AUTOMATIC CLASSIFICATION OF POLES IN ENERGY TRANSMISSION LINES WITH DEEP LEARNING
9 INTERNATIONAL CONGRESS ON ENGINEERING AND TECHNOLOGY MANAGEMENT, İstanbul, Turkey, 29 April - 01 May 2023, pp.124-132, (Full Text)
- Publication Type: Conference Paper / Full Text
- City: İstanbul
- Country: Turkey
- Page Numbers: pp.124-132
- Open Archive Collection: AVESIS Open Access Collection
- Bursa Uludag University Affiliated: Yes
Abstract
Abstract: The fact that different types of power transmission poles have different
properties and are designed to withstand different loads and weather conditions reveals
that the poles used in energy transmission lines should be classified. By classifying the
poles, engineers can ensure that the appropriate pole is selected for each location along
the transmission line. This helps ensure that the transmission line can transmit energy
safely, efficiently and reliably. It may be useful and necessary to use a deep learning model
trained and tested with visual data to classify poles used in power transmission lines.
Automating the pole classification process can save time and resources and help ensure
that the appropriate pole is selected for each location along the transmission line.
Classifying poles using deep learning techniques such as Convolutional Neural Networks
(CNNs) can contribute to science by improving our understanding of the physical
properties and behavior of power transmission poles. In this study, automatic
classification of poles used in power transmission lines was performed using CNN and an
accuracy rate of 96.6% was obtained.