Affiliations: [a] Key Laboratory of Power Electronics Energy-Saving Technologies & Equipment Xihua University,
Chengdu, China | [b] Graduate School of Science and Technology, Tokai University, Toroku, Kumamoto,
Japan | [c] College of Automation, Harbin Engineering University, Harbin, Heilongjiang, China | [d] College of Engineering and Science, Victoria University, Melbourne, VIC,
Australia
Abstract: As a new branch of natural computing, membrane computing (MC) has become a hot topic. Based on the combination of nested structure membrane optimization method, genetic algorithm (GA) and the distributed computing method, an efficient parallel genetic membrane computing (PGMC) is proposed. Some rules are proposed to improve the computational performance of PGMC such as communication and transportation rules between homo-core membranes and hetero-core membranes, elementary membrane crossover and division rules, mutation and dissolving rules. An application of PGMC to distribution network reconfiguration is presented. According to the features of radial distribution network operation, object generation of minimum loop and equal selection of crossover probability are used to further improve the computational efficiency. Finally, a typical example of 33-nodes net is simulated by comparing PGMC with general GA and genetic membrane computing (GMC). The results demonstrate superiority of PGMC on convergence, stability, global searching ability and so on.
