A novel sparrow search algorithm with integrates spawning strategy

The sparrow search algorithm (SSA), which has recently come under more scrutiny in the intelligent optimization groups. Nevertheless, the high randomicity and local optimum problems restrict the application of the algorithm. We provide a novel mixed sparrow search algorithm (NSSA) in this study that incorporates spawning techniques. Firstly, the good point set theory is suggested as an alternative to the conventional stochastic way to find the initial individual, this method can make the initial population more uniform allocation in the search space. Secondly, the oviposition strategy of cuckoo algorithm is integrated into the discovery phase, in which the global search capability is enhanced. This allows the algorithm to avoid a sudden drop in population diversity and precocious convergence. Then we use levy flight and brownian motion to disturb the individual's position dimension by dimension, and improve the power to escape from local optimal values in the later iterations of the algorithm. Finally, by contrasting the NSSA and SSA, DE, ALO and GOA algorithms with 12 common assessment functions and CEC-2017 test functions, the efficacy superiority and of the proposed scheme are validated.