Zero-order TSK-type fuzzy system learning using a two-phase swarm intelligence algorithm

摘要: This paper proposes zero-order Takagi-Sugeno-Kang (TSK)-type fuzzy system learning using a two-phase swarm intelligence algorithm (TPSIA). The first phase of TPSIA learns structure and parameters by on-line clustering-aided ant colony optimization (ACO). Phase two aims to further optimize all the free in particle (PSO). In ACO (CACO), is learned through clustering. Once new rule generated clustering, consequent selected from discrete set candidate values ACO. ACO, path an regarded as combination every rule. CACO helps locate good initial systems for subsequent learning. two, particles PSO are randomly according best solution found CACO. All designed optimally tuned PSO. Simulations on control three nonlinear plants conducted verify performance. Comparisons with other algorithms demonstrate