一种新型的被动启发式粒子群优化算法

A new passive heuristic particle swarm optimization algorithm

标准粒子群优化(SPSO)算法具有参数少、鲁棒性好、易于实现等优点,但同时也具有收敛慢、易于陷入局部极值点等缺点.在SPSO算法基础上,通过在其粒子速度更新公式的自我认知部分和社会经验部分引入惯性学习因子(ωc1,ωc2),提出一种新型的被动启发式粒子群优化算法(PHPSO).分别采用SPSO和PHPSO两种优化算法对测试函数进行求解,将这两种算法的优化过程进行比较分析,结果表明,与SPSO算法相比,该文提出的PHPSO算法收敛速度大幅提高,且更易得到全局最优解,收敛精度更高.

The standard particle swarm optimization（PSO） has the advantages of fewer parameters,robustness and it is easier to implement,but it also has the disadvantage of a slow convergence rate and easily falls into a local extremum point.In this paper,a new passive heuristic particle swarm optimization（PHPSO） algorithm was proposed by introducing the inertia-learning factors to the cognitive component and social component of a standard PSO＇s velocity updating formula.The author applied the standard PSO and the new PHPSO respectively to optimize the test functions.The comparison of the processes in these optimization algorithms shows that the convergence of the new PHPSO is much faster than the standard PSO,and is capable of obtaining a global solution which is much more accurate.