Papers by Keyword: Fractal

Abstract: Fractal is a science which studies very irregular and complicated natural shapes. It has been used widely in image processing. Firstly this thesis analyses the reasons that fractal can be used in 3D (Three Dimensions) terrain surface reconstruction. Then, it introduces fractional Brownian motion (fBm). On this basis, the thesis presents an interpolating algorithm to reconstruct 3D terrain surface. The algorithm uses fBm model to interpolate. It can be achieved simply and a good result is obtained.

Abstract: The fractal dimension of a large group of nano-silica, alumina, titanium oxide, and Gd₂O₃ and Y₂O₃ was calculated by means of original program. Nanopowders of used oxides are created in two main ways, such as evaporation substances by relativistic electron beam and high temperature hydrolysis. The fractal dimension was calculated upon transmission electron microscopy images obtained with different magnification. It was managed to establish that it's possible to make a comparative assessment of the degree of hydrophilicity of the nano-silica's samples by means of attitude fractal dimension to the specific surface.

Abstract: A kind of special energy aggregation structure is analyzed. The special structure can gather electromagnetic energy. The paper summarizes features of the structure and indicates that the structure is a fractal. By analyzing the fractal characteristics, deduces the math iterative that can produce the parameter matrix of the structure. Mathematical model is done for this special energy aggregation structure. A special energy aggregation structure is designed by the mathematical model. The experimental results proved that the sample has functions of gathering electromagnetic energy. The results prove that the mathematical model for the special energy aggregation structure is built correctly and effectively. The paper provides the basis of mathematical analysis for researching on energy aggregation and transmission mechanism of this structure.

Abstract: In order to improve safety, economy and reliability of NPP(nuclear power plant) and to make the best of human, material and information resources in the NPP, a specific function model developing method for integrated managing system based on CIMS and fractal conception is discussed in this paper. Under the guidance of general requirement, the developing process is deepened from integer to local, from macrocosm to microcosm. Furthermore, under the conditions of modularization, high inter-aggregation and low coupling, the developing process has the characteristics of layered detailing of the complexity degree.

Abstract: This paper extends the polynomial function to double logarithmic function, constructing a class of multi parameters iterative function, and uses this function to calculate the fractal interpolated surface for given interpolation points, and establishes the iterative function mathematical model of multi parameters fractal interpolation. In order to verify the effectiveness and reliability of this proposed model algorithm, this paper uses MATLAB numerical simulation method to calculate, and programs the Newton iterative function of multi parameters fractal interpolation surface, finally gets calculation nephogram of multi parameters fractal interpolation curved surface through calculating. Finally, using iterative method reduces the surface grid size, increasing the smoothness of the surface, so the surface is closer to the actual surface. It provides a new computer method research of fractal interpolation function.

Abstract: The stirred tank is widely used in many industries to obtain the desired type of fluid mixing. In the context of mixing process, two different fluids and have a different properties will mix in a single equipment to produce another fluid with a new property. In this research, a new approach of stirred tank which is containing a new design of baffles and impellers was proposed for fluid mixing. The new design of baffles and impellers that proposed here are used a fractal pattern for both parts in the stirred tank. Implementing a fractal pattern for baffles and impellers in stirred tank believe will influence the flow characteristic inside the stirred tank, hence will improve a mixing performance. In order to investigate the kinds of flow properties, a Particle Image Velocimetry (PIV) technique with 1 μm seeding particle was used. Four configurations were tested which are normal baffles and normal impellers, normal baffles and fractal impellers, fractal baffles and normal impellers, and the last configuration is fractal baffles and fractal impellers. In this study, dual Rushton impellers with 4 blades were used with the configurations mentioned. The result shows the significant flow field capture by PIV measurement on each configuration. By using fractal impeller some vortex are shown in the tank and high velocity vector on flow field compare with normal impeller while normal baffles gives high velocity vector depends on the configuration were used. From the results, it was showed that the fractal design can give a certain level of mixing efficiency in stirred tank. The PIV technique also gives good flow visualization in order to determine the flow pattern in stirred tank with a new concept of baffles and impellers.

Abstract: Fractal grids can be used to design turbulent flows with low power losses and high turbulence intensities for intense yet economic mixing over a region of designed length and location. The introducing circle grid perforated plate is the main aim of this present paper. In this numerical work, we want to ascertain a new approach in turbulence generators on the structure of premixes flames using perforated plate fractal-grids pattern. In this paper, we compared circle grid perforated plate by varies using its diameters, spacing and number of holes on the plate. The result showed good perceptivity of the fractal generated turbulence and the fractal flow physics. The turbulent intensity can be increased by a grid with higher blockage ratio.

Abstract: The classic problem of the existence of fractal coding time is too long, a kind of fast encoding algorithm was proposed in this paper, which is based on Wavelet and Fractal combined, using wavelet decomposition characteristics. This method reduces the amount of image data compression effectively, shorts coding time and improve the image encoding quality.

Abstract: Typical tight gas reservoirs in the Su53 Gas Field, Ordos Basin, are the demonstration region for deploying horizontal wells. In order to provide foundation for evaluating the volume fracturing adaptability of horizontal wells, through using the amplitude difference data between deep investigate double lateral resistivity (Rd) and shallow investigate double lateral resistivity (Rs), and other conventional logging data, in combination with the response characteristics of fracture in the logging curve, reservoir fracture was quantitatively identified with the variable scale fractal technique, the vertical heterogeneity of fracture distribution was analyzed, and the relationship between fractal dimension value of fracture and initial deliverability of gas well was researched. The results indicate that, this method is applied to quantitatively identify and evaluate the fracture development degree of single-well is feasible; the relationship between fractal dimension value and fracture development degree is positively relative, fractal dimension value is bigger, the fracture is more developed; there is a good corresponding relationship between fractal dimension value and deliverability of gas well, fractal dimension value decreases with a decrease of deliverability.

Abstract: To get better solution by improving the mutation strategy of Differential Evolution algorithm, a fractal mutation strategy is introduced. The fractal mutation factor of the proposed Fractal Mutation factor Differential Evolution (FMDE) algorithm is simulated by fractal Brownian motion with a different Hurst index. The new algorithm is test on 25 benchmark functions presented at 2005 IEEE Congress on Evolutionary Computation (CEC2005). The optimization results of at least 10 benchmark functions are significantly better than the results obtained by JADE and CoDE, and most of the rest of the test results are approximate. This shows that FMDE can significantly improve the accuracy and adaptability to solve optimization problems.