By Topic

Neural networks for combined control of capacitor banks and voltage regulators in distribution systems

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$31 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

2 Author(s)
Gu, Z. ; Oak Ridge Nat. Lab., TN, USA ; Rizy, D.T.

A neural network for controlling shunt capacitor banks and feeder voltage regulators in electric distribution systems is presented. The objective of the neural controller is to minimize total I2R losses and maintain all bus voltages within standard limits. The performance of the neural network for different input selections and training data is discussed and compared. Two different input selections are tried, one using the previous control states of the capacitors and regulator along with measured line flows and voltage which is equivalent to having feedback and the other with measured line flows and voltage without previous control settings. The results indicate that the neural net controller with feedback can outperform the one without. Also, proper selection of a training data set that adequately covers the operating space of the distribution system is important for achieving satisfactory performance with the neural controller. The neural controller is tested on a radially configured distribution system with 30 buses, 5 switchable capacitor banks and a nine tap line regulators to demonstrate the performance characteristics associated with these principles. Monte Carlo simulations show that a carefully designed and relatively compact neural network with a small but carefully developed training set can perform quite well under slight and extreme variation of loading conditions

Published in:

Power Delivery, IEEE Transactions on  (Volume:11 ,  Issue: 4 )