Applied Mechanics and Materials Vols. 241-244

Abstract: Through the analysis of its geology and geomorphology to calculate the relevant parameters , some hidden problems of slope stability of the mountain road are calculated and analyzed by the finite element method for solving the slope stress and strain and the analysis results of plastic zone , which confirms the slope stability conclusions drawn by the safety factor .

Abstract: In order to suppress the harmonic and protect the IGBT, an innovative buffer circuit for IGBT was designed and simulated in this paper. Compared with traditional buffer circuit, the proposed RCD buffer circuit is more reliable and can reduce the heat effectively when IGBT is turned off. The analytical expression of the buffer circuit is deduced by theoretical calculation. Finally, the performance of the presented buffer circuit was testified by simulation.

Abstract: In traditional aeroengine modeling, the nonlinear equations of engine model are generally solved through iterative algorithm. However, due to the strong nonlinear characteristics of the equations, the iterative model often fails to converge at some points of the full envelope and has a poor real time performance. In order to solve the problems, this paper proposes a non-iterative modeling method based on volume effect. In this method, several variables and differential equations of volume dynamics in aeroengine are introduced to the nonlinear equations, as a result, the whole set of equations becomes closed-form and can be solved without iteration. The non-iterative model of a gas turbine engine is written in matlab code. Furthermore, an open-loop simulation is carried out in matlab/simulink, both under groud and altitude condition. Meanwhile, the non-iterative model is verified by GSP11. The results illustrate that the non-iterative model provides good performance both in the stability and accuracy of solutions.

Abstract: In recent years, many application systems have generate large quantities of data, so it is no longer practical to rely on traditional database technique to analyze these data. Data mining offers tools for extracting knowledge from data, leading to significant improvement in the decision-making process. Association rules mining is one of the most important data mining technology. The paper first presents the basic concept of association rule mining, then discuss a few different types of association rules mining including multi-level association rules, multidimensional association rules, weighted association rules, multi-relational association rules, fuzzy association rules.

Abstract: A new algorithm for training radial basis function neural network (RBFNN) is presented in this paper. This algorithm is based on the dynamic fuzzy clustering method (DFCM). The algorithm has a number of advantages compared to the traditional method based on k-means. For example, it does not need to know the number of the hidden nodes and to predicts more accurately. Due to these advantages, this method proves to be suitable for developing models for complex nonlinear systems.

Abstract: Aiming at the problem that most of weighted association rules mining algorithms have not the anti-monotonicity, this paper presents a weighted support-confidence framework which supports anti-monotonicity. On this basis, weighted boolean association rules mining algorithm and weighted fuzzy association rules mining algorithm are presented, which use pruning strategy of Apriori algorithm so that improve the efficiency of frequent itemsets generated. Experimental results show that both algorithms have good performance.

Abstract: It usually need different ways to process different objects in the manufacturing, Therefore, firstly we need to distinguish the categories of objects to be processed, then the machine will know how to deal with the objects. In order to automatically recognize the category of the irregular object, this paper extracted the improved Hu's moments of each object as the feature by the way of processing images of the working platform that the irregular objects are putting on. This paper adopts the variable step BP neural network with adaptive momentum factor as the classifier. The experiment shows that this method can effectively distinguish different irregular objects, and during the training of the neural network, it has faster convergence speed and better approximation compared with the traditional BP neural network

Abstract: This paper is focused on the problem of finite-time H-infinity tracking control for the uncertain multi-agent systems with a leader. First, a nonlinear finite time H-infinity tracking control protocol is proposed for multi-agent systems. Second, it is proved that the given protocol can make the closed system reach consensus in finite time based on the theory of finite-time Lyapunov stability and the theory of robust H-infinity control theory. Finally, the simulation results are presented to illustrate the effectiveness of the obtained results.

Abstract: To solve the multi-pose ear recognition problem under the different illumination condition, a novel method which combines phase congruency with kernel discriminant analysis (KDA) is proposed. The phase congruency of ear image is first calculated using Log-Gabor filter with 5 scales and 8 orientations, and then the phase congruency of different orientation is constructed as high dimensional vector including ample information. The high dimensional vector is mapped to kernel space to acquire discriminant feature. Experimental results show that the proposed method obtains higher recognition rate compared with the other related methods. The method of the phase congruency can eliminate the influence of illumination and phase congruency with KDA is effective to multi-pose ear recognition.

Abstract: The performance of support vector machine (SVM) depends on the selection of model parameters, however, the selection of SVM model parameters more depends on the empirical value. According to the above deficiency, this paper proposed a parameters optimization method of support vector machine based on immune memory clone strategy (IMC). This method can solve the multi-peak model parameters selection problem better which is introduced by n-folded cross-verification and automatic acquire the optimum model parameters. Proved by the simulation results on standard data, this method has higher precision and faster optimization speed. In a word, it can be used as an effective and feasible SVM parameters optimization method.