摘要
通过引入梯度信息来影响粒子速度的更新,构造了一种带有梯度加速的粒子群算法.为减小陷入局优的可能性,当群体最优信息陷入停滞时,对群体进行部分初始化来保持群体的活性,并讨论了改进算法的适用范围.仿真结果表明,对于单峰函数和多峰函数,改进算法都能够取得较好的优化效果.
By adding gradient information to influence the update of velocities of the particles, a kind of particle swarm optimization (PSO) algorithm with gradient acceleration is proposed. When the optimum information of the swarm is stagnant, some particles in the population are initialized again to reduce the possibility of trapping in local optimum. The scope of application is also discussed, and the result of computer simulation indicates that the improved PSO could get better performance in one-peak functions and multi-peak functions.
出处
《控制与决策》
EI
CSCD
北大核心
2004年第11期1298-1300,1304,共4页
Control and Decision
关键词
粒子群算法
演化计算
随机搜索
Computer simulation
Convergence of numerical methods
Gradient methods
Iterative methods
Stochastic control systems