非线性时间序列的小波分频预测

基于噪声的小波变换特点，结合小波包分解和模极大重构来抽取含噪信号的主分量，提出了一种基于最佳尺度分解和Volterra自适应滤波的分频预测算法，使用较少的模型训练样本，同时具有强的抗噪能力.该算法克服了传统小波分解尺度选取的盲目性及单纯Volterra预测器抗噪性能的不足，数值仿真表明，针对含强噪声的非线性信号可进行有效预测. 关键词： 小波分解 Volterra自适应滤波器 分频预测     

应用小波分析进行油菜籽红外光声光谱去噪

为探究小波分析对油菜籽红外光声光谱的去噪效果,利用db6小波4尺度分解对其进行去噪研究.对比分析低频系数重构、缺省阈值、Birge-Massart阈值和4种自适应阈值(Rigrsure、Minimaxi、Rigsure和Sqtwolog)模型等模型的去噪效果.同时与Savitzky-Golay卷积平滑和快速傅里叶变换的去噪效果进行比较.研究表明,Birge-Massart阈值模型的综合小波去噪效果最好,同时小波去噪的方法较Savitzky-Golay卷积平滑和快速傅里叶变换去噪可以更好地捕获光谱的尖峰特征.     

多尺度有限差分法模拟复杂介质波传问题

利用有限差分和紧支集正交小波变换对波动方程的时间、空间进行联合近似求解,提出一种适合于一般边界非均匀耗散介质中波传问题数值模拟的快速自适应混合算法——多尺度有限差分(multiresolution finite difference缩写为MRFD)方法.将波传问题的求解转换到小波域中进行,利用小波基的自适应性与消失矩特性,使偏微分算子矩阵稀疏化,有效改善了计算量等.地球物理勘探中的数值实例显示了算法具有良好效率 关键词： MRFD 小波变换 多尺度 波传问题     

基于双边滤波和双树复小波的图像去噪算法

提出了基于空域双边滤波和双树复小波变换的图像去噪算法.该算法使用双树复小波变换对含噪图像进行多尺度和多方向的分解,对各个高频方向子带使用带有方向窗的局部维纳滤波算法进行去噪.在重构过程中,对每一个尺度上重构得到的低通图像使用空域的双边滤波算法进一步的去除噪声.实验结果表明本文提出的图像去噪算法获得了明显的去噪性能改善.     

基于双边滤波和双树复小波的图像去噪算法

提出了基于空域双边滤波和双树复小波变换的图像去噪算法.该算法使用双树复小波变换对含噪图像进行多尺度和多方向的分解,对各个高频方向子带使用带有方向窗的局部维纳滤波算法进行去噪.在重构过程中,对每一个尺度上重构得到的低通图像使用空域的双边滤波算法进一步的去除噪声.实验结果表明本文提出的图像去噪算法获得了明显的去噪性能改善.     

结合边缘信息和图像特征信息的曲波域遥感图像融合

曲波变换是一种更适合于图像处理的多尺度几何分析方法,具有比小波变换更强的方向选择和辨识能力,而且对图像边缘的表达更优于小波.结合色度-饱合度-亮度变换将其应用于合成孔径雷达图像和多光谱图像融合可以更好地表示图像中的有用特征.首先对多光谱图像进行色度-饱合度-亮度变换,得到亮度分量Ⅰ,对雷达图像和Ⅰ分量进行曲波变换得到粗尺度系数和细节尺度系数;将雷达图像的粗尺度系数和细节尺度系数进行叠加,计算归一化的曲波系数直方图,定义边缘有效因子,利用合成孔径雷达图像的特征信息将曲波变换系数分为均匀区、非均匀区和亮点目标区.然后采用相应的融合规则对融合图像的粗尺度系数进行处理,对细节尺度系数采用简单的直接取大方法,逆变换后得到新的亮度分量.用新的亮度分量替代原亮度分量进行逆色度-饱合度-亮度变换得到最终融合结果,利用统计类指标对融合结果进行评价.实验结果表明,该方法在保持光谱信息和提高空间分辨率上都有较好的效果.     

结合边缘信息和图像特征信息的曲波域遥感图像融合

曲波变换是一种更适合于图像处理的多尺度几何分析方法,具有比小波变换更强的方向选择和辨识能力,而且对图像边缘的表达更优于小波.结合色度-饱合度-亮度变换将其应用于合成孔径雷达图像和多光谱图像融合可以更好地表示图像中的有用特征.首先对多光谱图像进行色度-饱合度-亮度变换,得到亮度分量I,对雷达图像和I分量进行曲波变换得到粗尺度系数和细节尺度系数;将雷达图像的粗尺度系数和细节尺度系数进行叠加,计算归一化的曲波系数直方图,定义边缘有效因子,利用合成孔径雷达图像的特征信息将曲波变换系数分为均匀区、非均匀区和亮点目标区.然后采用相应的融合规则对融合图像的粗尺度系数进行处理,对细节尺度系数采用简单的直接取大方法,逆变换后得到新的亮度分量.用新的亮度分量替代原亮度分量进行逆色度-饱合度-亮度变换得到最终融合结果,利用统计类指标对融合结果进行评价.实验结果表明,该方法在保持光谱信息和提高空间分辨率上都有较好的效果.     

