Applied Mechanics and Materials Vols. 380-384

Abstract: By analyzing the current Particle Swarm Optimization, especially the analysis of weighting coefficient descending and random disturbance term improving, an improved Particle Swarm Optimization combining periodic weighting adjustment and random disturbance was put forward and its effectiveness is verified by experiments in this article.

Abstract: To avoiding e the problem of premature convergence of GA algorithm, a novel algorithm was advanced based on niche and discrete. The improved GA algorithm use hierarchical coding and selection method in the classic GA to abundance the individual choices. Then use the multilevel optimization to quicken the algorithm. Tests show the new algorithm has better performance than classic GA in effect and time, especially suit to multi peak problems.

Abstract: The pressure testing of the building materials has been a hot topic of research in the field of architecture. Traditional building materials pressure testing methods all calculate the larger bearing pressure fragile area, and are difficult to be accurate to the very point. This is mainly because of the larger range of signal distribution, which avianizes the correlation of the signal. To address the problem mentioned above, a bearing pressure fragile support point positioning algorithm for the building materials is proposed. The algorithm The combines the quantum computing with the neuron model in neural network to form the quantum neurons, and then expands them into a quantum neural network to achieve the functions of the traditional neural network, enhance the optimization capability of computing the small surface area of the building materials and ensure that the bearing pressure fragile support area of the building materials is further reduced, and shorten the positioning range. Simulation results show that the proposed method has better positioning effect on bearing pressure fragile points computing of the building and higher positioning accuracy.

Abstract: In this paper we propose a new kind of geometry driven subdivision scheme for curve interpolation. We use cubic Lagrange interpolatory polynomial to construct a new point, selecting parameters by accumulated chord length method. The new scheme is shape preserving. It can overcome the shortcoming of the initial four point subdivision scheme proposed.

Abstract: with the development of computer information science and technology, the acquirement and access of data from huge database become more and more convenient. Not all the algorithm of different database is the same. Data mining finds the method of characteristic demand data from a large amount of information. It searches data according to relevance and clustering of data. This paper presents a new data mining system---B / S framework and establishes rules of data mining algorithms and mathematical models using fuzzy membership function and Apriority algorithm. This paper establishes a set of data mining system related with subjects learn of driving school taking the driving school for example and using multi-layer B / S framework mathematical model and results of driving candidates. It also finds the importance of each study subject. It proposes data reference for the order of the learn subjects of driving school which provides a theoretical reference for the study of data mining algorithms and systems.

Abstract: By means of abstract analysis to the hostile planning, this paper decomposes a series of hostile action (hostile planning) which executed by hostile agent to element sets that can be expressed by certain numbers of operators, thus through the abstract description of hostile action operator and attacks operating unit which executed by hostile agent, it achieves the representation of the hostile planning process.

Abstract: Contour line map and digital terrain model are widely used in practical work. With the rapid development of computer technology, computer graphics and geographic information system, they become more and more practical and their roles have become more prominent. Contour line has incomparable advantage of expressing both qualitative and quantitative information especially in the terrain analysis. Many algorithms of contour line map are automatically generated based on the digital terrain model.

Abstract: To solve the problem that the central guidance system takes too long time to calculate the shortest routes between all node pairs of network which can not meet the real-time demand of central guidance, this paper presents a central guidance parallel route optimization method based on parallel computing technique involving both route optimization time and travelers preferences by means of researching three parts: network data storage based on an array, multi-level network decomposition with travelers preferences considered and parallel shortest route computing of deque based on messages transfer. And based on the actual traffic network data of Guangzhou city, the suggested method is verified on three parallel computing platforms including ordinary PC cluster, Lenovo server cluster and HP workstations cluster. The results show that above three clusters finish the optimization of 21.4 million routes between 5631 nodes of Guangzhou city traffic network in 215, 189 and 177 seconds with the presented method respectively, which can completely meet the real-time demand of the central guidance.

Abstract: The pattern matching algorithm is the mainstream technology in the instruction detection system, and therefore as a pattern-matching methods core string matching algorithm directly affect an intrusion detection system performance and efficiency. So based on the discussions of the most fashionable pattern matching algorithms at present, an improved algorithm of AC-BM is presented. From the experiments in the Snort ,it is concluded that the improved algorithm of the performance and efficiency is higher than AC-BM algorithm.

Abstract: Concerning the defect of fuzzy membership as a function of distance between the point and its class center in feature space for some current Fuzzy Support Vector Machines (FSVM), a new FSVM based on entropy and Genetic Algorithm (GA) named EGFSVM was proposed in this paper. Making use of evaluation of entropy and intelligence of GA, EGFSVM enhances the classification capability and makes clustering center more suitable and membership more accurate. Experimental results show EGFSVM has better precision and classification performance, especially to multi-class and large scale data.