Fruit fly optimization algorithm is a novel metaheuristic optimization algorithm based on foraging of fruit fly proposed by Taiwan scholar Wen-Tsao Pan in 2011. Compared with some Swarm intelligence optimization algorithm, fruit fly optimization algorithm has the characteristics of a simple structure, less parameters, easy to adjust, and it’s easy to understand and program, so the algorithm attracts more and more attention to the scholars, and becomes one of the important research in the field of computational intelligence. At present, fruit fly optimization algorithm has been successfully applied to the parameter optimization of support vector machine, the parameter optimization of generalized regression neural network, the parameter optimization of grey neural network, permutation flow shop scheduling problem, TSP problem and multidimensional knapsack problem and so on. With the application of in-depth study, people found that the algorithm still has some shortcomings, such as easily trapped into local optimal value, low convergence precision, and it is not suitable to handle the independent variable which is negative value. In view of the shortcomings of the algorithm, This thesis will use the integration strategy improving the shortcomings of fruit fly optimization algorithm and proposes several improved fruit fly optimization algorithm. The aim is to improve the performance of fruit fly optimization algorithm and improve the research on the theory of the fruit fly optimization algorithm.The research results obtained are as following:(1) In view of the problems of easily relapsing into local extremum and low convergence precision of fruit fly optimization algorithm, this paper proposes an adaptive fruit fly optimization algorithm based on velocity variable. The idea of this algorithm based on the flight characteristics of fruit fly, using particle swarm optimization concept of particle velocity, improving the convergence speed of fruit fly optimization algorithm by adding the particle velocity variable parameter based on fruit fly optimization algorithm. test results show that the convergence speed and precision of the improved fruit fly optimization algorithm are improved obviously(2) To expand the information of individuals, the idea of diploid of complex-valued encoding is introduced in fruit fly optimization algorithm, and proposing a fruit fly optimization algorithm based on complex-valued encoding. The independent variables of the objective function are determined by the modules and angles of their corresponding complex numbers. experimental results show that the algorithm is effective.(3) The idea of Levy Flights is introduced in fruit fly optimization algorithm, taking The idea of Levy Flights can enhance algorithm’s ability of jumping out of local optimum, proposing a fruit fly optimization algorithm based on Levy Flights Trajectory. experimental results show that improved algorithm effectively improve the accuracy and accelerate the convergence speed of the algorithm. |