窄频带Lamb波频散特性研究

利用连续小波变换对从铝合金板结构中俘获的Lamb波信号进行分析,获得波信号在时间-尺度域的等高线和等高线脊线.根据Lamb波的频散特征、时间-尺度域等高线脊线的斜率和波在不同尺度下的到达时间,识别了Lamb波信号中各信息包的模式,并匹配出基础阶模式窄频带Lamb波在铝合金板结构中传播的实际频散曲线.对试验的Lamb波信号分析的结果表明该方法对于研究和应用窄频带Lamb波的频散特性是有效的. 关键词： 连续小波变换 Lamb波 频散     

小波变换软阈值去噪在粗晶材料超声检测中的应用

本研究多尺度小波分析方法在粗晶材料超声检测信号增强中的应用。在分析结构噪声，电噪声和缺陷信号的小波变换特性的基础上，提出一种用一个尺度间变化的门限闭值来抑制噪声的小波变换系数，并以各尺度缺陷信号的能量关系形成权值，加权重构信号来提高信噪比。实验结果表明，该方法具有很好的去噪效果，增强了粗晶材料缺陷的检出率。     

基于小波变换及多尺度分析的星点检测方法

为提高从包含复杂天空背景的红外图像中检测出星点目标的能力,提出了一种基于小波变换及多尺度分析的星点检测算法。在对图像进行小波变换的基础上,利用多尺度分析方法,确定图像自适应阈值,实现图像小波域的去噪,达到星点目标检测的目的。实验表明,该方法能有效地检测出红外图像中的星点目标,检测能力较普通背景预测方法提高近40%。     

Quantitative temperature field measurements on a non-gray multi-materials scene by thermoreflectometry

This article addresses the problem of measuring an accurate temperature field on a multi-materials scene composed of two dielectric and one metallic materials. The measurements of thermo-radiative properties demonstrate that the scene exhibit very different emissivity spectra and thermal conductivities inducing high thermal gradients. From these radiative properties, the calculation of the theoretical temperature error of conventional passive methods highlights that a method may be suitable for measuring only one material but no method provides satisfactory measurements of the sets of materials. The proposed method, called thermoreflectometry, performs a simultaneous measurement on all materials thanks to an on-line indirect determination of emissivity based on a bidirectional reflectivity measurement. Its temperature error is compared to that of the selected passive method for each material through an experimental validation on the multi-materials scene. These results highlight the accuracy of thermoreflectometry and shows opening prospects for the on-line temperature control of dynamical processes.     

介孔二氧化硅基导电聚合物复合材料热导率的实验研究

本文首先制备并表征了介孔二氧化硅SBA-15、 填充导电聚合物的复合材料PANI/SBA-15和复合材料PPy/SBA-15, 并建立双流计实验台开展了材料压片情况下的热导率研究. 在测量得到复合材料热导率的基础上, 引入当量孔径, 结合测量孔径对 PANI/SBA-15和PPy/SBA-15复合材料热导率随填充量的变化进行了定性分析. 分析表明: PANI/SBA-15和PPy/SBA-15复合材料的热导率比基材SBA-15的热导率大得多; 在相同的测量孔径和当量孔径情况下, PANI/SBA-15复合材料的热导率比PPy/SBA-15复合材料的热导率大; 导电聚合物填充到复合材料孔道内和孔道外都有助于热导率的提高, 填充到孔道内比填充到孔道外对热导率提高的贡献更大.     

An Overview of Infrared Spectroscopy Based on Continuous Wavelet Transform Combined with Machine Learning Algorithms: Application to Chinese Medicines,Plant Classification,and Cancer Diagnosis

Abstract

Infrared spectroscopy has been a workhorse technique for materials analysis and can result in positively identifying many different types of material. In recent years there have been reports using wavelet analysis and machine learning algorithms to extract features of Fourier transform infrared spectrometry (FTIR). The machine learning algorithms contain back-propagation neural network (BPNN), radial basis function neural network (RBFNN), and support vector machine (SVM). This article reviews the important advances in FTIR analysis employing a continuous wavelet transform (CWT) and machine learning algorithms, especially in the applications of the method for Chinese medicine identification, plant classification, and cancer diagnosis.     

Multi-scale bending,buckling and vibration analyses of carbon fiber/carbon nanotube-reinforced polymer nanocomposite plates with various shapes

Using a finite element-based multi-scale modeling approach, the bending, buckling and free vibration of hybrid polymer matrix composites reinforced by carbon fibers and carbon nanotubes (CF/CNT-RP) are analyzed herein. Thick composite plates with rectangular, circular, annular and elliptical shapes are considered. First, the equivalent material properties of CF/CNT-RP are calculated for different volume fractions of CF and CNT. To accomplish this aim, a two-step procedure is presented through which the coupled effects of nano- and micro-scale are taken into account. In the first step, modeling of dispersion of CNTs into the polymer matrix is done with considering interphase formed by their chemical interaction with the matrix, and the equivalent properties of resulting composite material are determined accordingly. CFs are then dispersed into CNT-RP which is considered a homogenous material in this step. Both distributions of CNTs and CFs are assumed to be random. After computing the equivalent properties of CF/CNT-RP for different volume fractions of its constituents, the bending, buckling and free vibration analyses of plates with different shapes are performed. It is shown that the reinforcement of the polymer matrix with both CF and CNT significantly affects the bending, buckling and free vibration characteristics of plates.     

