【作者】 郑东亮； 薛云灿； 李斐； 杨启文；

【Author】 Zheng Dongliang~1 Xue Yuncan~2 Li Fei~2 Yang Qiwen~2 (College of Computer and Information Engineering Hohai University Changzhou 21302 China)

【机构】 河海大学计算机及信息工程学院；

【摘要】 针对离散粒子群优化收敛速度、精度低,但其充分利用了粒子的局部最优和全局最优值信息和郭涛算法收敛速度和精度较高但其当前粒子向种群中其它粒子学习时具有盲目性而限制了算法的进化速度的特点,本文提出了一种基于序列倒置的改进离散粒子群优化算法。实验研究表明,该算法和普通离散粒子群优化算法相比,不论是收敛速度和还是求解精度都有了较大提高。改进离散粒子群优化算法应用于罩式退火车间钢卷自动组合堆垛,并取得很好的效果。更多还原

【Abstract】 The discrete particle swarm optimization(DPSO) can make the best of the particles’ local optima and the particles’ global optima,but it has the disadvantages of slow convergence and search precision.The Gu Tao’s algorithm is fast converged and high precise but it is blindfold when it learns from the others’ particles.This disadvantage limits its evolution velocity.Based on these,a novel modified discrete particle swarm optimization algorithm base on the inver-over operator(IDPSO) is presented.Experiments carried out on the travel salesman problem show that the modified DPSO achieve good results.Compared with the general DPSO,the new algorithm can improve the performance of the DPSO and speed up the velocity of the DPSO convergence.The modified discrete particle swarm optimization algorithm was applied to automatic combination stacks of steel roll.更多还原