Abstract: Lately, it makes some progress in research on denoising mutation about Floating Point Representation（FPR）genetic algorithm. A little positive result is gained by some researchers. The theory and method are focal points in research on FPR denoising mutation all the time. The field needs to emerge some newer theory and worthwhile achievement in application. So far, there is research on Tight Wavelet Frame on Bounded Domain（TWFBD）used in FPR denoising mutation. In this paper, it carries out that it emphasizes on analysis on the properties of TWFBD, makes denoising mutation operation in the algorithm with TWFBD, proposes the TWFBD-based Floating Point Representation Genetic Algorithm（TFPRGA）, and makes many experiments. The results of the research and the experiments indicate that the proposed method has reliable theory base, more obvious than other methods in effect.

Key words: bounded domain, tight wavelet frame, denoising mutation, floating point representation, genetic algorithm

摘要： 近年来，有关浮点数编码遗传算法的消噪变异研究有了一定的进展，取得了一些成果。浮点数编码消噪变异的理论和方法研究一直是该领域研究的重点，需要有更新更有理论和应用价值的研究成果出现。有界域的紧小波框架用于浮点数编码消噪变异尚处于无人问津的研究领域。着重分析了有界域的紧小波框架的性质，用有界域的紧小波框架在算法中进行消噪变异操作，提出了基于有界域紧小波框架的遗传算法，并进行了实验。研究和实验结果表明，将有界域的紧小波框架用于浮点数编码消噪变异，具有可靠的理论基础，与其他方法相比，其效果也十分明显。

关键词: 有界域, 紧小波框架, 消噪变异, 浮点数编码, 遗传算法