Machine learning approach for the prediction and optimization of thermal transport properties

Traditional simulation methods have made prominent progress in aiding experiments for understanding thermal transport properties of materials, and in predicting thermal conductivity of novel materials. However, huge challenges are also encountered when exploring complex material systems, such as formidable computational costs. As a rising computational method, machine learning has a lot to offer in this regard, not only in speeding up the searching and optimization process, but also in providing novel perspectives. In this work, we review the state-of-the-art studies on material’s thermal properties based on machine learning technique. First, the basic principles of machine learning method are introduced. We then review applications of machine learning technique in the prediction and optimization of material’s thermal properties, including thermal conductivity and interfacial thermal resistance. Finally, an outlook is provided for the future studies.     

Accurate determination of anisotropic thermal conductivity for ultrathin composite film

Highly anisotropic thermal conductive materials are of significance in thermal management applications. However, accurate determination of ultrathin composite thermal properties is a daunting task due to the tiny thermal conductance, severely hindering the further exploration of novel efficient thermal management materials, especially for size-confined environments. In this work, by utilizing a hybrid measuring method, we demonstrate an accurate determination of thermal properties for montmorillonite/reduced graphene oxide (MMT/rGO) composite film with a thickness range from 0.2 μ m to 2 μ m. The in-plane thermal conductivity measurement is realized by one-dimensional (1D) steady-state heat conduction approach while the cross-plane one is achieved via a modified 3ω method. As-measured thermal conductivity results are cross-checked with different methods and known materials, revealing the high measurement accuracy. A high anisotropic ratio of 60.5, independent of composite thickness, is observed in our measurements, further ensuring the negligible measurement error. Notably, our work develops an effective approach to the determination of ultrathin composite thermal conductivity, which may promote the development of ultrathin composites for potential thermal-related applications.     

周期孔洞区域中热力耦合问题的双尺度有限元计算

复合材料的研究中经常遇到具有周期孔洞结构的材料,由于区域的小周期性及剧烈振荡性,用传统的有限元计算方法来计算这些材料对应的问题时需要大量的计算机存储空间及计算时间.对这类材料的热力耦合问题给出了一种新型的高阶双尺度渐近解,得到了对应的均匀化常数、均匀化方程及对应的有限元算法.数值算例表明,周期单胞的局部结构对局部应力与应变有较大的影响.算法对数值模拟这类材料的力学行为是高效和可行的. 关键词： 双尺度方法 热力耦合 周期孔洞区域 有限元方法     

Metal Matrix Composites for Thermal Management: A Review

The most challenging objective in the electronic industries is to develop materials that demonstrate a tunable thermal property with today's microelectronic devices. The development of composite material with balanced thermal properties is highly appreciated and currently competing the traditional monolithic conductive material. However, the tailored thermal properties of the composite are significantly influenced by the composites constituents and their fabrication routes. This article presents a review of thermal properties of particulate as well as fiber-reinforced composite proportional to matrix microstructure, reinforcement architecture. The processing techniques used to fabricate composites have been addressed with an objective to achieve suitable thermal properties. The developments in the analytical and numerical simulation approach to predict the thermal conductivity and CTE of the developed composites have been critically reviewed. Lastly, future work needs attention is summarized.     

一种基于小波系数综合能量特征的多算子图像融合算法

提出了一种新的多算子小波分解图像融合算法,算法对输入图像进行多尺度小波分解,综合考虑同层各子带及相邻层子带小波系数图像特征描述的相关一致性,基于局部空间复合能量和局部相对能量差特征测度,采用多算子自适应融合规则构造融合图像,得到含有丰富细节特征的融合图像。     

Synthesis and characteristics of form-stable n-octadecane/expanded graphite composite phase change materials

N-octadecane/expanded graphite composite phase-change materials were prepared by absorbing liquid n-octadecane into the expanded graphite. The n-octadecane was used as the phase-change material for thermal energy storage, and the expanded graphite acted as the supporting material. Fourier transformation infrared spectroscopy (FT-IR), X-ray diffraction (XRD), scanning electron microscopy (SEM) and thermal diffusivity measurement were used to determine the chemical structure, crystalline phase, microstructure and thermal diffusivity of the composite phase-change materials, respectively. The thermal properties and thermal stability were investigated by differential scanning calorimetry (DSC) and thermogravimetric analysis (TGA). The DSC results indicated that the composite phase-change materials exhibited the same phase-transition characteristics as the n-octadecane and their latent heat increased with the n-octadecane content in composite phase-change materials. The SEM results showed that the n-octadecane was well absorbed in the porous network of the expanded graphite, and there was no leakage of the n-octadecane from the composites even when it was in the molten state.