基于小波低频分量的量化择时策略及仿真模拟

文章运用Best Basis Selection（BBS）算法选取最优小波包基,对上证综指收盘价进行小波包非线性阈值消噪,在消除随机性干扰的基础上,针对传统均线策略买卖信号滞后性的不足,根据不同分解水平的小波低频分量能够反映信号基本和次级趋势且不具滞后性的特点,提出了一种基于小波低频分量的量化择时策略,并对该策略和传统的均线策略分别用R语言进行仿真模拟交易和回测,实验表明在类似风险的情况下,该策略在提示基本和次级趋势买卖信号的同时可以缩短交易信号的滞后性,具有更好的投资表现。

基于小波低频分量微调的IHS变换融合法及其应用

针对IHS变换融合法和小波变换融合法存在的弊端,结合这两种方法,提出了一种基于小波低频分量微调的IHS变换融合法,并通过IKONOS和QuickBird影像数据,验证了新方法的可行性和优越性。新融合法首先利用IHS变换技术提取真彩色影像的亮度分量;接着,为了提高融合影像的光谱保真度,采用小波变换技术,联合亮度分量和全色影像数据,生成更为合理的新亮度分量;同时提出了小波低频分量微调环节,并利用该环节进一步改善了融合影像的高频信息融入效果;最后,执行IHS逆变换,获得真彩色融合影像。研究表明:新融合法兼具IHS变换融合法和小波变换融合法的优势,融合结果的质量得到较大程度提高。

Wavelet analysis of pressure fluctuation signals in a gas-solid fluidized bed

It has been shown that much dynamic information is hidden in the pressure fluctuation signals of a gas-solid fluidized bed. Unfortunately, due to the random and capricious nature of this signal, it is hard to realize reliable analysis using traditional signal processing methods such as statistical analysis or spectral analysis, which is done in Fourier domain. Information in different frequency band can be extracted by using wavelet analysis. On the evidence of the composition of the pressure fluctuation signals, energy of low frequency (ELF) is proposed to show the transition of fluidized regimes from bubbling fluidization to turbulent fluidization. Plots are presented to describe the fluidized bed's evolution to help identify the state of different flow regimes and provide a characteristic curve to identify the fluidized status effectively and reliably.