Flower pollination algorithm(FPA) is a new nature-inspired algorithm designed by YANG X.S. at Cambridge University in 2012 through simulating the flowerâ€™s behavior of pollination. Flower pollination algorithm has been a hot research area in intelligent optimization algorithms. The FPA has been applied in solving optimization problems and engineering application problems successfully because of its efficient ability in exploring the search domain and converging to the global optimal solution. In this paper, aiming at the problem of easily fall into local optimal solution, an improved algorithm of Flower pollination algorithm- Chaos-based Flower Pollination Algorithm(CBFPA) is put forward,and use it to deal with the Radar positioning problem. The researches in this paper are as follows:(1) The biological mechanism, basic steps and the domestic and foreign research situation of the pollen algorithm are described. And the advantages and disadvantages of the algorithm are analyzed.(2) Aiming at the phenomena of early convergence and local optimization, flower pollination algorithm is improved by improving the convergence speed and precision of the algorithm. In the initial stage of flower pollination algorithm, the chaotic sequence is introduced to initialize pollen grains. Then, the reverse learning mechanism and the Gauss disturbance strategy are introduced in the iterative process. Finally, a new algorithm based on chaos is proposed. The CBFPA algorithm and FPA algorithm are simulated and compared by 10 typical test functions(respectively, for the case of 2, 10 and 50). Which include the aspects of convergence rate comparison, average optimal value and variance contrasts and t test contrast. The simulation results show that the performance of CBFPA algorithm is better than performance of FPA algorithm.(3) The CBFPA algorithm is used to solve practical problem- radar positioning. The experimental results show that this algorithm can effectively solve such problems. |