| The failure of both electric cable in a electric system and pipes in a gas or petroleum transportation system can cause disastrous consequence. It is important to monitor the health of electric cable and pipeline in order to prevent accidents and save maintenance expenses. In both applications, joint time-frequency analysis is found to be an effective tool to provide significant information about the health status of electric cable and pipeline.;The Joint Time-Frequency Domain Reflectometry (JTFDR) which is based on time-frequency cross-correlation function is applied to health monitoring of cross-lined polyethylene (XLPE), ethylene propylene rubber (EPR), and silicone rubber (SIR) insulated cables under accelerated aging tests for diagnostics and prognostics. The experimental results presented in this dissertation show that JTFDR has the prognostic potential to predict the remaining life of the cables.;In the second part of this dissertation, a new definition of cross time-frequency distribution is presented and the theory is applied to a corrosion detection problem in order to detect and locate a corrosion damage on an aluminium plate. The experimental results illustrate that the proposed cross time-frequency analysis can detect the corrosion progression more accurately in comparison with three classical methods: damage index, zero-crossing and cross power spectral analysis, which either work in time domain only or in frequency domain only. |
