Evolutionary Approaches for the Optimal Restoration of Sections in Distribution Systems

Abstract

Due to loss of diversity of loads, the restoration of distribution feeders after long interruptions creates cold load pickup conditions. As a result, the total load briefly exceeds the substation transformer rated load. In order to prevent overheating of these transformers, the distribution system load may have to be restored in a step-by-step manner using sectionalizing switches. The restoration time is dependent on the order in which sections are restored. We propose genetic algorithms and ant colony algorithms as two stochastic optimization algorithms to compute the globally best restoration sequence of sections. Both approaches belong to a class of algorithms known as evolutionary algorithms. While genetic algorithms are well known techniques for optimization, ant colony algorithms have been proposed very recently. Results obtained using both methods for two test cases are presented.