In this paper, a novel optimization method is proposed based on the sparrow search algorithm, namely, multi-strategy-guided sparrow search algorithm (MGSSA). It is well-known that the basic SSA has limitations such as the slow convergence speed and vulnerability to local optimality. Surrounding these two issues, some strategies are presented in the MGSSA. Firstly, the newly introduced ring topology search strategy not only maintains the diversity of the entire population but also enhances the exploration ability of the SSA. Secondly, the proposed leader-based search strategy can improve exploitation ability of the SSA to prevent falling into local optimum as much as possible. Moreover, the coordinated learning strategy is put forward to better balance between the exploration and exploitation abilities. Finally, the MGSSA is compared with seventeen advanced algorithms on two well-known benchmark suites (i.e., CEC-2017 and CEC-2020). Meanwhile, the MGSSA-based forecasting approach is applied to predict the remaining useful life for lithium-ion batteries. The statistical results indicate that the MGSSA is a high-performance optimizer, which can not only solve the defects of the original SSA, but also obtain satisfactory solutions in both complex numerical optimization and real-world application problems.

本文提出了一种基于麻雀搜索算法的优化方法，即多策略引导麻雀搜索算法（MGSSA）。众所周知，基本SSA存在收敛速度慢、易陷入局部最优等局限性。围绕这两个问题，MGSSA 提出了一些策略。首先，新引入的环形拓扑搜索策略不仅保持了整个种群的多样性，而且增强了SSA的探索能力。其次，所提出的基于领导者的搜索策略可以提高SSA的开发能力，以尽可能防止陷入局部最优。此外，提出了协调学习策略，以更好地平衡探索和利用能力。最后，MGSSA 在两个著名的基准测试套件（即 CEC-2017 和 CEC-2020）上与 17 种先进算法进行了比较。同时，采用基于MGSSA的预测方法来预测锂离子电池的剩余使用寿命。统计结果表明，MGSSA是一种高性能优化器，不仅可以解决原始SSA的缺陷，而且在复杂的数值优化和实际应用问题中都可以获得满意的解决方案。