Date of Award
Master of Science (MS)
Dr. Elham Makram, Committee Chair
Dr. Anthony Martin
Dr. Keith Corzine
With the increasing amount of overhead lines being converted to underground cables in the distribution system, the need to be able to determine the health of these underground cables becomes imperative. Since the health of underground cables cannot be determined by visual means like overhead lines, an on-line measurement method is needed to determine the health of these cables. By sending a high frequency voltage pulse down the cable and measuring the return pulse, a method called time domain reflectomentry (TDR), an on-line measurement method becomes feasible. One of the main causes of cable failure is known as water-trees, and they are formed through dielectric breakdown of the cables insulation. They are formed from electrical stress at the interface of the cables’ insulation and conductors. To determine an on-line measurement method to detect water trees, an accurate model of water-trees in underground cables is developed. Two different cable types are modeled with water-trees, concentric neutral and tape shield cables. These models are developed in COMSOL Multiphysics®. With this developed water-tree model, it is then integrated into a distribution feeder located along the coast of South Carolina, with parameters provided by Santee Cooper®. To perform TDR and monitor the health of all the three-phase cables in the distribution feeder an optimal pulse generators placement algorithm was used to determine the location of pulse generators to monitor all cables. Finally, an algorithm for monitoring every cable was created and the method tested in PSCAD®. Based on these results an on-line measurement water-tree detection method is presented.
Burkes, Klaehn William, "Water Tree Analysis and On-Line Detection Algorithm Using Time Domain Relectometry" (2014). All Theses. 2521.