Premium
Genetic algorithm sampling the solution space selectively depending on difficulty of power distribution network restoration
Author(s) -
Kato Yoshifumi,
Hamagami Tomoki
Publication year - 2012
Publication title -
electronics and communications in japan
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.131
H-Index - 13
eISSN - 1942-9541
pISSN - 1942-9533
DOI - 10.1002/ecj.10389
Subject(s) - genetic algorithm , table (database) , computer science , sampling (signal processing) , accident (philosophy) , algorithm , power (physics) , artificial intelligence , data mining , computer vision , machine learning , filter (signal processing) , quantum mechanics , philosophy , physics , epistemology
A new genetic algorithm for a multiagent‐based autonomous power distribution network restoration system is proposed. The state of the art of this study is to realize a new genetic algorithm using selective sampling for improving the restoration performance. The proposed method realizes the selective sampling by a virtual accident selecting algorithm that changes the probability of selecting the virtual accident area. The virtual accident selecting algorithm consists of a weight table and an area‐value list. The weight table represents the difficulty of restoration in each accident area. The area‐value list represents the difficulty of restoration in the latest generation, and affects the weight table in the next generation. This architecture enables the system to alter the probability of changing each virtual accident area autonomously from restoration simulation. The simulation results show that the proposed method is capable of improving performance. © 2012 Wiley Periodicals, Inc. Electron Comm Jpn, 95(2): 16–24, 2012; Published online in Wiley Online Library ( wileyonlinelibrary.com ). DOI 10.1002/ecj.10389