基于结晶过程的分子动理论优化算法

A kinetic-molecular theory optimization algorithm based on crystallization process

针对分子动理论优化算法(KMTOA)存在易陷入局部最优、寻优精度低等问题,提出一种基于结晶过程的分子动理论优化算法(C-KMTOA)。该算法通过模拟结晶过程设计了一种分离算子,该算子将种群分为最优个体、优秀个体、较差个体三个子群,并通过引导操作使较差个体向优秀个体附近移动、优秀个体向最优个体附近移动,从而使搜索范围快速缩小到最优解附近。实验结果表明,该算法在优化精度、动态性能等方面均优于GA、DE、QPSO和KMTOA。

We propose a kinetic-molecular theory optimization algorithm based on crystallization process （C-KMTOA） to solve the problem that the KMTOA is easily stuck into local optimal and low accuracy. We also design a separation operator by simulating the crystallization process, which divides the population into three subgroups： the best individuals, the excellent individuals and the worst individuals. In addition, with the help of guiding operation, the worst individuals can move toward the excellent individuals and the excellent individuals move toward the best individuals, so that the search range is narrowed down to the optimal solution quickly. Experimental results show that the proposed algorithm is superior to the GA, DE, QPSO, and KMTOA algorithms in terms of optimization precision and dynamic performance.