Azam Bagheri
Azam Bagheri, PhD in Electric Power Engineering at Luleå University of Technology. Photo: Kersti Bergkvist. View original picture , opens in new tab/window

Machine learning behind new methods for analyzing power quality

Published: 11 December 2018

New methods for analyzing large amounts of measurement data regarding power quality is now presented in a doctoral dissertation at Luleå University of Technology in Skellefteå. The extracted information can show potential impacts of power quality disturbances and how sensitive equipment can be improved to increase the reliability of the power supply.

The area of power quality is growing rapidly, both in terms of the need for good quality of transmission of electricity, but also when it comes to managing large volumes of complex measurement data. Power quality data has all the characteristics of "Big Data" and requires special treatment methods.

– In my doctoral dissertation I have developed advanced methods for analyzing large amounts of data. The methods are based on machine learning and deep learning techniques and aim to extract additional information from data that measures power quality. The extracted information may refer to the potential consequences of electrical damage to sensitive equipment, but also to the underlying cause of the interference. The power quality information can also be used to improve fault ride through capability of the equipment and to increase the reliability of the power supply, says Azam Bagheri, new PhD in Electric Power Engineering at Luleå University of Technology.

Tags