一种混合一维搜索的矢量拟态物理学算法

针对约束优化问题,首先引入收缩系数,使得越界个体沿着速度方向回到问题空间,并且保持搜索方向不变。然后利用违反约束量函数来判断个体是否在可行域内,用一维搜索的方法保证个体在迭代过程始终都在可行域内。最后利用矢量拟态物理学优化算法搜索目标问题的最优解。通过对10个典型约束优化测试函数的仿真测试,并与类电磁算法比较发现,本算法在最优适应值的精度上优于类电磁算法,方差的结果也说明了算法的稳定性,这表明该算法是简单有效的。

The Vector Model of Artificial Physics Optimization Algorithm with a Mixed One-dimensional Search

Aiming at the constrained optimization problem,the shrinkage coefficient is firstly introduced to guarantee individual over boundary to go back to the problem space along the direction of speed without changing its search direction.Then the violation degree function is used to determine whether every individual is in the feasible region or not,one-dimensional search method is selected in the algorithm to ensure that all individuals are in the feasible region.VM-APO algorithm is employed to search the optimal solutions of the objective function.Based on 10 typical test functions of the constrained optimization and the comparison with the simulation test of electromagnetic algorithm,this algorithm is superior to the electromagnetism-like algorithm in the precision of optimal fitness,variance result also proves the stability of this algorithm,which indicates that the VM-APO-ODS algorithm is simple and effective.