Partial Discharges: Experimental Investigation, Model Development, and Data Analytics
| dc.contributor.author | Razavi Borghei, Seyyed Moein | en |
| dc.contributor.committeechair | Ghassemi, Mona | en |
| dc.contributor.committeemember | Centeno, Virgilio A. | en |
| dc.contributor.committeemember | Min, Chang Woo | en |
| dc.contributor.committeemember | Safaai-Jazi, Ahmad | en |
| dc.contributor.committeemember | Bansal, Manish | en |
| dc.contributor.department | Electrical Engineering | en |
| dc.date.accessioned | 2022-02-12T09:00:08Z | en |
| dc.date.available | 2022-02-12T09:00:08Z | en |
| dc.date.issued | 2022-02-11 | en |
| dc.description.abstract | Insulation system is an inseparable part of electrical equipment. In this study, one of the most important aging factors in insulation systems known as partial discharge (PD) is targeted. PD phenomenon has been studied for more than a century and yet new technologies still demand the investigation of PD impact. Nowadays, electrification is penetrating into various fossil-fuel-based industries such as transportation system that demands the reliability of electrical equipment under various harsh environmental conditions. Due to the lack of knowledge on the behavior of insulation systems, research in this area is intensively needed. The current study probes into the partial discharge phenomenon from two aspects and the groundwork for both aspects are provided by experimentation of multiple PD types. In the first goal, a finite-element analysis (FEA) approach is developed based on measurement data to estimate electric field distribution. The FEA model is coupled with a programming scheme to evaluate PD conditions, calculate PD metrics, and perform statistical analysis of the results. For the second target, it is aimed to use deep neural networks to identify and discriminate different sources of PD. The measurement data are used to generate thousands of phase-resolved PD (PRPD) images that will be used for training deep learning models. To meet the characteristics of the dataset, a deep residual neural network is designed and optimized to discriminate PD sources in an accurate, stable, and time-efficient way. The outcome of this research enhances the reliability of electrical apparatus through a better understanding of the PD behavior and lays a foundation for automatic monitoring of PD sources. | en |
| dc.description.abstractgeneral | Electrical equipment functions properly when its conductive elements are electrically insulated. The science of dealing with insulation systems has become more prominent in recent years due to the novel challenges and circumstances introduced by the rapid electrification trend. As an instance, the electrification trend in transportation systems can impose a multitude of environmental, thermal, and mechanical constraints which were not traditionally considered. These new challenges have led to an accelerated deterioration rate of insulation materials. To address this concern, this study targets the experimentation and modeling of the main aging mechanism in electrical equipment known as partial discharge (PD). A numerical model based on finite-element analysis (FEA) is developed that agrees with the test results and can accurately predict the aging of insulating materials due to the PD phenomenon. Moreover, the growing interest toward electrification of the aviation industry (as a response to the climate change crisis) requires the study of insulating materials under low-pressure (high-altitude) conditions. Theoretical and experimental data confirm the more frequent occurrence of PDs and their higher intensity under low-pressure conditions. Safety of operation is the highest priority in airborne transportation, yet no study has addressed the condition monitoring system as a necessary asset of the electric aircraft. To address this research gap, this work develops a dielectric online condition monitoring system (DOCMS) that actively monitors the deterioration level of insulation using deep learning methods. Based on standardized measurements under low-pressure conditions, the data are preprocessed to train the deep neural network with the pattern of PD activities. The proposed scheme can achieve >82% with short-term signals emitted measured from the system. | en |
| dc.description.degree | Doctor of Philosophy | en |
| dc.format.medium | ETD | en |
| dc.identifier.other | vt_gsexam:33375 | en |
| dc.identifier.uri | http://hdl.handle.net/10919/108321 | en |
| dc.language.iso | en | en |
| dc.publisher | Virginia Tech | en |
| dc.rights | Creative Commons Attribution 4.0 International | en |
| dc.rights.uri | http://creativecommons.org/licenses/by/4.0/ | en |
| dc.subject | Deep learning | en |
| dc.subject | electric aircraft | en |
| dc.subject | finite-element analysis | en |
| dc.subject | high voltage engineering | en |
| dc.subject | high frequency | en |
| dc.subject | insulation systems | en |
| dc.subject | partial discharge | en |
| dc.title | Partial Discharges: Experimental Investigation, Model Development, and Data Analytics | en |
| dc.type | Dissertation | en |
| thesis.degree.discipline | Electrical Engineering | en |
| thesis.degree.grantor | Virginia Polytechnic Institute and State University | en |
| thesis.degree.level | doctoral | en |
| thesis.degree.name | Doctor of Philosophy | en |