论文部分内容阅读
针对传统粒子群算法在求解高维空间中复杂多峰函数时容易陷入局部最优的问题,提出带反向预测和斥力因子的改进粒子群优化算法.算法通过引入反向预测因子改进速度更新方式,以降低粒子在运动过程中产生惰性而出现早熟收敛的概率,并给出带斥力因子的位置修正策略,使粒子均匀分散于搜索空间,从而避免陷入局部最优.实验分析表明,在对高维空间中复杂多峰函数进行优化求解时,改进的粒子群优化算法较传统粒子群算法更加优越.
Aiming at the problem that the traditional particle swarm optimization is apt to fall into the local optimum when solving complex multimodal functions in high dimensional space, an improved particle swarm optimization algorithm with backward prediction and repulsive force factor is proposed. The algorithm improves the speed update mode by introducing backward predictors , So as to reduce the probability of premature convergence caused by the inertia of the particles during the movement and give the location correction strategy with the repulsive force factor so that the particles are uniformly dispersed in the search space so as to avoid falling into the local optimum.The experimental analysis shows that, When solving complex multimodal functions in dimension space, the improved particle swarm optimization algorithm is superior to the traditional particle swarm optimization algorithm.