Research on Floating Point Representation Denoising Mutation Based on GFMRA

摘要

Multiresolution analysis (MRA) was an important method of constructing wavelet. Generalized frame multiresolution analysis (GFMRA) could construct any orthonormal wavelet based on single mother function. Floating point representation (FPR) was superior to other representation in function optimization and restriction optimization. The noise that FPR brings about influenced badly the performance of genetic algorithm in genetic operation environment. This paper was dependent on theoretical analysis. It presented floating point repreaentation genetic algorithm (FPRGA) based on GFMRA (FPRGAG). FPRGAG was a method of FPR denoising mutation by orthonormal wavelet. The experiments were made on FPRGAG The results of the theoretical research and the experiments in it indicate which FPRGAG is superior to other used algorithms, in convergence efficiency and precision. The method is reliable in theory, is feasible in technique.