To solve global optimization problems, this paper proposed a novel improved version of sine cosine algorithm—the dimension by dimension dynamic sine cosine algorithm (DDSCA). In the update equation of sine cosine algorithm (SCA), the dimension by dimension strategy evaluates the solutions in each dimension, and the greedy strategy is used to form new solutions after combined them with other dimensions. Moreover, in order to balance the exploration and exploitation of SCA, a dynamic control parameter is designed to modify the position equation of this algorithm. To evaluate the effectiveness of DDSCA in solving global optimization problems, it is compared with state-of-art algorithms and modified SCA on 23 benchmark functions. The experimental results reveal the DDSCA has better robustness and efficiency. The IEEE CEC2010 large-scale functions are selected to solve high-dimensional optimization problem, the results show that the performance of the DDSCA is better than other algorithms. In addition, five engineering optimization problems are also verified the effectiveness of the DDSCA. The results of accuracy and speed show that the improved sine cosine algorithm (DDSCA) is competitive in solving global optimization problems.

为了解决全局优化问题，本文提出了一种新颖的改进版本的正弦余弦算法-逐维动态正弦余弦算法（DDSCA）。在正弦余弦算法（SCA）的更新方程中，逐维策略评估每个维的解，而贪婪策略则与其他维组合后用于形成新解。此外，为了平衡对SCA的探索和开发，设计了动态控制参数来修改该算法的位置方程。为了评估DDSCA解决全局优化问题的有效性，将其与最新的算法和经过修改的SCA进行了23种基准测试功能的比较。实验结果表明，DDSCA具有更好的鲁棒性和效率。选择IEEE CEC2010大规模函数解决高维优化问题，结果表明DDSCA的性能优于其他算法。此外，五个工程优化问题也得到了DDSCA有效性的验证。准确性和速度的结果表明，改进的正弦余弦算法（DDSCA）在解决全局优化问题方面具有竞争力。