Applied Mechanics and Materials Vols. 380-384

Abstract: The core technology of apparel layout is the intelligent algorithm of apparel marker software, this paper puts forward an optimized particle swarm algorithm under the condition of comparing and analyzing the merits and drawbacks of genetic algorithm, simulated annealing algorithm and particle swarm optimization algorithm, in addition, a demonstration test is done by C++ under the environment of VS2007. The result concluded in this paper shows that the optimized particle swarm algorithm can achieve ideal material utilization.

Abstract: In order to show the time cumulative effect in the process for the time series prediction, the process neural network is taken. The training algorithm of modified particle swarm is used to the model for the learning speed. The training data is sunspot data from 1700 to 2007. Simulation result shows that the prediction model and algorithm has faster training speed and prediction accuracy than the artificial neural network.

Abstract: This paper introduces a Load-balancing Clustering Algorithm of WSN for Data Gathering (LCA-DG), so that the energy is distributed evenly between each cluster, thus prolonging the lifetime of network. A fast local maintenance and update algorithm is proposed on the basis of the optimization algorithm in this article, to make sure that the network will be able to restore quickly and operate normally after adding or removing nodes, so that this algorithm has good expansibility and self-recovery ability, and more in line with the needs of practical application.

Abstract: The application of data mining technology in education, finding the hidden, useful information from a large number of data, will contribute to the reform and development of education. This article presents an improved C4.5 algorithm, which only use simple add, subtract, multiply and divide operations instead of the logarithm of original C4.5 algorithm. This algorithm greatly improves the operation speed and the decision tree generation efficiency, and its application to the analysis of students' achievement, implied to identify the factors affecting the performance, promotes the improvement of teaching quality.

Abstract: The hazard of pool fire is mainly thermal radiation damage. In this paper, a multi-point source model of pool fire thermal radiation is proposed. The thermal sources are the particles constituting the flame which is established by incorporating particle system with Monte Carlo method. A study case of gasoline pool fire is simulated by this model, and the radiation results evaluated by this model are compared with those evaluated by the traditional point source model and the measured data.

Abstract: In this paper we present a key data structure and associated render-time algorithm for the combined display of multi-resolution 3D terrain and traditional GIS vector data, by combining a fast GPU-based terrain solver, which is designed to create fully 3D scenes that deliver the best possible quality and do not require dynamic texture generation and handling. Level of detail terrain models and vector maps are created, and the server-client architecture is presented. The application provides an effective way for powerful access and manipulation of large-scale real datasets, which represents a basic mechanism for 3D GIS interface and exploration tools.

Abstract: Ramanujan Sums (RS) and their Fourier transforms attract many attentions in signal processing in recent years. Thanks to their non-periodic and non-uniform spectrum, RS are widely used in low-frequency noise processing, e.g., Doppler spectrum estimation and time-frequency analysis. We proved the transforms can be perfectly reconstructed under certain circumstance and built a multi-tone system using Ramanujan Fourier transforms as modulation and demodulation. This system got a lower BER in AWGN channel compare to OFDM.

Abstract: In this paper we combine the largest minimum distance algorithm and the traditional K-Means algorithm to propose an improved K-Means clustering algorithm. This improved algorithm can make up the shortcomings for the traditional K-Means algorithm to determine the initial focal point. The improved K-Means algorithm effectively solved two disadvantages of the traditional algorithm, the first one is greater dependence to choice the initial focal point, and another one is easy to be trapped in local minimum [1][2].

Abstract: Traditional method mining for association rules between items in large and grand data sets is inefficient. In this paper we present an efficient method called BPMRA which is based on mapreduce and partition. We have compared BPMRA algorithm based multi-node and partition based single node method and performed some experiments. It turns out that BPMRA possesses high parallelism good stability and scalability, especially suitable for mining for association rules in large and grand data sets.

Abstract: FCM method and ant colony algorithm are all traditional algorithms in image segmentation. The two algorithms can complement each other. The combination of two algorithms will improve image segmentation and speed up algorithms convergence. Tests prove new hybrid algorithm is more effective than single algorithm in image segmentation detection.