A novel paddy field algorithm based on pattern search method

摘要

Paddy field algorithm (PFA) is a fast random algorithm with global search capability. But when the number of solutions is over the range, the algorithm efficiency becomes low because it executes a lot of redundant iterations. Pattern search algorithm is sensitive to the initial condition. In order to strength the local search ability, paddy field algorithm is introduced and a novel paddy field algorithm is proposed in this paper. The hybrid algorithm operates by initially scattering seeds at random in the parameter space. The final result is found by pattern search based on the result of PFA algorithm. This algorithm is tested on three sample functions alongside the basic paddy field algorithm and pattern search method. The simulation results show that the algorithm performs well.