Fractals

Fractal model provides an alternative and useful means for studying the transport phenomenon in porous media and analyzing the macroscopic transport properties of porous media, as fractal geometry can successfully characterize disordered and heterogeneous geometrical microstructures of porous media on multi scales. Recently, fractal models on porous media have attracted increasing interests from many different disciplines. In this mini-review paper, a review on fractal models for number-size distribution in porous media is made, and a unified fractal model to characterize pore and particle size distributions is proposed according to the statistical fractal property of the complex microstructure in porous media. Using the fractal scaling laws for pore and fracture size distributions, a fractal capillary bundle model and a fractal tree-like network model are presented and summarized for homogenous and fractured porous media, respectively. And the applications of the fractal capillary bundle model and fractal tree-like network model for analysis of transport physics in porous media are also reviewed.

A new method termed as three-dimensional root-mean-square (3D-RMS) method, is proposed to calculate the fractal dimension (FD) of machined surfaces. The measure of this method is the root-mean-square value of surface data, and the scale is the side length of square in the projection plane. In order to evaluate the calculation accuracy of the proposed method, the isotropic surfaces with deterministic FD are generated based on the fractional Brownian function and Weierstrass–Mandelbrot (WM) fractal function, and two kinds of anisotropic surfaces are generated by stretching or rotating a WM fractal curve. Their FDs are estimated by the proposed method, as well as differential boxing-counting (DBC) method, triangular prism surface area (TPSA) method and variation method (VM). The results show that the 3D-RMS method performs better than the other methods with a lower relative error for both isotropic and anisotropic surfaces, especially for the surfaces with dimensions higher than 2.5, since the relative error between the estimated value and its theoretical value decreases with theoretical FD. Finally, the electrodeposited surface, end-turning surface and grinding surface are chosen as examples to illustrate the application of 3D-RMS method on the real machined surfaces. This method gives a new way to accurately calculate the FD from the surface topographic data.

The box counting algorithm is a well-known method for the computation of the fractal dimension of an image. It is often implemented using a recursive subdivision of the image into a set of regular tiles or boxes. Parallel implementations often try to map the boxes to different compute units, and combine the results to get the total number of boxes intersecting a shape. This paper presents a novel and highly efficient method using Open Computing Language (OpenCL) kernels to perform the computation on a per-pixel basis. The mapping and reduction stages are performed in a single pass, and therefore require the enqueuing of only a single kernel. Each instance of the kernel updates the information pertaining to all the boxes containing the pixel, and simultaneously increments the box counters at multiple levels, thereby eliminating the need for another pass to perform the summation. The complete implementation and coding details of the proposed method are outlined. The performance of the method on different processors are analyzed with respect to varying image sizes.

Animal models represent the basis of our current understanding of the pathophysiology of asthma and are of central importance in the preclinical development of drug therapies. The characterization of irregular lung shapes is a major issue in radiological imaging of mice in these models. The aim of this study was to find out whether differences in lung morphology can be described by fractal geometry. Healthy and asthmatic mouse groups, before and after an acute asthma attack induced by methacholine, were studied. In vivo flat-panel-based high-resolution Computed Tomography (CT) was used for mice's thorax imaging. The digital image data of the mice's lungs were segmented from the surrounding tissue. After that, the lungs were divided by image gray-level thresholds into two additional subsets. One subset contained basically the air transporting bronchial system. The other subset corresponds mainly to the blood vessel system. We estimated the fractal dimension of all sets of the different mouse groups using the mass radius relation (mrr). We found that the air transporting subset of the bronchial lung tissue enables a complete and significant differentiation between all four mouse groups (mean D of control mice before methacholine treatment: 2.64 ± 0.06; after treatment: 2.76 ± 0.03; asthma mice before methacholine treatment: 2.37 ± 0.16; after treatment: 2.71 ± 0.03; p < 0.05). We conclude that the concept of fractal geometry allows a well-defined, quantitative numerical and objective differentiation of lung shapes — applicable most likely also in human asthma diagnostics.

Based on the fractal game which is performed by the majority and the minority, the fractal market theory (FMT) is employed to describe the features of investors' decision-making. Accordingly, the process of fractal games is formed in order to analyze the statistical features of conversion between momentum and contrarian. The result shows that among three fractal game mechanisms, the statistical feature of simulated return rate series is much more similar to log returns on actual series. In addition, the conversion between momentum and contrarian is also extremely similar to real situation, which can reflect the effectiveness of using fractal game in analyzing the conversion between momentum and contrarian. Moreover, it also provides decision-making reference which helps investors develop effective investment strategy.

