A multi-strategy improved sparrow search algorithm
Chengtian Ouyang, Yaxian Qiu, Donglin Zhu
Abstract
Abstract As a novel algorithm, the sparrow search algorithm has better optimization performance than other intelligent optimization algorithms. However, in complex problems, there is still the possibility of falling into a local optimum and relying on the initial population stage. In response to these shortcomings, a multi-strategy improved sparrow search algorithm (KLSSA) is proposed. First, in the initial population stage, K-means clustering method is used to cluster and differentiate the individual positions of sparrows, which speeds up the work efficiency of the population and gets rid of the influence of randomness. Then, the levy flight mechanism and adaptive local search strategy are respectively introduced in the calculation of the location update of the discoverer and the follower, so that the discoverer can conduct a wide range of searches more flexibly, and the follower has a more detailed search method. Through the 10 standard test functions, it can be seen that the multi-strategy improved sparrow search algorithm has stronger optimization ability and better optimization speed.