictive analysis, identifying influential variables such as reservoir levels and ambient temperatures. KDE is then used to estimate the density of historical data, allowing for dynamic calibration of warning thresholds. In regions of low data density—where prediction uncertainty is higher—the thresholds are widened to reduce false alarms, while in high-density regions, stricter thresholds are maintained to preserve sensitivity. The methodology was validated using data from an arch dam, demonstrating improved anomaly detection capabilities. It successfully reduced false positives in data-sparse conditions while maintaining high sensitivity to true anomalies in denser data regions. These results confirm that the proposed methodology successfully meets the goals of enhancing reliability and adaptability in dam safety monitoring. This adaptive framework offers a robust enhancement to dam safety monitoring systems, enabling more reliable detection of structural issues under variable operating conditions.
- 26 June 2025
- Published by MDPI AG in Infrastructures
- Vol. 10 (7) , 158
Open AccessAdaptive Warning Thresholds for Dam Safety: A KDE-Based Approach
Dams are critical infrastructures that provide essential services such as water supply, hydroelectric power generation, and flood control. As many dams age, the risk of structural failure increases, making safety assurance more urgent than ever. Traditional monitoring systems typically employ predictive models—based on techniques such as the finite element method (FEM) or machine learning (ML)—to compare real-time data against expected performance. However, these models often rely on static warning thresholds, which fail to reflect the dynamic conditions affecting dam behavior, including fluctuating water levels, temperature variations, and extreme weather events. This study introduces an adaptive warning threshold methodology for dam safety based on kernel density estimation (KDE). The approach incorporates a boosted regression tree (BRT) model for predictive analysis, identifying influential variables such as reservoir levels and ambient temperatures. KDE is then used to estimate the density of historical data, allowing for dynamic calibration of warning thresholds. In regions of low data density—where prediction uncertainty is higher—the thresholds are widened to reduce false alarms, while in high-density regions, stricter thresholds are maintained to preserve sensitivity. The methodology was validated using data from an arch dam, demonstrating improved anomaly detection capabilities. It successfully reduced false positives in data-sparse conditions while maintaining high sensitivity to true anomalies in denser data regions. These results confirm that the proposed methodology successfully meets the goals of enhancing reliability and adaptability in dam safety monitoring. This adaptive framework offers a robust enhancement to dam safety monitoring systems, enabling more reliable detection of structural issues under variable operating conditions.
- Open Access
- 10 March 2025
- Published by Universitat Politècnica de Catalunya
Evaluation of potential hazard due to off-stream reservoir failure using machine learning techniques
(English) Hazard classification for dams and off-stream reservoirs, which entails identifying potential damages in the event of structural failure, is a crucial tool for implementing local-level risk reduction plans. Consequently, national administrations have developed guidelines including suggested methodologies and tools. The modification of the Spanish Regulation of the Public Hydraulic Domain (Royal Decree 849/1986, of April 11), carried out through Royal Decree 9/2008, of January 11, obliges owners of off-stream reservoirs with a height of 5 meters or a capacity higher than 100,000 m3, whether public or private, to develop a classification study based on the potential risk of their failure (Articles 356 and 367). This represented a significant paradigm shift, under the former regulation, such a study was only required for dams exceeding 15 m in height or those with a height between 10 and 15 m and a capacity exceeding 100,000 m3. The procedure for this classification is time and resource-consuming, and in the specific case of owners of off-stream reservoirs, they may not have these assets. Therefore, this research proposes a simplified methodology to classify off-stream reservoirs, utilizing a surrogate Machine Learning (ML) model that is simpler and has a