Voltage quality enhancement with wavelet-artificial intelligence current limiting devices

Due to the increased use of digital computers, PWM adjustable speed drives, and automated controllers in industries, utilities are faced with rising c oncerns about the quality of power supplied. This paper, based purely on si mulations, focuses on the application of current limiting devices on radial distribution systems to enhance the voltage quality at the distribution bu ses. The first section of the paper demonstrates the effectiveness of the l imiting devices in reducing the dips in voltages and associated phase-angle jumps. Because the current devices need to be introduced to the systems al most instantaneously after the occurrence of a fault, the second section in troduces a new adaptive fault-sensing scheme for use with these devices. Th is scheme utilizes wavelet packets analysis as a preliminary feature extrac tor and a neural network as a pattern classifier. Simulations show that the scheme required no user interference and can be applied to any network con figuration.