Salp swarm algorithm (SSA) based on the swarming behaviour of salps found in ocean, is a very competitive algorithm and has proved its worth as an excellent problem optimizer. Though SSA is a very challenging algorithm but it suffers from the problem of poor exploitation, local optima stagnation and unbalanced exploration and exploitation operations. Thus in order to mitigate these problems and improve the working properties, seven new versions of SSA are proposed in present work. All the new versions employ new set of mutation properties along with some common properties. The common properties of all the algorithms include division of generations, adaptive switching and adaptive population strategy. Overall, the proposed algorithms are self-adaptive in nature along with some added mutation properties. For performance evaluation, the proposed algorithms are subjected to variable initial population and dimension sizes. The best among the proposed is then tested on CEC 2005, CEC 2015 benchmark problems and real world problems from CEC 2011 benchmarks. Experimental and statistical results show that the proposed mutation clock SSA (MSSA) is best among all the algorithms under comparison.

Salp群算法（SSA）基于海洋中出现的大量蜂群的行为，是一种非常有竞争力的算法，并已证明其作为出色的问题优化器的价值。尽管SSA是一个非常具有挑战性的算法，但是它遭受着开发不善，局部最优停滞以及不平衡的勘探和开发操作的问题。因此，为了减轻这些问题并改善工作性能，在当前工作中提出了七个新版本的SSA。所有新版本都采用了一组新的突变属性以及一些常见属性。所有算法的共同属性包括世代划分，自适应切换和自适应种群策略。总体而言，所提出的算法本质上具有自适应性以及一些附加的突变特性。为了进行绩效评估，所提出的算法受到可变的初始种群和维度大小的影响。然后，在CEC 2005，CEC 2015基准测试问题和CEC 2011基准测试中的实际问题上测试建议中的最佳方法。实验和统计结果表明，所提出的变异时钟SSA（MSSA）是所有比较算法中最好的。