Abstract
Transformer insulation paper is a key indicator for transformer remaining operational life. Paper decomposition is evaluated using the degree of polymerization (DP) which calls for samples of insulation paper from operating transformers. Since collecting such paper samples is extremely difficult, other indicators have been used to indirectly reveal the DP value of insulation paper. This includes dissolved gases in transformer oil such as carbon oxides and hydrocarbon gases, furan compounds, methanol, ethanol, and moisture along with some oil characteristics such as interfacial tension. However, for the same oil sample, these individual parameters lead to different DP values. This is attributed to the lack of accuracy of the established mathematical and artificial intelligence models correlating DP with each of the above mentioned individual parameters. This paper presents a self-learning method to estimate the DP value of transformer insulation paper based on multiple transformer oil aging parameters. The proposed method comprises data processing, fuzzy c-means and linear regression. Results reveal that estimating the DP value based on multiple aging parameters is more accurate than estimating it using one single parameter as per the current practice. The proposed method not only helps to understand the correlation between multiple oil aging parameters and the DP value of paper insulation, but also promotes the establishment of more accurate life assessment models for power transformers based on these oil aging parameters.
Highlights
Oil impregnated paper (OIP) is widely used as a solid insulating material in power equipment, such as power transformers [1]
The useful remaining operational life of a power transformer is identified based on the condition of its solid insulation that is measured using the degree of polymerization (DP) [3]
This is attributed to the lack of accuracy of the developed mathematical and artificial intelligence models correlating these individual indicators with the DP value
Summary
Oil impregnated paper (OIP) is widely used as a solid insulating material in power equipment, such as power transformers [1]. Estimating the DP based on one single indicator results in inconsistent DP values for the same oil-paper sample This is attributed to the lack of accuracy of the developed mathematical and artificial intelligence models correlating these individual indicators with the DP value. The main contribution of this paper is the presentation of a new self-learning method to estimate the DP value of paper insulation based on multiple oil aging parameters that can be measured using transformer oil samples. In this context, series of thermal aging experiments on oil-paper samples are conducted and the proposed indicators are measured and randomly split into training and testing datasets. The proposed FCM-LR method is applied on other thermal aging experimental conditions and different types of oil and paper to further validate the robustness of the developed model when employed by different oil-paper materials
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
