An Enhance Sparrow Search Algorithm Based on Individual Dispersion and Niche Mechanism

Aiming at the problem of Sparrow Search Algorithm(SSA)'s lower optimization accuracy and being easily trapped into local optimal, this paper proposes an enhanced sparrow search algorithm based on individual dispersion and niche mechanism, namely DNSSA. At First, DNSSA adopts an individual dispersion and fitness value based adaptive selection approach to select producers in order to reducing the aggregation around local optimal. Secondly, DNSSA applies a species formation strategy based on adaptive niche mechanism to maintain the diversity of the whole population, so as to avoid premature convergence of the algorithm effectively. 6 functions from CEC2017 test suit were used as the benchmark tested to evaluate DNSSA and other 5 state-of-the-art improved SSA algorithms, namely t-GSSA, MSSA, KSSA, HSSA and CSSOA. Experimental results show DNSSA can achieve supreme performance by applying all strategies simultaneously, and also illustrate that DNSSA even can raise convergence accuracy up by 10% compared with the second ranked evaluated algorithm on most of multimodal objective test functions.