Hybrid Flower Pollen Algorithm Based on Modified Search Strategy

Abstract

The pollen algorithm has many characteristics, such as fewer parameters, easy adjustment, and the excellent search path, and has been applied to many engineering optimization problems. But there are easy to premature convergence, slow convergence and easy to fall into the local optimum and other deficiencies, in order to improve the convergence speed and precision of the algorithm, the pollen algorithm is improved in the early stage of the pollen algorithm, the theory of good lattice points is introduced, the pollen gamete are initialized, and the pollen gamete are initialized, In the iterative process, we add the opposition learning mechanism and introduce the improved cross strategy and the Gaussian disturbance strategy of the genetic algorithm in the optimal position, and propose a hybrid pollen algorithm based on modified search strategy based on the improved search strategy(HMFPA). The HMFPA algorithm and the FPA algorithm are simulated by four typical complex test functions in the optimization problem. The results show that the convergence speed and accuracy of the HMFPA algorithm are better than those of the FPA algorithm.