lower computational cost than conventional approaches. Additionally, the influence of two sources of uncertainties on hazard classification is analysed. This research is based on the generation of synthetic data. A specialized tool in Iber was developed to generate massive 2D hydraulic models of synthetic off-stream reservoir failures, which make all the processes of construction, calculation and extraction of results automatic. The first analysis was focused on the effect of selecting a breach parametric model on the hydraulic variables, the potential damages, and the hazard classification of the structures. Three common parametric models were compared, using a set of synthetic cases and a real off-stream reservoir. The results highlighted that the choice of the model has significant effects. Notably, the erodibility of the material exerts a high influence, surpassing that of the failure mode. The use of an inappropriate model or a lack of information regarding dike material can lead to overly conservative or underestimated results, consequently affecting hazard classification. The ML model constructed for the simplified methodology was a Random Forest classifier capable of identifying potential damages at any point in the vicinity of an off-stream reservoir. This ML model was trained using synthetic data, offering an estimation of potential damages by considering the physical characteristics of the structure, the surrounding terrain, and vulnerable areas. During a real case application, the simplified methodology achieved an accuracy rate of 91%. The simplified methodology allows owners and administration to obtain a pre-classification without the need to make a 2D hydraulic model, which saves time and money. Furthermore, an interface called ACROPOLIS was developed, integrating the ML model. Users can apply the simplified methodology through ACROPOLIS, which guides them step by step, providing the overall classification of the off-stream reservoir based on Spanish regulations. Finally, the analysis of uncertainties related to breach formation and the location of the breaking points in reservoirs was integrated. This involved comparing the current deterministic approach for hazard classification and a newly proposed fourth-step probabilistic approach that accounts for uncertainties in constructing hazard maps. The study revealed variations in classifications between scenarios, as different breaking points and breach formations generate diverse classifications that can affect emergency plans. Additionally, the proposed visualization can be used for various purposes, including tracking the evolution of categorization over time due to land use changes. (Català) La classificació de preses i basses en funció del risc potencial implica la identificació de possibles danys en cas de fallida estructural, i és una eina crucial per a la implementació de plans d'emergència per a la reducció de riscos a nivell local. En conseqüència, les administracions nacionals han desenvolupat directrius que inclouen metodologies i eines suggerides per a dur a terme aquesta anàlisi. Amb la modificació del Reglament del Domini Públic Hidràulic espanyol (Reial decret 849/1986, de l'11 d'abril), realitzada a través del Reial decret 9/2008, de l'11 de gener, obliga els propietaris de basses amb una altura de 5 metres o un volum superior a 100.000 m³, siguin públics o privats, a desenvolupar un estudi de classificació basat en el risc potencial (Articles 356 i 367). Això va representar un canvi de paradigma significatiu, ja que sota la regulació anterior, aquest estudi només era requerit per a preses que excedissin els 15 m d'altura o aquelles amb una altura entre 10 i 15 m i una capacitat superior a 100.000 m³. Realitzar un estudi de classificació requereix una alta inversió de temps i recursos econòmics, els quals, en molts casos, els propietaris de basses poden no tenir disponibles. Per tant, aquesta investigació proposa una metodologia simplificada per classificar basses, utilitzant un model de Machine Learning, que és més senzill i té un cost computacional menor que els mètodes convencionals. A més, es presenta una anàlisi de la influència de dues fonts d'incertesa en la classificació de les basses. Aquesta investigació es basa en dades sintètiques. Es va desenvolupar una eina especialitzada en Iber que permet generar models hidràulics massius en 2D de falles de basses amb geometria sintètica, automatitzant tots els processos de construcció, càlcul i...
