Voltage Stability Index and Voltage Deviation Improvements using Intelligent Algorithms for Capacitor Sizing and Placement

Abstract

Voltage stability of each bus of an electrical distribution system along with the deviation of voltage magnitudes from the tolerance limits are two of the most common but significant issues in the present day distribution networks. This work is carried out with the objective of identifying the optimal location and ideal sizes of capacitors to be used in a radial distribution network for alleviating the above problems. An intelligent two-stage methodology is used that employs genetic algorithm and neural networks in order to achieve this objective. By analyzing the load flow study of the base case of load profile, voltage deviation and voltage stability indices are calculated for each node of the system. From these indices, the candidate buses are identified for the capacitor allocation in stage one of the methodology. Then, the combination of two artificial intelligent algorithms based on modern learning methods are employed to identify the ideal sizes of the capacitors for operation throughout the day. This methodology is implemented on a 33 bus radial distribution network. The results of these intelligent algorithms predict the ideal sizes of capacitors with optimal locations, providing a stable system with a smooth voltage profile across the entire duration of the day.