Interpretation technique of partial discharge pulse count and surface defect analysis for evaluation of composite materials

Abstract

High voltage insulation must be designed in such a way that it is very resistant to ageing including that from partial discharge (PD). Many studies were previously carried out on composites based on low density polyethylene (LDPE). However, the use of natural rubber (NR) and nanosilica (SiO2 nanoparticle) in the LDPE-based composites are relatively new. Furthermore, the PD resistant performance of the composites is yet to be extensively researched. It is desired to know the weight percentage of each component in the LDPE-NR-SiO2 composite, especially the nanosilica filler weight percentage, for an optimum PD resistant performance. Due to specific research requirements, a customized laboratory scale PD data acquisition system (PD-DAQS) is desired to be developed. With the availability of several parameters obtained from the PD experiment, there is another need to devise an interpretation tool that can correctly correlate all parameters to the PD resistance. This work aims to develop a new PD system for measuring, analysing and interpreting PD signals, which can then be used to determine a new nanocomposite material with high PD resistance. A new PD-DAQS comprising CIGRE Method II test cell, Picoscope interfacing device, and LabVIEW based program was successfully developed. Scanning electron as well as normal microscopes were used for composite sample’s image capturing and morphological analyses. A new PD interpretation technique based on PD pulse count and its surface image analysis was obtained and successfully categorised the PD performance of a given dielectric sample on scores of 1 to 5 corresponding to very bad to very good. PD improvement index and a scoring system were also introduced and utilised for the new dielectric work. Three groups of new composites with varying compositions and nanosilica content were made and tested by applying high voltage stress using the CIGRE Method II test cell for 60 minutes. Results have shown that the addition of nanosilica filler into LDPE and LDPE-NR base polymers have increased the PD resistance of the new composites. The highest PD resistance score of 5 was achieved by the best nanocomposite sample of LDPE-SiO2 with 4.5 weight percent of nanosilica. Even though the addition of natural rubber to LDPE matrix has caused a decrease in PD resistance, the addition of 6 weight percent of nanosilica as fillers in the LDPE-NR (80:20) composite has tremendously improved its PD resistant performance. The LDPE as well as LDPE-NR based nanocomposites can be potentially developed as a high voltage insulating material. The developed PD measuring and interpretation system can also be utilised for future PD and nanodielectric studies.