- 10 March 2025
- Published by Universitat Politècnica de Catalunya
Open AccessEvaluation of potential hazard due to off-stream reservoir failure using machine learning techniques
(English) Hazard classification for dams and off-stream reservoirs, which entails identifying potential damages in the event of structural failure, is a crucial tool for implementing local-level risk reduction plans. Consequently, national administrations have developed guidelines including suggested methodologies and tools. The modification of the Spanish Regulation of the Public Hydraulic Domain (Royal Decree 849/1986, of April 11), carried out through Royal Decree 9/2008, of January 11, obliges owners of off-stream reservoirs with a height of 5 meters or a capacity higher than 100,000 m3, whether public or private, to develop a classification study based on the potential risk of their failure (Articles 356 and 367). This represented a significant paradigm shift, under the former regulation, such a study was only required for dams exceeding 15 m in height or those with a height between 10 and 15 m and a capacity exceeding 100,000 m3. The procedure for this classification is time and resource-consuming, and in the specific case of owners of off-stream reservoirs, they may not have these assets. Therefore, this research proposes a simplified methodology to classify off-stream reservoirs, utilizing a surrogate Machine Learning (ML) model that is simpler and has a lower computational cost than conventional approaches. Additionally, the influence of two sources of uncertainties on hazard classification is analysed. This research is based on the generation of synthetic data. A specialized tool in Iber was developed to generate massive 2D hydraulic models of synthetic off-stream reservoir failures, which make all the processes of construction, calculation and extraction of results automatic. The first analysis was focused on the effect of selecting a breach parametric model on the hydraulic variables, the potential damages, and the hazard classification of the structures. Three common parametric models were compared, using a set of synthetic cases and a real off-stream reservoir. The results highlighted that the choice of the model has significant effects. Notably, the erodibility of the material exerts a high influence, surpassing that of the failure mode. The use of an inappropriate model or a lack of information regarding dike material can lead to overly conservative or underestimated results, consequently affecting hazard classification. The ML model constructed for the simplified methodology was a Random Forest classifier capable of identifying potential damages at any point in the vicinity of an off-stream reservoir. This ML model was trained using synthetic data, offering an estimation of potential damages by considering the physical characteristics of the structure, the surrounding terrain, and vulnerable areas. During a real case application, the simplified methodology achieved an accuracy rate of 91%. The simplified methodology allows owners and administration to obtain a pre-classification without the need to make a 2D hydraulic model, which saves time and money. Furthermore, an interface called ACROPOLIS was developed, integrating the ML model. Users can apply the simplified methodology through ACROPOLIS, which guides them step by step, providing the overall classification of the off-stream reservoir based on Spanish regulations. Finally, the analysis of uncertainties related to breach formation and the location of the breaking points in reservoirs was integrated. This involved comparing the current deterministic approach for hazard classification and a newly proposed fourth-step probabilistic approach that accounts for uncertainties in constructing hazard maps. The study revealed variations in classifications between scenarios, as different breaking points and breach formations generate diverse classifications that can affect emergency plans. Additionally, the proposed visualization can be used for various purposes, including tracking the evolution of categorization over time due to land use changes. (Català) La classificació de preses i basses en funció del risc potencial implica la identificació de possibles danys en cas de fallida estructural, i és una eina crucial per a la implementació de plans d'emergència per a la reducció de riscos a nivell local. En conseqüència, les administracions nacionals han desenvolupat directrius que inclouen metodologies i eines suggerides per a dur a terme aquesta anàlisi. Amb la modificació del Reglament del Domini Públic Hidràulic espanyol (Reial decret 849/1986, de l'11 d'abril), realitzada a través del Reial decret 9/2008, de l'11 de gener, obliga els propietaris de basses amb una altura de 5 metres o un volum superior a 100.000 m³, siguin públics o privats, a desenvolupar un estudi de classificació basat en el risc potencial (Articles 356 i 367). Això va representar un canvi de paradigma significatiu, ja que sota la regulació anterior, aquest estudi només era requerit per a preses que excedissin els 15 m d'altura o aquelles amb una altura entre 10 i 15 m i una capacitat superior a 100.000 m³. Realitzar un estudi de classificació requereix una alta inversió de temps i recursos econòmics, els quals, en molts casos, els propietaris de basses poden no tenir disponibles. Per tant, aquesta investigació proposa una metodologia simplificada per classificar basses, utilitzant un model de Machine Learning, que és més senzill i té un cost computacional menor que els mètodes convencionals. A més, es presenta una anàlisi de la influència de dues fonts d'incertesa en la classificació de les basses. Aquesta investigació es basa en dades sintètiques. Es va desenvolupar una eina especialitzada en Iber que permet generar models hidràulics massius en 2D de falles de basses amb geometria sintètica, automatitzant tots els processos de construcció, càlcul i...
