CAJ | 학술논문

提出了一种动态多子群协作 QPSO 算法（Dynamic Multiple Sub-population Collaboration Quantum-behaved Particle Swarm Optimization，简称 DMPQPSO），该方法动态构建各子群，并采用混沌策略分2个阶段优化 QPSO，同时对各子群的收缩扩张系数分别进行自适应调整。采用该方法优化 RBFNN，并将 DMPQPSO 算法与标准 PSO和 QPSO 算法对比，仿真实验验证了该方法的优化效果。
제출료일충동태다자군협작 QPSO 산법（Dynamic Multiple Sub-population Collaboration Quantum-behaved Particle Swarm Optimization，간칭 DMPQPSO），해방법동태구건각자군，병채용혼돈책략분2개계단우화 QPSO，동시대각자군적수축확장계수분별진행자괄응조정。채용해방법우화 RBFNN，병장 DMPQPSO 산법여표준 PSO화 QPSO 산법대비，방진실험험증료해방법적우화효과。
A Dynamic Multiple Sub-population Collaboration Quantum-behaved Particle Swarm Optimization is proposed for parameters identification of RBFNN.The method dynamically builds each sub-population,and the chaotic strategy is adopted to optimize the Quantum-behaved Particle Swarm Optimization (QPSO)algorithm in the two stages of search process,at the same time,the contraction expansion coefficient of the algorithm is adjusted adaptively in the evolutionary process according to the fitness of each particle.The proposed method is used to optimize RBFNN,and compared with standard PSO and QPSO.The simulation results show that the opti-mized effect is enhanced.