Applied Mechanics and Materials Vols. 513-517

Abstract: Prognostic capability is a very important character for PHM distinguishing from other diagnosis systems, fault prognostic method based on SVR was researched on in this paper. In order to solve kernel function parameters, penalty coefficient and insensitivity loss coefficient of SVR, improved QPSO was put forward to train SVR in this paper. And then a certain power supply combination in beam control system of a certain control and guide radar was taken as an example to collect voltage signal and forecasted, and its precision was very perfect and could satisfy prognostic demand.

Abstract: This paper introduces the three dimensional animation to simulate the CBKI gear pump assembly procedure with three dimensional manufacture software 3DMAX, AUTOCAD2000 and AE in which AUTOCAD2000 is used to create the gear pump line, 3DMAX is for modeling and AE is used to create the animation and post-processing. This paper mainly focuses on the scheme of the gear pump in the assembly which is from the design and manufacture to the final performance.

Abstract: This paper puts forward a network anti-interference method based on multi-state hidden layers' distribution with iterations. It first calculates the distribution characteristics of multi-state data, and analyzes the law of network data; secondly, statistically analyzes the distribution of multi-state hidden layers, introduces constant iterations and update, and extracts system noise characteristics; then separates noise and network data, so as to implement anti-interference; finally carries out experiments with a set of test data in the actual network, and the results show that under strong interference, network anti-interference method based on multi-state hidden layers' distribution with iterations effectively reduces the network error rate, basically does not affect the normal communication, and has a good application value.

Abstract: The special service website is an effective way to conduct special services in network environment. Recently many problems exist in website construction for special topic service: the style is not uniform, the database is not compatible, communication protocol is not practical, technical architectures have serious differences and rights management is confused. All of the problems mentioned above lead to a great waste of hardware, software, human and financial resources. This paper proposes a Drupal-based rapid website establishment solution which effectively improves the construction and management efficiency of special topic service website and increases compatibility between different topic service sites, while greatly reducing the cost of topic services.

Abstract: To solve the problem of inefficiency of DM artificial modeling and difficulty of knowledge reuse, this paper studies DM application characteristics, technical characteristics and business data characteristics, probes for the automatic DM modeling method and designs the evaluation system of DM modeling. Based on the DM automatic modeling research and combined MAS, we set up the framework of MAS DM modeling automatic selection, and assess the feasibility of severity selection process by data analysis.

Abstract: Information transmission is widespread in business process of the organizations. In order to facilitate access to information transmission, this article divides information transmission into three categories: process, rules as well as tricks and set up the flow chart with various factors of communication and information among organizations. This paper proposes the double closed-loop organization model, a reconstruction method based on information management. Through the study of the effective management of information, the organization of information is enriched and the sharing of information is enhanced. Therefore, the organization has a certain intelligence, which provides theoretical guidance for the organizational reconstruction.

Abstract: Many original signers can put a delegation of powers to many proxy signers, it is a representation of all the original signers to generate a plurality of proxy signers in the multi-proxy multi-signature scheme. It is analyzed to the existing multi-proxy multi-signature schemes in this paper , the verification equation is improved. A new secure and efficient scheme is proposed. The security analysis shows that the verification equations of the new scheme is more safe. The new scheme can resist the public-key substitution attack, can resist the coalition attack.

Abstract: Based on system analysis of information security risk factors and evaluation process, aiming at the uncertainty information is difficult to quantify the evaluation process, Bayesian network inference algorithm, and combined with the inference rules of conditional probability matrix is given by expert knowledge of Bayesian network, the evaluation model of information security risk. Finally an instance of the risk assessment approach on the model is analyzed which demonstrates the rationality and feasibility of this method. So it provides a new method for information security assessment.

Abstract: Recently, with rapid development of computer/network technology and algorithms for composing test paper, cyber-based online examination system is a practically valuable hot research concern. In the paper, the mathematical model is created for solving problems with the online test paper composition system. Through comparative analysis of merits and shortcomings of various coding schemes, and to overcome the shortcoming that traditional genetic algorithms easily fall into premature convergence, it utilizes the adaptive adjustment method of dynamic parameters and elitist strategy to improve to develop the online test paper forming scheme based on adaptive genetic algorithm. For the selection of each parameter, simulation test is conducted to obtain the solution approximate to the best one.

Abstract: In order to solve the problem of information overload, which is a consequence of the extreme abundance of digital television (TV) program resource, this paper proposed a digital TV program recommendation system based on latent factor model (LFM). This system constructed a digital TV program recommendation system structure including information inputting unit, system analysis unit and information recommendation sending unit. The proposed program type analysis method searches for the relation between audience interest and the programs which audiences watch. To classify the audience crowd, there are two types of analysis method which are program type threshold analysis and program type cluster analysis that based on artificial classification of experience value and automatic clustering with arbitrary number of clusters respectively. The feasibility of the algorithm has been proved by simulated analysis.