A pilot study of ASTRA revealed significant uncertainties in the assessment of the state of cantilever retaining walls. Both non-destructive and destructive techniques could not provide satisfactory results. Therefore, research is required on how to monitor these walls, especially since associated costs of repair measures and other consequences are very high. The present proposal develops the basis for an automated measuring system and demonstrates its functionality on a pilot wall.
First, a system based on fiber-optic strain sensors is designed and tested. Fiber-optic sensors offer reliability, accuracy and minimal space requirements. The focus is on the sensor integration into the structure: bonding to the structure and placement within the structure. Laboratory tests shall prove the bonding. The placement shall be optimized by studying the deformation and failure modus of the walls.
Second, a pilot wall is instrumented including temperature and piezometer sensors and an automated monitoring system will be operated over two years.
Large-scale tests of an ETH research project will be instrumented in parallel with the same system as the pilot wall. The laboratory results will help to understand values from daily and seasonal fluctuation in comparison to extreme values recorded at failure of the wall.
Finally, recommendations for further development of the system as well as an assessment of the cost effectiveness is provided. The ultimate goal is an automated alarm system.