HERI – Research Actions Database

Welcome to the HERI Research Actions Database. The database provides details of past and present research actions related to all aspects of hydropower. The database has been structured following the categorisation developed for the HYDROPOWER EUROPE research and innovation agenda, and Strategic Industry Roadmap.

The database contents have been collated from a variety of organisations working on research actions across all sectors of the hydropower industry. The information has been share freely and is available as open access. You can search the database or add details of new initiatives to the database via the buttons below.

A brief video introduction to HERI:

 

What, why and how we use data ➧



Definition of warning thresholds for dam safety using artificial intelligence and a non-euclidean affinity assessment metric

Description

The AlDA project aimed to improve dam safety through the application of artificial intelligence techniques and the development of a non-Euclidean affinity metric to identify similarities between dams. Recognising the need for more objective methods in dam safety management, the project addressed the limitations of traditional approaches for defining emergency warning thresholds, which typically rely on simplified statistical analyses and expert judgement. By applying advanced AI methods—including neural networks, random forests, Bayesian networks, and other machine learning techniques—the project sought to develop objective methodologies for defining emergency thresholds based on dam monitoring data. In addition, it developed methods to estimate reference behavioural models for newly commissioned dams by analysing data from similar existing dams, thereby improving safety assessments during the critical early stages of operation when historical monitoring data are limited. The project combined expertise in computational modelling from CIMNE with dam engineering expertise from the Universidad Politécnica de Madrid (UPM) to develop innovative tools for risk-informed dam safety assessment and emergency planning.

Organisations

  • International Centre for Numerical Methods in Engineering (CIMNE) - Partner
  • Universidad Politécnica de Madrid - Partner

Funding

  • National funding

Specific Funding Sources:
Ministerio de Economía y Competitividad

Level of Action:

National: SPAIN

Applications

Large-hydro, Storage Hydropower

Keywords

Civil / Structural engineering, Dams, Digitalization, Maintenance, Modeling / Simulation

Areas of Research

Research and Innovation Agenda

  • Development of artificial intelligence-based methods for defining objective emergency warning thresholds and predictive monitoring systems to improve dam safety assessment and maintenance planning.

Strategic Industry Roadmap

Last Updated: 30/07/2026 15:57

AlDA

Start Date
2014
End Date
2016

Budget Range
< €100,000

Project website