论文部分内容阅读
在许多固态材料的表面存在着一种天然的非线性振荡花样,这些花样经计算机图像处理在时间和空间上呈现明显的分形特征。利用普通光学显微镜对Cu Zn Al合金表面的金相组织进行观察和分析,用高速摄像机拍摄了典型的振荡现象。借助计算机图像处理技术系统地研究该非线性振荡现象的形貌、结构特征及形成动力学规律,通过计算机编程技术构建系统的非线性动力模型,并重构系统的相空间。用R/S方法对系统时间序列的演化特征进行分析论证,结果表明,这种相空间重构方法具有较好的非线性预报效果。
There is a natural non-linear oscillation pattern on the surface of many solid materials. These patterns show obvious fractal characteristics in time and space by computer image processing. The microstructure of Cu-Zn-Al alloy was observed and analyzed by ordinary optical microscope, and the typical oscillation phenomenon was recorded by high-speed camera. The morphology, structural characteristics and formation dynamics of this nonlinear oscillation are studied systematically by means of computer image processing technology. The nonlinear dynamic model of the system is constructed by computer programming technique and the phase space of the system is reconstructed. The R / S method is used to analyze the evolvement characteristics of the system time series. The results show that this phase space reconstruction method has a good nonlinear predicting effect.