Design of fuzzy logic controllers using genetic algorithms

摘要: Since membership functions and fuzzy control rules are interdependent in designing a logic controller (FLC), GA-based approach is proposed for simultaneous design of these two components. With triangular functions, the left right widths locations their peaks, output set corresponding to every possible combination input sets then chosen as parameters be optimized. In proportional scaling method, transformed into real-coded chromosomes, over which arithmetical crossover nonuniform mutation implemented. Meanwhile, enlarged sampling space ranking mechanism also used evolution process. To show application cart-centering example given. From simulation results, we find that designed FLC robust can drive cart system from any given initial state desired final state, verifies feasibility validity method.