CAJ | 학술논문

针对传统粒子群算法寻优精度不高、易早熟的缺点，提出了基于黄金分割评判准则的混沌云粒子群（CCGPSO）算法。该算法利用黄金分割评判准则，将粒子群按照适应度大小分为标准粒子、混沌云粒子、云粒子三个子群，分别进行不同的算法操作。黄金分割的引入使整个粒子群可以搜索到全部解空间，解决了标准粒子群算法易陷入局部最优解和寻优精度不高的问题。选取了四种典型函数测试，并与混沌云粒子群算法（CCPSO）比较。仿真结果表明CCGPSO具有较高的寻优精度和收敛速度。
침대전통입자군산법심우정도불고、역조숙적결점，제출료기우황금분할평판준칙적혼돈운입자군（CCGPSO）산법。해산법이용황금분할평판준칙，장입자군안조괄응도대소분위표준입자、혼돈운입자、운입자삼개자군，분별진행불동적산법조작。황금분할적인입사정개입자군가이수색도전부해공간，해결료표준입자군산법역함입국부최우해화심우정도불고적문제。선취료사충전형함수측시，병여혼돈운입자군산법（CCPSO）비교。방진결과표명CCGPSO구유교고적심우정도화수렴속도。
To improve the low accuracy and premature convergent in traditional Particle Swarm Optimization(PSO)algorithm, the chaos cloud particle swarm algorithm based on golden section evaluation criteria(CCGPSO)is proposed in this paper. This method divides the particle swarm into standard particle, chaos cloud particle and cloud particle using the golden section judge principles according to fitness level, every sub swarm particle has respective different algorithm operations. The golden section enables particle swarm to search the entire solution space, solves the problems of easily falling into local optimum and low accu-racy in basic PSO. This paper chooses four reference functions to have a test and compared with chaos cloud particle swarm opti-mization(CCPSO). The simulation results demonstrate CCGPSO has high optimization precision and convergence speed.