Journal of System Simulation ›› 2020, Vol. 32 ›› Issue (8): 1473-1480.doi: 10.16182/j.issn1004731x.joss.19-0019

Previous Articles     Next Articles

Structure Learning of Fuzzy-tree Based on Rigorous Binary Tree Code and Genetic Algorithm

Liu Changliang1,2, Wang Ziqi2   

  1. 1. State Key Laboratory of Alternate Electrical Power System with Renewable Energy Sources, North China Electric Power University, Beijing 102206, China;
    2. School of Control and Computer Engineering, North China Electric Power University, Baoding 071003, China
  • Received:2019-01-11 Revised:2019-03-27 Online:2020-08-18 Published:2020-08-13

Abstract: To solve the problems of information redundancy and low optimization efficiency in the structure learning of fuzzy-tree model, a method based on rigorous binary tree code and genetic algorithm is proposed. The structure of fuzzy-tree model is coded by rigorous binary tree code, which improves the information redundancy of the existing matrix code. Considering the particularity of the code and the convergence of the algorithm, an improved genetic algorithm is proposed to optimize the structure of fuzzy-tree model. The experimental results show that the algorithm has good stability and computing speed on different data sets, and can find a better binary tree structure, and that improves the modeling accuracy of fuzzy-tree model.

Key words: fuzzy-tree, rigorous binary tree code, genetic algorithm, structure learning

CLC Number: