一种改进的人工电场算法

An Improved Artificial Electricfield Algorithm for Global Optimization

人工电场算法(AEFA)是受库伦定律和运动定律启发而发展起来的一种新型启发式算法。在整个搜索过程中,个体位置变化是基于步长计算的,步长依赖于一个常数,即引力常数(K)。AEFA算法中采用指数函数来计算常数K(t)。较大的引力常数可以增加AEFA的探测能力,较小的引力常数可以提升AEFA的开采能力。然而这种指数变化使得K的大小随着迭代次数的迅速的递减,会使AEFA的搜索能力快速变弱,使得AEFA在早期阶段对整个搜索空间的搜索不充分,从而导致搜索结果不是最优的,为了提高算法的探索能力,论文提出一种新的引力常数计算方法。实验将改进后的算法和原始的AEFA算法在3个标准测试函数上进行对比,实验结果表明改进后算法的性能具有明显的提升。

Artificial electric field algorithm is a noval population-based metaheuristic algorithm which inspired by Coulomb's Law and law of motion.In whole search process,the agent's position change is calculated based on the step size which depends on the gravitational constant(K).In AEFA algorithm an exponentially function is used to calculate the constant K(t).A larger gravita⁃tional constant can increase the exploration ability of AEFA,and a smaller gravitational constant can enhance the exploitation capac⁃ity of AEFA.However,this exponential change causes the size of K to decrease rapidly with the number of iterations,and will make AEFA's search capability weaken,make AEFA insufficient search of the entire search space in the early stages,and miss the opti⁃mal solution.In order to improve the exploration ability of the algorithm,this paper proposes a new calculation method of gravitation⁃al constant.The experiments are performed on 3 benchmark functions.The experimental results show that the improved gravitational constant performance of the algorithm is significantly improved and compared with the original AEFA algorithm.