Aiming at the problems that the dung beetle optimizer is prone to local optimization and low convergence accuracy in global optimization,an improved dung beetle optimization algorithm was proposed.By replacing the initial population randomly generated in the original algorithm using the good nodes set sequence,the diversity of the population was improved.The frac-tional calculus method was introduced to adjust regional dynamic boundary to separate overlapping population individuals and improve the local mining performance of the algorithm.A mechanism was proposed to update the global optimal position,preventing it from falling into the local optimal position.The global optimization experiments of 24 benchmark functions and the feature selection experiments of 5 data sets show that the improved algorithm has better optimization performance than the same type of algorithms.
维普
万方数据