An alternative approaches to predict flashover voltage on polluted outdoor insulators using artificial intelligence techniques

Ali. A. Salem (University Tun Hussein Onn Malaysia (UTHM))
Rahisham Abd Rahman (University Tun Hussein Onn Malaysia (UTHM))
M. S. Kamarudin (University Tun Hussein Onn Malaysia (UTHM))
N. A. Othman (University Tun Hussein Onn Malaysia (UTHM))
N. A. M. Jamail (University Tun Hussein Onn Malaysia (UTHM))
H. A. Hamid (Universiti Malaysia Perlis (UniMAP))
M. T. Ishak (National Defence University of Malaysia (UPNM))

Bulletin of Electrical Engineering and Informatics
Vol 9, No 2: April 2020

This paper presents an alternative approach for predicting critical voltage of pollution flashover by using Artificial Intelligence (AI) technique. Data from experimental works combined with the theoretical results from well-known theoretical modelling are used to derive algorithm for Artificial Neural Network (ANN) and Adaptive Neuro-fuzzy Inference System (ANFIS) for determining critical voltage of flashover. Series of laboratory testing and measurement are carried for 1:1, 1:5 and 1:10 ratios of top to bottom surface salt deposit density on cup and pin insulators. Insulators variables such as height H, diameter D, form factor F, creepage distance L, equivalent salt deposit density (ESDD) and flashover voltage correction are identified and used to train the AI network. Comparative studies have evidently shown that the proposed (AI) technique gives the satisfactory results compared to the analytical model and test data with the Coefficient of determination R-Square value of more than 97%.