Publication: Determination of transfomer lifespan using artificial intelligence
Date
2020
Authors
Abid Akalil Akmal A Hamid
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Transformer is one of the important assets in electrical power system. In general, transformer is designed to operate for a long period of time. The expected life of a transformer is between 25 years to 40 years. However, as time goes by, transformer will experience degradation and any undesired event could happened in sudden. Hence, Health Index (HI) assessment is performed in order to access the condition of the transformer and determine its remaining lifespan. Still and all, it could be time consuming to perform the HI assessment since it requires some substantial amount of data. In this modern era, a strategy that could be used to solve the issue is by integrating the HI assessment with Artificial Intelligence (AI) to automate the remaining lifespan of the transformer. To implement this, the Al is required to perform classification tasks in order to classify the condition of the transformers. The Al techniques which will be used are the Support Vector Machine (SVM) and the K-Nearest Neighbours (KNN). These AI techniques will be trained with 100 sets of data. It is observed that the SVM performed better compared to KNN as the SVM obtained higher accuracy than the KNN in predicting the desired output.
Description
TK2551.A25 2020
Keywords
Electric transformers