Malfunctions of monitoring system and lack of systematic feedback analysis from monitoring data are common problems for dam safety management in Taiwan. On the other hand, there is a great demand for effective non-destructive methods to examine the interior of the reservoir structures. The objectives of this 3-year study are to develop improved monitoring techniques, evaluate and improve non-destructive testing methods, and eventually establish a more effective diagnosis system for dam safety.
In developing improved monitoring techniques, this year focused on improvement and durability test of Fiber Bragg Grating (FBG) piezometer and the development of standard procedure and quantitative approach of localized shear deformation monitoring by Time domain Reflectometry (TDR). A data acquisition system based on Field Programmable Gate Array (FPGA) was also developed. TDR physical model that mimics the field condition was developed to study the effect of soil-grout-cable interaction. Furthermore, a signal processing technique (wavelets analysis) was proposed to achieve early detection of shear displacement. Installation and data reduction procedure were recommended from experiemtal results. A TDR data acquisition system capable of interrogating multiple physical parameters based on TDR and integrating conventional monitoring techniques was proposed.
The seismic surface wave method is the focus for this year’s study on non-destructive testing methods. A new survey method, named High Lateral Resolution Surface Wave Method, was proposed to overcome the dilemma in selecting field parameters. The effects of topography and higher modes on surface wave testing were also discussed and some possible countermeasures were proposed. Furthermore, the guideline for applying non-destructive testing methods in dam safety evaluation was outlined and partly drafted.
The system identification method for safety diagnosis of concrete dames based on seismograph data was applied to gravity dams. Due to the lack of seismograph data at the foundation level, the feasibility of the system identification method in absence of input excitation data was validated by using the recorded earthquake time history data of Ku-Kuan and Tien-Loon dams provided by Taiwan Power Company. Unlike arch dams, the vibration frequency of the gravity dam does not have a single peak value, implying the system identification approach is less applicable to gravity dams.