A Sparrow Search Algorithm Based on Linnik Flight Improvements

In view of the problems of low convergence accuracy, slow convergence speed, and serious convergence in the later stage of population search, resulting in insufficient local exploration in the sparrow search algorithm. In this paper, an improved Sparrow Search Algorithm (LMSSA) based on Linnik flight is proposed. First, the chaos mapping theory was improved to obtain a more uniform and diverse initial sparrow population. Secondly, Linnik flights are added to enhance a wide range of perturbation search operations, thereby expanding the variability of the search, increasing the diversity of the search, and avoiding falling into a local minimum. Thirdly, the search characteristics are dynamically adjusted through the adaptive T-distribution to adapt to different search spaces and search states. Finally, a new dynamic boundary control has been added to speed up convergence. Experimental results show that the convergence and accuracy of LMSSA on the classical benchmark function and the CEC2017 test function are better than other improved sparrow search algorithms and advanced meta-heuristic optimization algorithms.