Multi-Strategy-Driven Salp Swarm Algorithm for Global Optimization

In response to the shortcomings of Salp Swarm Algorithm (SSA) such as low convergence accuracy and slow speed, a Multi-Strategy-Driven (MSD-SSA) was proposed. First, food sources or random leaders were associated with current bottle sea squirt at beginning iteration, which Levy flight walk crossover operators small probability added improve global search ability jump out local optimum. Secondly, position mean leader used establish link followers, effectively avoided blind following followers greatly improved speed algorithm. Finally, Brownian motion stochastic steps introduced populations near sources. The method switched under changes in adaptive parameters, balancing exploration development SSA. simulation experiments, performance algorithm examined using SSA MSD-SSA on commonly CEC benchmark test functions CEC2017-constrained optimization problems, effectiveness verified by solving three real engineering problems. results showed that algorithm, achieved good practical