Advanced Materials Research Vols. 945-949

Abstract: an improved algorithm of Bayesian classifier combined with Fisher Linear Discriminant Analysis is proposed. This algorithm is the key to search the projection space of maximum separation. The original samples are projected to maximum separation space and new samples are obtained. These new samples are classifed by Bayes classifier. Experimental results show that improved Bayesian classifier has higher accuracy of classification and better performance of classification in the given data collection.

Abstract: There is a great deal of history involving the research of Tibetan network public opinion dissemination behavior. In order to monitor Tibetan web information and real-time construction and supplementary of Tibetan sensitive information, we investigate the random walk dissemination model to analyze the overall information of Tibetan Web. According to Monte Carlo’s random walk models, we simulate the Brownian motion in the different dimensional space and discuss some basic ideas and applications.

Abstract: In this paper, we establish two new differential-integral inequalities. Under several practical assumptions, the inequality is solved by adopting novel analysis techniques, such as: change of variable, amplification method, differential and integration, and the dialectical relationship between constants and variables, and explicit bounds for the unknown functions are given clearly. The derived results can be applied in the study of some practical problems in engineering.

Abstract: In this paper, we discuss a class of new weakly singular difference inequality, which is solved using change of variable, discrete Jensen inequality, Beta function, the mean-value theorem for integrals and amplification method, and explicit bounds for the unknown functions is given clearly. The derived results can be applied in the study of fractional difference equations in Engineering.

Abstract: In the below work the problem of optimal recovery of functions in Hardy class is covered. Namely, by the values ​​of these functions in a finite number of points lying in the unit circle determined their value at a given point. Coefficients of the linear best approximation method and error of the best method are calculated. The functions are considered with some given weight function.

Abstract: When remanufacturing complex surface parts, such as twisted turbine blade, it is difficult to obtain an accurate model. A segmentation and iterative surface reconstruction algorithm for repair of twisted turbine blade is described. The algorithm reconstructs the blade surface by segmenting the data points base on design intent and iterating the control points. The continuity of the fitted closed surface is tested by UG and the error of the fitted surface can be verified with Imageware by comparison with three different blades. It shows that the time complexity of the algorithm is reduced greatly, and the error is within the acceptable limits.

Abstract: Starting from the Riccari equation() expansion approach and a linear variable separation approach, some new complex wave solutions with of the related SchrÖdinger (RS) equation , are derived.

Abstract: A finite-difference time-domain method based on the auxiliary differential equation (ADE) technique is used to obtain the formulation of 2-D TM wave propagation in lossy Lorentz media. In the paper, the reflected coefficients calculated by ADE-FDTD method and the exact theoretical result are better agreement.

Abstract: According to the equal principle, to establish the relationship between function and curved trapezoid area, designed the fluid differential measurement model, analysis of main parameters, introduced working principle, combining with the different forms of curve, depicting the derived function curve measurement method is given.

Abstract: At present, the mine has only realized the real-time monitoring of gas, but not the prediction of gas.There were some limitation of the traditional prediction method, such as modeling subjectivism and statistical prediction. Because it can dynamically adjust the parameters of the model, adaptive prediction method can get the current time according to the prediction error of data and the current time, real-time fault prediction model parameters, this is a very consistent with the prediction method for practical use.This paper presents the gas emission chaos time series method by using volterra series prediction, and on the basis to establish time-series prediction models. The results show that the method not only avoids the phase space reconstruction, but also avoid the points in the neighborhood search, in real-time, with very high efficiency.