Loading...
Please wait, while we are loading the content...
Similar Documents
The Electrical Performance of Thermoplastic Polymers When Used As Insulation in Cables
| Content Provider | Semantic Scholar |
|---|---|
| Author | Qin, Siyi Timoshkin, Igor V. Wilson, Mark P. Gregor, Scott J. Mac Maclean, Matthew Anderson, Joanne Wang, Tzu-Kai Pilgrim, James A. Lewin, P. L. Haddad, Abdolhosein Elnaddab, Khalifa Griffiths, Hugh Wallace, Philip R. Nekahi, Azam McMeekin, Scott G. |
| Copyright Year | 2013 |
| Abstract | No. 1 Classification of PD patterns from single void and two void arrangements in poly-ethylene-terephthalate A. Abubakar Mas’ud, B. G. Stewart, S.G.McMeekin and A. Nesbitt School of Engineering and Built Environment, Glasgow Caledonian University, Glasgow, UK. E-mail: abdullah.abubakar@gcal.ac.uk Partial discharge (PD) measurements are commonly applied in monitoring the degradation of dielectric insulation subjected to high voltage (HV) stress. One important PD defect of HV insulation system is the void. This is because of its unpredictable behaviour [1]. The number of discharges and the PD amplitude in a void can sometimes increase or decrease and in some cases even disappear. Over the years, a number of researchers have analysed PD activity within voids [2] but few have analysed voids in poly-ethyleneterephthalate (PET) [3]. HV insulation may contain more than one void at any localised position. Thus it is important to study the discharge patterns emanating from single and two void arrangements in PET and compare them to observe whether there exist variations in the PD patterns and statistics over longer degradation periods. To this end the aim of the work presented in this paper is as follows: 1. To evaluate and compare the statistical parameters of the φ-q-n (phase-amplitude-number) patterns for different single and two void arrangements within PET insulations. To achieve this, experiments were carried out and φ-q-n patterns established for the different voids (see example Figure 1) enabling statistical parameters to be captured over long test periods i.e. when the degradation of the insulation initiates. The voids considered have. 0.6mm diameters. 2. To apply an ensemble neural network (ENN) to classify and capture any statistical distinctions existing among the void arrangements. The results of the ENN are compared with the widely applied single neural network (SNN). The results demonstrate that both the ENN and SNN are capable of distinguishing between the three different void arrangements, with the ENN always providing an improved performance over the SNN. |
| File Format | PDF HTM / HTML |
| Alternate Webpage(s) | http://www.uhvnet.org.uk/downloads/Abstract%20booklet%206th%20UHVnet%20Colloquium.pdf |
| Language | English |
| Access Restriction | Open |
| Content Type | Text |
| Resource Type | Article |