The entropy production and the variational functional of a Laplacian diffusional field around the first four fractal iterations of a linear self-similar tree (von Koch curve) is studied analytically and detailed predictions are stated. In a next stage, these predictions are confronted with results from numerical resolution of the Laplace equation by means of Finite Elements computations. After a brief review of the existing results, the range of distances near the geometric irregularity, the so-called "Near Field", a situation never studied in the past, is treated exhaustively. We notice here that in the Near Field, the usual notion of the active zone approximation introduced by Sapoval et al. [M. Filoche and B. Sapoval, Transfer across random versus deterministic fractal interfaces, Phys. Rev. Lett. 84(25) (2000) 5776;¹ B. Sapoval, M. Filoche, K. Karamanos and R. Brizzi, Can one hear the shape of an electrode? I. Numerical study of the active zone in Laplacian transfer, Eur. Phys. J. B. Condens. Matter Complex Syst. 9(4) (1999) 739-753.]² is strictly inapplicable. The basic new result is that the validity of the active-zone approximation based on irreversible thermodynamics is confirmed in this limit, and this implies a new interpretation of this notion for Laplacian diffusional fields.

An exploratory study is made on the dynamics of complex maps endowed with delay type memory of past states. Particular attention is paid to the quadratic map. Other maps, i.e. the cubic map, the map induced by the Newton–Raphson numerical method, and the purely real logistic map are also scrutinized.

A model for the yield stress of aggregates is presented that incorporates fractal dimension taking into account the solid volume fraction and the aggregate diameter. The model shows the yield stress (σ_y) of aggregates increases with the solid volume fraction (ϕ_s) as a power law, and is given by , where the exponent (m) is related to fractal dimension (D), and σ_y0 is a referenced parameter. The relationship between exponent (m) and fractal dimension is validated by published data of aggregates and represents the measured data very well, over a wide range of the solid volume fractions. The referenced parameter (σ_y0) is calibrated from experiments of yield stress using power law fittings. The agreement between theory and experiments supports the idea that yielding is ultimately caused by the rupture of a few interparticle bonds within aggregates. In addition, the proposed model for the yield stress of aggregates is found to match better with experiments by comparing with all models in literature.

In this study, the analytical expressions for the hydraulic permeability and Kozeny–Carman (KC) constant of porous nanofibers are derived based on fractal theory. In the present approach, the permeability is explicitly related to the porosity and the area fractal dimensions of porous nanofibers. The proposed fractal models for KC constant is also found to be a function of the microstructural parameters (porosity, area fractal dimensions). Besides, the present model clearly indicates that KC constant is not a constant and increases with porosity. However, KC constant is close to a constant value which is 18 for ϕ > 0.8. Every parameter of the proposed formulas of calculating permeability and KC constant has clear physical meaning. The model predictions are compared with the existing experimental data, and fair agreement between the model predictions and experimental data is found for different porosities.

We introduce weighted tetrahedron Koch networks with infinite weight factors, which are generalization of finite ones. The term of weighted time is firstly defined in this literature. The mean weighted first-passing time (MWFPT) and the average weighted receiving time (AWRT) are defined by weighted time accordingly. We study the AWRT with weight-dependent walk. Results show that the AWRT for a nontrivial weight factor sequence grows sublinearly with the network order. To investigate the reason of sublinearity, the average receiving time (ART) for four cases are discussed.

We study the asymptotic behavior of the aggregated fractional Ornstein–Uhlenbeck processes with random coefficients that follows a Gamma-type distribution. We also analyze the properties of the limit process.

In this paper, first we equip the automorphism group of the p-ary rooted tree X* with a natural metric and define a family of contractions on Aut(X*). Then, we construct an iterated function system (IFS) whose attractor is the closure of the adding machine group on Aut(X*). Finally, we show that this group is a strong self-similar group in the sense of IFS.

The detrending moving average (DMA) algorithm is one of the best performing methods to quantify the long-term correlations in nonstationary time series. As many long-term correlated time series in real systems contain various trends, we investigate the effects of polynomial trends on the scaling behaviors and the performances of three widely used DMA methods including backward algorithm (BDMA), centered algorithm (CDMA) and forward algorithm (FDMA). We derive a general framework for polynomial trends and obtain analytical results for constant shifts and linear trends. We find that the behavior of the CDMA method is not influenced by constant shifts. In contrast, linear trends cause a crossover in the CDMA fluctuation functions. We also find that constant shifts and linear trends cause crossovers in the fluctuation functions obtained from the BDMA and FDMA methods. When a crossover exists, the scaling behavior at small scales comes from the intrinsic time series while that at large scales is dominated by the constant shifts or linear trends. We also derive analytically the expressions of crossover scales and show that the crossover scale depends on the strength of the polynomial trends, the Hurst index, and in some cases (linear trends for BDMA and FDMA) the length of the time series. In all cases, the BDMA and the FDMA behave almost the same under the influence of constant shifts or linear trends. Extensive numerical experiments confirm excellently the analytical derivations. We conclude that the CDMA method outperforms the BDMA and FDMA methods in the presence of polynomial trends.