Adaptive Warning Thresholds for Dam Safety: A KDE-Based Approach
Open Access
- 26 June 2025
- journal article
- research article
- Published by MDPI AG in Infrastructures
- Vol. 10 (7) , 158
Abstract
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.Keywords
Funding Information
- Ministerio de Ciencia e Innovación (PID2021-122661OB-I00, TED2021-129969B-C33, CEX2018-000797-S)
- MCIN/AEI/10.13039/501100011033 (PID2021-122661OB-I00, TED2021-129969B-C33, CEX2018-000797-S)
- MCIN/AEI/10.13039/501100011033 (PID2021-122661OB-I00, TED2021-129969B-C33, CEX2018-000797-S)
This publication has 22 references indexed in Scilit:
- SOLDIER: SOLution for Dam behavior Interpretation and safety Evaluation with boosted Regression treesSoftwareX, 2023
- Dam monitoring data analysis methods: A literature reviewStructural Control and Health Monitoring, 2019
- Dynamic monitoring of a concrete arch dam during the first filling of the reservoirEngineering Structures, 2018
- Integrated Multi-Level Control Value and Variation Trend Early-Warning Approach for Deformation Safety of Arch DamWater Resources Management, 2017
- An empirical comparison of machine learning techniques for dam behaviour modellingStructural Safety, 2015
- Data-Based Models for the Prediction of Dam Behaviour: A Review and Some Methodological ConsiderationsArchives of Computational Methods in Engineering, 2015
- Performance improvement method of support vector machine-based model monitoring dam safetyStructural Control and Health Monitoring, 2015
- Two online dam safety monitoring models based on the process of extracting environmental effectAdvances in Engineering Software, 2013
- Optimal sensor placement for structural health monitoring based on multiple optimization strategiesThe Structural Design of Tall and Special Buildings, 2011
- Multivariate Statistical Process Monitoring of an Industrial Polypropylene Catalyzer Reactor with Component Analysis and Kernel Density EstimationChinese Journal of Chemical Engineering, 2007