- Open Access
- 7 February 2025
- Published by Springer Nature in Journal of Civil Structural Health Monitoring
- Vol. 15 (6) , 1771-1792
Detection of outliers in dam monitoring time series with autoencoders
Dam monitoring is crucial for behavior analysis and safety assessment. The data recorded by monitoring systems are the basis of behavioral models. Ensuring the quality of these data is vital for making informed decisions and improving prediction accuracy. However, monitoring data often contain errors that need to be corrected before use. As data acquisition systems become increasingly automated, the resulting large databases present challenges that conventional methods cannot effectively address. This work proposes a methodology based on autoencoders for the automatic detection of outliers in dam monitoring data. The model is calibrated with a general procedure, which showed to be effective in all situations considered. The application of this method to data series from deformations, piezometers, joint openings and seepage flow in two different dams demonstrates its ability to detect all strong outliers without false positives. Only a few potential outliers, some of which are hard to classify after an exploratory analysis, were overlooked by the model. While the final decision on record validation should be made by experienced technicians, this approach can effectively screen large databases of measurements, aiding in the efficient identification of outliers.
- 7 February 2025
- Published by Springer Nature in Journal of Civil Structural Health Monitoring
- Vol. 15 (6) , 1771-1792
Open AccessDetection of outliers in dam monitoring time series with autoencoders
Dam monitoring is crucial for behavior analysis and safety assessment. The data recorded by monitoring systems are the basis of behavioral models. Ensuring the quality of these data is vital for making informed decisions and improving prediction accuracy. However, monitoring data often contain errors that need to be corrected before use. As data acquisition systems become increasingly automated, the resulting large databases present challenges that conventional methods cannot effectively address. This work proposes a methodology based on autoencoders for the automatic detection of outliers in dam monitoring data. The model is calibrated with a general procedure, which showed to be effective in all situations considered. The application of this method to data series from deformations, piezometers, joint openings and seepage flow in two different dams demonstrates its ability to detect all strong outliers without false positives. Only a few potential outliers, some of which are hard to classify after an exploratory analysis, were overlooked by the model. While the final decision on record validation should be made by experienced technicians, this approach can effectively screen large databases of measurements, aiding in the efficient identification of outliers.
- Open Access
- 6 November 2024
- Published by Wiley in Journal of Flood Risk Management
- Vol. 18 (1)
Effect of uncertainties in breach location and breach mechanisms on risk‐related classification of off‐stream reservoirs
Off‐stream reservoirs are artificial water storage structures that increase the flood risk of an area. In some places, related risk reduction plans are based on a risk classification of these structures, which follows local water resource management regulations. These classification methods typically follow deterministic qualitative guidelines that do not account for uncertainties. This study introduces a fourth‐step probabilistic approach that accounts for uncertainties related to simultaneous breach formation and breaking point location of off‐stream reservoirs, and proposes an alternative visualisation for their classification. The methodology is applied to a set of Spanish off‐stream reservoirs that are classified according to the Spanish normative. Results show that different breaking points and breach formations generate diverse classifications that can affect risk reduction plans. Additionally, we demonstrate that the proposed visualisation can be used for various purposes, including the case of the evolution of the categorisation in time, due to land use changes, which could be used by decision‐makers to understand which off‐stream reservoir requires a category update. These findings introduce a novel approach to managing uncertainties, which is crucial for developing resilient flood management strategies and contributes to the innovation discourse in flood risk management.
