求解高维优化问题的改进正弦余弦算法

Improved sine cosine algorithm for solving high-dimensional optimization problems

提出一种改进的正弦余弦算法(简记为ISCA)。受粒子群优化(PSO)算法的启发,引入惯性权重以提高正弦余弦算法的收敛精度和加快收敛速度。此外,采取反向学习策略产生初始个体以提高种群的多样性和解的质量。采用八个高维基准测试函数进行仿真实验:在相同的最大适应度函数评价次数下,ISCA总体性能上均优于基本SCA和HGWO算法;当维数较高(D=1 000)时,ISCA所用计算量远小于HDEOO算法。实验结果表明ISCA在收敛精度和收敛速度指标上均优于对比算法。

This paper proposed an improved sine cosine algorithm(ISCA)for solving high-dimensional function optimization problems.It inspired by particle swarm optimization(PSO)algorithm,introduced inertia weight to enhance the convergence precision and accelerate the convergence speed.In addition,in order to enhance the diversity of population and solution quality,when producing the initial individuals,it employed the opposite-based learning method.It conducted simulation experiments on the 8 benchmark high-dimensional functions.The compute consumption of ISCA was far less than HDEOO in high dimension(D=1 000),and its overall performance was much better than the basic SCA and HGWO algorithm in the same number of maximum fitness function evaluation.The experimental results demonstrate that the proposed ISCA has better performance in convergence precision and convergence speed.