Advanced Materials Research Vol. 831

Abstract: This paper presents a novel method of the speech recognition in combining the empirical mode decomposition with radical basis function neural network. Speech signals which pretreated are decomposed by empirical mode decomposition to get a set of intrinsic mode functions. It extracts mel frequency cepstrum coefficient from intrinsic mode function. Features parameters are made up of the coefficients. For BP Neural Network, RBF Neural Network has advantages on approximating ability and learning speed. So using RBF Neural Network as a recognition model is a good method. Experiments show that this new method has good robustness and adaptability. The speech recognition rate of this method reach ninety-one percents accurately under no noise environment. Speech signal recognition is feasible and effective in noisy environment.

Abstract: Based on digital image correlation method (DICM), this paper promotes a new method for tracking the movements of coarse aggregates of asphalt mixtures. The four-point bending fatigue test is applied as the observed test. The fatigue damage process of AC-16 asphalt mixture is observed by CCD camera. Results show that the new method is able to well reflect the exact displacement of coarse aggregates under different loading repetitions. And the movement trends also reflect the influences of cracks in the specimens. Compared to the original method, the new method provides more detail movement information of different materials (aggregates and asphalt binder). This method helps to the research of asphalt mixtures in meso-scale.

Abstract: An H-shaped steel frame in fire combined blast loading was numerically developed in this paper using finite element method by ANSYS thermal analysis. At the first step, a temperature field has been numerically analyzed. At the second step, each element of the steel frame was assigned different modulus of elasticity and yield stress based on the temperature analysis results. A dynamic simulation of the steel frame subjected to a blast load was analyzed under such a temperature field. The failure modes of the frame structure has been discussed. Finally, a same shape frame at the room temperature under the same blast load was simulated to compare with that at the elevated temperature.

Abstract: With the fast development of science and technology, as an emerging Web technology, JSP provides quality tools for professionals and non-professionals to develop high-level web sites. But, currently there are few visualized learning tools about JSP compiling, the most are written interpretations, which offers no intuitive sense for the learners. Aimed at this situation, the compiler compiled a standalone version of the basic learning software of JSP compiling by integrating what have learned and conforming the relative contents. This software contains rudiments, viewing courseware, experimental video, focused explanation, dynamic compiling, exercise, online test, help, etc., besides, the interface is simple and pleasing to the eye, the content is solid and comprehensive, and the operation is convenient and swift, which furnishes the JSP fans with a good guide. Specially, the dynamic compiling module provides the visualized compilation process which could effectively help the users understanding the dynamic compilation process of JSP.

Abstract: Common algorithms of selecting hidden unit data center in RBF neural networks were first discussed in this essay, i.e. k-means algorithm, subtractive clustering algorithm and orthogonal least squares. Meanwhile, a hybrid algorithm mixed of k-means algorithm and particle swarm optimization algorithm was put forward. The algorithm used the position of the particles in particle swarm optimization algorithm to help deal with the defects of local clusters resulted from k-means algorithm and to make optimization with the optimal fitness of k-means particle swarm with the aim to make the final optimal fitness better satisfy the requirements.

Abstract: Face recognition technology is a significant branch of the study of artificial intelligence, the recognition precision is easily affected by facial expressions, skin colors, beam angles in the images and apparels. This essay tests human face images in the format of 24 BMP and realizes face location and mark of five sense organs. Firstly, color space model is adopted to set up skin color distribution model to segment skin regions; secondly, the obtained regions are judged and screened preliminarily, and optimized based on the characteristics of segmented regions with region optimization algorithm of depth-width ratio, rejecting the region with the similar color of the skin caused by some disturbing factors and other naked parts of the body, through which the rough region of human face could be attained and face location could be realized; finally, five organs of the obtained face region is located with the method of grey level region in combination with searching rectangle.

Abstract: This is the study to utilize logistic regression before apply grey system theory to predict construction contractors default.Differ from previous research, this research try to reduce the number of financial ratio for default prediction and employs the area under curve (AUC) to assess the model performance. The empirical results show that the most favorable financial ratio has an acceptable discriminatory power under AUC curve. The result of this research indicates debt ratio plays an important role in construction contractors default prediction.

Abstract: Currently in the process of engineering, but increasingly implemented simulation methods since they are an economical and feasible to predict the behavior of some variable you wish to benefit. The problem of fluid simulation is a broad field of study, traditionally in this area are implemented domain discretization methods, volumes, differences or finite elements (Computational Fluid Dynamics), in this work, a different approach where the discretization is made on the physical properties of fluid and the fluid for reconstruction from its microscopic properties, simulating these, propagating Boltzmann distribution functions for the grid of nodes, this set is comprised of a fluid group of nodes, nodes fluid the border and nodes structure, docked the method to the boundary conditions necessary to simulate Glycerol in a pipe.

Abstract: Based on the dynamic wave routing algorithm of SWMM model, this paper simulated the residential area drainage system. Then combining with the status reports, such as link summary, node summary, time series plot and the link profile plot, the drainage system of residential area is studied. And the drainage capacity of drainage system, the status of surcharging node and surcharging conduit under heavy rain situation are given. In accordance with the requirements of criterion, this paper used the method of iterative algorithm to improve the drainage system, and finally built the drainage system which meets 2 year storm.

Abstract: Dai writing as the product of Dai civilization. Because of the limitation of culture and territory, it cant spread widely. In todays society, the utilization rate of it still very low, only few people are familiar with it and know it. Remain resources still very limited, cant touched by people. This article uses .net platform to develop Dai writing online inquiry system, meeting the learning requirement of people who want to learn Dai writing.