- 6 November 2024
- Published by Wiley in Journal of Flood Risk Management
- Vol. 18 (1)
Open AccessEffect of uncertainties in breach location and breach mechanisms on risk‐related classification of off‐stream reservoirs
Off‐stream reservoirs are artificial water storage structures that increase the flood risk of an area. In some places, related risk reduction plans are based on a risk classification of these structures, which follows local water resource management regulations. These classification methods typically follow deterministic qualitative guidelines that do not account for uncertainties. This study introduces a fourth‐step probabilistic approach that accounts for uncertainties related to simultaneous breach formation and breaking point location of off‐stream reservoirs, and proposes an alternative visualisation for their classification. The methodology is applied to a set of Spanish off‐stream reservoirs that are classified according to the Spanish normative. Results show that different breaking points and breach formations generate diverse classifications that can affect risk reduction plans. Additionally, we demonstrate that the proposed visualisation can be used for various purposes, including the case of the evolution of the categorisation in time, due to land use changes, which could be used by decision‐makers to understand which off‐stream reservoir requires a category update. These findings introduce a novel approach to managing uncertainties, which is crucial for developing resilient flood management strategies and contributes to the innovation discourse in flood risk management.
- 31 August 2024
- Published by Springer Nature
- p. 289-300
Flood Maps Definition for Off-Stream Reservoir Failure: Deterministic Versus Probabilistic Approach
- 31 August 2024
- Published by Springer Nature
- p. 289-300
Flood Maps Definition for Off-Stream Reservoir Failure: Deterministic Versus Probabilistic Approach
- Open Access
- 31 January 2024
- Published by Universitat Politecnica de Valencia in Ingeniería del Agua
- Vol. 28 (1) , 47-63
Avances en Iber para la clasificación de balsas: proyecto ACROPOLIS
Se estima que en España hay alrededor de 70.000 balsas. Según la legislación vigente, usuarios y administraciones tienen la obligación de clasificarlas en función del riesgo potencial frente una eventual rotura si tienen más de 100.000 m3 o el dique mide más de 5 m. Con el objetivo de construir un sistema de ayuda para su clasificación que facilite y agilice este proceso, se ha llevado a cabo el proyecto ACROPOLIS. A tal fin, la herramienta de modelización numérica Iber fue elemento esencial y de carácter transversal. Por un lado, el empleo de Iber permitió desarrollar un método simplificado para pre-clasificar balsas en base a técnicas de machine learning y, por otro lado, se desarrollaron metodologías ad hoc en Iber que agilizan y automatizan los procesos, como la extracción automática de información hidráulica y de peligrosidad en puntos vulnerables y la paralelización del código para el cálculo mediante técnicas de computación de alto rendimiento.
- 31 January 2024
- Published by Universitat Politecnica de Valencia in Ingeniería del Agua
- Vol. 28 (1) , 47-63
Open AccessAvances en Iber para la clasificación de balsas: proyecto ACROPOLIS
Se estima que en España hay alrededor de 70.000 balsas. Según la legislación vigente, usuarios y administraciones tienen la obligación de clasificarlas en función del riesgo potencial frente una eventual rotura si tienen más de 100.000 m3 o el dique mide más de 5 m. Con el objetivo de construir un sistema de ayuda para su clasificación que facilite y agilice este proceso, se ha llevado a cabo el proyecto ACROPOLIS. A tal fin, la herramienta de modelización numérica Iber fue elemento esencial y de carácter transversal. Por un lado, el empleo de Iber permitió desarrollar un método simplificado para pre-clasificar balsas en base a técnicas de machine learning y, por otro lado, se desarrollaron metodologías ad hoc en Iber que agilizan y automatizan los procesos, como la extracción automática de información hidráulica y de peligrosidad en puntos vulnerables y la paralelización del código para el cálculo mediante técnicas de computación de alto rendimiento.
