Papers by Keyword: Random Distribution

Abstract: A three dimensional analysis model is developed on the fatigue life prediction of composite laminates based on a progressive damage analysis. This model consists of stress analysis, fatigue failure analysis and material property degradation. Teserpe’s failure criteria is used to fatigue damage analysis. Fiber tensile/compressive breakage, matrix tensile/compressive cracking, matrix/fiber shear failure and tension/compression delamination are considered in fatigue damage analysis. The methodologies of sudden degradation and gradual degradation are both applied in the material property degradation. The stiffness and strength gradual degradation is based on the Shokrieh fatigue model, which is based on fatigue test for unidirectional laminates. In order to consider the scatter of the material in the practical structures, the stiffness and strength of the material are randomly distributed using normal distribution in the numerical model. The progressive fatigue damage model is developed in finite element code ABAQUS through user subroutine UMAT, which can simulate the fatigue damage process. Fatigue life of different ply stacking sequences and geometries composite laminates under different cycle loading are predicted. The predicted fatigue life is in good agreement with the experimental results.

Abstract: This study presents a statistical two-scale method to predict the viscoelastic properties of composite materials with consistent random distribution of particles. The explicit formulation for predicting the effective viscoelastic relaxation modulus is given. At first, the Laplace transformation is used to the linear viscoelastic problem, the effective generalized relaxation modulus in Laplace domain for composites is derived. Then, the effective relaxation modulus in time domain is obtained by the least-square and inverse Laplace transformation. At the end of this paper, some numerical examples are given to validate that the presented method is feasible and effective.

Abstract: In this paper we use the wireless Android smart network to control data collection network of wireless sensor networks, and establish the wireless sensor network data acquisition model based on particle swarm layout algorithm, and design the data signal receiving algorithm by introducing wireless fence monitoring method. In order to verify the reliability and applicability of the algorithm, we compare the particle swarm layout algorithm with PSO, random distribution and genetic algorithm. Through experiments, data gathering integrity of particle swarm layout algorithm reaches 96%, the time of data acquisition is 1258 seconds, the iterative number is 12587 times. It is the best of the three algorithms. It provides a theoretical reference for the research on data acquisition and control technology of wireless sensor network.

Abstract: In this paper, we propose a Density-based Energy-Efficient Unequal Clustering (DEEUC) algorithm to solve the energy consumption imbalance problem of the nodes deployed randomly. The algorithm balances the energy consumption of the cluster by making nodes as cluster head in turn and taking the remaining energy into consideration, which makes the generation of the cluster head more reasonable. It balances the energy consumption of the cluster by generating unequal cluster with different radius based on nodes density and the distance to the sink. In the routing strategy, the cluster head node with high density is chosen as the next hop, which makes the selected node undertake more data forwarding tasks, and balances the regional energy consumption. The simulation experiments show that this algorithm can effectively balance the network energy consumption and prolong the network lifetime.

Abstract: DDA(Discontinuous Deformation Analysis) was used to study the mechanical characteristics of coarse granular material (CGM). The numerical simulation model was found consulting the two Dimensional test of CGM,and the stress-strain curves obtained by the numerical simulation were basically in agreement with the results of laboratory test. It indicates that the numerical simulation is suitable for the mechanical characteristic research of CGM. In addition,the distribution statistic of several fabric elements about interaction between particles was also presented. Some relation between the stress-strain and fabric element change are found after analysis.

Abstract: In this paper, Representative Volume Element with random distribution pattern has been built and applied to study and analyze the macro mechanical properties of the carbon black filled rubber composites by the micromechanical finite element method. And numerical simulations under uniaxial compression have been made by two-dimensional plane stress model. The periodic boundary conditions are imposed on each Representative Volume Element in order to ensure the compatibility of the deformation field. The dependence of the macroscopic stress-strain behavior and the effective elastic modulus of the composites, on particle distribution pattern, particle volume fraction and particle stiffness has been investigated and discussed. It is shown that the stiffness of the composite is increased considerably with the introduction of carbon black filler particles, and the effective elastic modulus of the composite is increased with the increase of the particle volume fraction.

Abstract: The properties of plain fiber assembly are discussed, and the fiber alignment in the assembly is investigated. Some theories in yarn irregularity research are applied to the study of the random and parallel fibers in fiber assembly and a stochastic method is used to generate the fiber assembly. For a given fiber length distribution, the combing process is simulated on the basis of the plain fiber assembly. If the main parameters in cotton combing are known, the simulation can be conducted, yielding the fiber length distributions in combed cotton slivers and in waste cotton that is combed down. With the variation of combing parameters, the influence of combing parameters on the combing results can be discussed from the simulation.

Abstract: The resources are divided into curable and non-curable morphologies to improve the efficiency of virtual computing systems, and we determine the quantities of such two morphologies based on resources required and resources that system owes. In the environment of tasks presenting random distribution, we describe the trend of resources required and compute the range interval of curable morphology by theoretical analysis. It is concluded that the amounts of tasks have great effect on resources required and the experiments show that the trend of decision variables in our numerical example is consistent with the theory discussed. The decision making methodology presented in this paper has some significance to the management of resources in virtual computing systems.