In the first edition, models initially developed to describe wave propagation in porous media saturated by heavy fluids are used to predict the acoustical performances of air saturated sound absorbing porous media. In this expanded and revised edition, we have retained, with slight modifications, most of the basic material of the first edition and expanded it by revisiting several original topics and adding new topics to integrate recent developments in the domain of wave propagation in porous media and practical numerical prediction methods that are widley used by researchers and engineers. Chapters 1 to 3 dealing with sound propagation in solids and fluid and Chapter 9 dealing with the modelling of perforated facings were slightly modified. Chapters 4 to 6 were greatly revisited. A more detailed description of sound propagation in cylindrical pores is presented (Chapter 4), related to the more general presentation of new parameters and new models for sound propagation in rigid-framed porous media (Chapter 5). Also in Chapter 5 a short presentation of homogenization, with some results concerning double porosity media, is added. In Chapter 6, different formulations of the Biot theory for poroelastic media are given, with a simplified version for the case of media with a limp frame. In Chapter 11 we have revisited the original representation of the modelling of layered media (Chapter 7 of the first edition) and extended it to cover the systematic modelling of layered media using the Transfer Matrix Method (TMM). In particular, a step by step presentation of the numerical implementation of the method is given with several application examples. Major additions include five new chapters. Chapter 7 discusses the acoustic field created by a point source above a rigid framed porous layer, with recent advances concerning the poles of the reflection coefficient and the reflected field at grazing incidence. Chapter 8 is concerned by the poroelastic layers excited by a point source in air or by a localized stress source on the free face of the layer, with a description of the Rayleigh waves and the resonances. Axisymmetrical poroelastic media are studied in Chapter 10. In Chapter 12, complements to the transfer matrix method are given. They concern mainly the effect of the finite lateral extend, and the excitation by point loads, of sound pack- ages. Several examples illustrating the practical importance of these extensions are given (e.g. size effects on the random incidence absorption and transmission loss of porous media; airborne vs. structure borne insertion loss of sound packages). In Chapter 13, an introduction to the finite element modelling of poroelastic media is presented. Emphasis is put on the use of the mixed displacement-pressure formulation of the Biot theory, xiv PREFACE TO THE SECOND EDITION which appears in the Appendix of Chap. 6. Detailed description of coupling conditions between various domains including a waveguide are presented together with sections on the breakdown of the power dissipation mechanisms within a porous media and radiation conditions. Several applications are chosen to illustrate the practical use of the presented methods including modelling of double porosity materials and smart foams. As in the first edition, the goal of the book remains to provide in a concrete manner a physical basis, as simple as possible, and the developments, analytical calculations and numerical methods, that will be useful in different fields where sound absorption and transmission and vibration damping by air saturated porous media are concerned.

一起 k 均值 matlab 代码动态丛传播 该存储库包含动态丛传播算法的 Matlab 实现。 该算法最初是在 (Viles 2013) 中开发的，它提供了一种机制，用于在不确定的情况下随着时间的推移识别和跟踪功能网络中的社区。 这与导致癫痫发作的大脑连接的处理（例如，由 ECoG 测量）具有特定的相关性。 受 (Palla 2005) 中的 clique percolation 方法的启发，该算法在函数网络的每个时间步识别 k-plex（使用 k-plex 使算法对噪声具有鲁棒性）并将它们在每个时间步内和跨时间步骤。 因此，这允许随着时间的推移和在不确定性下跟踪功能社区的诞生和消亡。 这个 Matlab 实现是算法的基本实现，以及一些相关的工具： 模拟动态网络 汇总统计信息 可视化动态网络 这仍然是一项正在进行的工作。 本文档和存储库将使用有关如何使用算法的文档和示例进行更新。 运行算法 为了运行算法，你必须有一个无向二元图的动态邻接矩阵。 给定n个顶点和t时间步长，动态邻接矩阵C应该是n × n × t 。 矩阵中的每个值表示在特定时间两个顶点之间是否存在边。 例如C(a, b,

用有限差分模拟高斯脉冲在自由空间传播，运行该脚本，你会得到一个表面，它是由脉冲在1微米的步骤传播。

利用BP 神经网络来预测汽油中 辛烷值的成分，有详细的注释

MATLAB模拟粒子散射代码光束传播法（BPM）的大规模全息粒子3D成像 MatLab实现文件。 我们提供了光束传播方法的正向模型和重建算法代码，模拟的样本对象，全息图，重建的粒子以及样本实验捕获的全息图，重建结果。 引文 如果您发现该项目对您的研究有用，请考虑引用我们的论文： 抽象的 我们为3D粒子场的大规模全息重建开发了一种新颖的算法。 我们的方法基于结合稀疏正则化的多散射光束传播方法（BPM），该方法可从单个全息图中恢复高折射率对比度的致密3D粒子。 我们显示，BPM计算的全息图生成的强度统计数据与实验测量值非常匹配，并且比单散射模型提供高达9倍的精度。 为了解决反问题，我们设计了一种计算效率高的算法，与基于最新技术的基于多重散射的技术相比，该算法将计算时间减少了两个数量级。 我们在不同散射强度下的仿真和实验中均展示了卓越的重建精度。 我们表明，对于深成像深度和高粒子密度，BPM重建显着优于单散射方法。 概述图 如何使用程式码 正向模型： 重建： 数据 dz = lambda / 16的模拟对象：object / simulatedData / density_1.6 重建对象的

General perturbations element sets generated by NORAD can be used to predict position and velocity of Earth-orbiting objects. To do this one must be careful to use a prediction method which is compatible with the way in which the elements were generated. Equations for ¯ve compatible models are given here along with corresponding FORTRAN IV computer code. With this information a user will be able to make satellite predictions which are completely compatible with NORAD predictions.

高速数字设计经典教材，电子版

基于随机游走的社团发现算法Hadoop版 以及一个graph生成程序。整个是个eclipse项目，没有把lib放上来。内容在 http://blog.csdn.net/lgnlgn/article/details/6561876 的下一篇博客

贝叶斯网络上的信念传播 这是一个在贝叶斯网络 (BN) 上运行循环信念传播的程序，并为网络上的每个节点生成边缘化概率。 算法细节参考 ，随着使用bethe聚类图而不是BN的纯因子图的变化。 输入格式应为 用法 $ python bp < .bif file path > [-o output file] [-t threshold] Options: -o, --output output file name, default to ' result.txt ' -t, --threshold threshold for convergence default to 1e-10 致谢 BIF 解析器由提供。 项目中的代码用于华盛顿大学的作业 3。

信仰传播 该存储库包含有关信念传播的项目的代码，作为“图形模型：离散推理和学习”课程的一部分（主MVA）。 我们进行了两个实验来测试信念传播的两种实现： （湿草示例） （分类归纳） 项目演示的幻灯片在，而项目报告在共享。 作者：Charbel-RaphaëlSégerie，克莱门特·邦内（ClémentBonnet）。 2021年3月31日