- Open Access
A Probabilistic Approach for Off-Stream Reservoir Failure Flood Hazard Assessment
Off-stream reservoirs are hydraulic structures that might cause severe flood damages in case of failure or improper operation. Their safety regulations usually require hydraulic studies for flood hazard and inundation zone mapping. The selection of the break point is not trivial because the topography in its surroundings is commonly highly anthropic. A wrong selection would not provide the worst scenario in terms of maximum flood hazard extent. This work presents a probabilistic approach based on a stochastic definition of the break point along the dyke. A number of failure scenarios are generated automatically, corresponding to different breach formations. Then, an in-cascade calculation process simulates each scenario, providing a framework to carry out statistical analysis. The simulation of the breach formation and the flood wave propagation is performed through a GPU parallelised two-dimensional hydraulic numerical model, which provides a probabilistic inundation zone and flood hazard mapping of all scenarios simulated in a suitable timeframe.
Open AccessA Probabilistic Approach for Off-Stream Reservoir Failure Flood Hazard Assessment
Off-stream reservoirs are hydraulic structures that might cause severe flood damages in case of failure or improper operation. Their safety regulations usually require hydraulic studies for flood hazard and inundation zone mapping. The selection of the break point is not trivial because the topography in its surroundings is commonly highly anthropic. A wrong selection would not provide the worst scenario in terms of maximum flood hazard extent. This work presents a probabilistic approach based on a stochastic definition of the break point along the dyke. A number of failure scenarios are generated automatically, corresponding to different breach formations. Then, an in-cascade calculation process simulates each scenario, providing a framework to carry out statistical analysis. The simulation of the breach formation and the flood wave propagation is performed through a GPU parallelised two-dimensional hydraulic numerical model, which provides a probabilistic inundation zone and flood hazard mapping of all scenarios simulated in a suitable timeframe.
- Open Access
A Machine Learning-Based Surrogate Model for the Identification of Risk Zones Due to Off-Stream Reservoir Failure
Approximately 70,000 Spanish off-stream reservoirs, many of them irrigation ponds, need to be evaluated in terms of their potential hazard to comply with the new national Regulation of the Hydraulic Public Domain. This requires a great engineering effort to evaluate different scenarios with two-dimensional hydraulic models, for which many owners lack the necessary resources. This work presents a simplified methodology based on machine learning to identify risk zones at any point in the vicinity of an off-stream reservoir without the need to elaborate and run full two-dimensional hydraulic models. A predictive model based on random forest was created from datasets including the results of synthetic cases computed with an automatic tool based on the two-dimensional numerical software Iber. Once fitted, the model provided an estimate on the potential hazard considering the physical characteristics of the structure, the surrounding terrain and the vulnerable locations. Two approaches were compared for balancing the dataset: the synthetic minority oversampling and the random undersampling. Results from the random forest model adjusted with the random undersampling technique showed to be useful for the estimation of risk zones. On a real application test the simplified method achieved 91% accuracy.
Open AccessA Machine Learning-Based Surrogate Model for the Identification of Risk Zones Due to Off-Stream Reservoir Failure
Approximately 70,000 Spanish off-stream reservoirs, many of them irrigation ponds, need to be evaluated in terms of their potential hazard to comply with the new national Regulation of the Hydraulic Public Domain. This requires a great engineering effort to evaluate different scenarios with two-dimensional hydraulic models, for which many owners lack the necessary resources. This work presents a simplified methodology based on machine learning to identify risk zones at any point in the vicinity of an off-stream reservoir without the need to elaborate and run full two-dimensional hydraulic models. A predictive model based on random forest was created from datasets including the results of synthetic cases computed with an automatic tool based on the two-dimensional numerical software Iber. Once fitted, the model provided an estimate on the potential hazard considering the physical characteristics of the structure, the surrounding terrain and the vulnerable locations. Two approaches were compared for balancing the dataset: the synthetic minority oversampling and the random undersampling. Results from the random forest model adjusted with the random undersampling technique showed to be useful for the estimation of risk zones. On a real application test the simplified method achieved 91% accuracy.