This research Project is enshrined in a new research line with the final objective of defining criteria for the usage of intelligent systems as a support tool in dam safety analysis. The general objective of the project is to establish a new methodology to analyze dam safety and potential risk, using analysis techniques based on Neural Networks.

 
Figure 1. La Baells dam

These are the specific objectives:

1. Obtain conclusions about the benefits of applying new technologies in obtaining, transmitting and processing monitoring data, in order to improve the knowledge of dam safety in real time.
2. Set criteria about the range and amount of data needed by the analysis to reach its objectives.
3. Identify advantages and disadvantages of using neural networks instead of traditional analysis techniques.
4. Know the predictive capacity of the developed models depending on whether its training has been done with data obtained from numerical simulation or from dam monitoring.
5. Obtain conclusions and set criteria for indentifying dambreak scenarios, safety thresholds and risk assessment.

The developments of the Project will be applied to the Baells dam, as a test case. It's an arch dam of 102 m height and 300 m of crown length. This dam has been selected because of the large amount of monitoring data available, since the first filling up of the reservoir until nowadays.

CIMNE is developing a mathematical model for both the dam and the foundation to be used in the calculation of the response to different situations.