Master Thesis project Signal Processing System for acoustic emission from dikes
Context
TNO and GeoDelft are major participants in the IJkdijk (http://nl.wikipedia.org/wiki/IJkdijk) project. By destroying 80 (!) dikes equipped with sensors, sensor technology and scientific models for dikes are developed, leading ultimately to an effective early warning system. This research project will investigate the optimal signal processing configuration of sensors that measure acoustic activities in dikes. For this purpose a dike has been equipped with two microphones that are placed in the ends of a long hollow tube in a dike. Data from this system will be multicast over the internet. UvA will be one of the subscribers. Most, but not all, research will be done at UvA. The student(s) working on this project will be expected to participate in implementing experiments at the test facility, and to collaborate with other researchers in the actual dike breach experiments. As a result, this project will gain insight in the application of (acoustic) sensor technology for the detection of dike failures.
Research Questions
In this master thesis project a signal processing system for dikes is developed that answers the following questions: 1) What observables (and hence signal processing) are useful for the measurement and analysis of acoustic activities in dikes · as a function of external (weather, canal) conditions (e.g. is there more noise between 1000-and1500 Hz if the humidity in the dike is higher) · when the dikes is being destroyed 2) What is the potential of external acoustic sources to determine the condition of the dike (e.g. echoscopy to see fractures / water levels) In addition the following question should be addressed 3) Which design patterns exist for hard and software architectures that process acoustic signals from dikes?

Master Thesis project

Signal Processing System for acoustic emission from dikes

Informatie: Prof. R.J. Meijer/Prof F.C.A. Groen/Drs. R.J.Strijkers (UvA), Dr. A. Koelewijn, Geodelft

0653720517 / rmeijer@science.uva.nl

Context

TNO and GeoDelft are major participants in the IJkdijk (http://nl.wikipedia.org/wiki/IJkdijk) project. By destroying 80 (!) dikes equipped with sensors, sensor technology and scientific models for dikes are developed, leading ultimately to an effective early warning system.

This research project will investigate the optimal signal processing configuration of sensors that measure acoustic activities in dikes. For this purpose a dike has been equipped with two microphones that are placed in the ends of a long hollow tube in a dike. Data from this system will be multicast over the internet. UvA will be one of the subscribers.

Most, but not all, research will be done at UvA. The student(s) working on this project will be expected to participate in implementing experiments at the test facility, and to collaborate with other researchers in the actual dike breach experiments. As a result, this project will gain insight in the application of (acoustic) sensor technology for the detection of dike failures.

Research Questions

In this master thesis project a signal processing system for dikes is developed that answers the following questions:

1) What observables (and hence signal processing) are useful for the measurement and analysis of acoustic activities in dikes

· as a function of external (weather, canal) conditions (e.g. is there more noise between 1000-and1500 Hz if the humidity in the dike is higher)

· when the dikes is being destroyed

2) What is the potential of external acoustic sources to determine the condition of the dike (e.g. echoscopy to see fractures / water levels)

In addition the following question should be addressed

3) Which design patterns exist for hard and software architectures that process acoustic signals from dikes?