Applied Mechanics and Materials Vols. 411-414

Abstract: Context-aware computing is an important aspect of Pervasive Computing, Cyber-physical Systems (CPS), Cloud computing, etc. Recently, few works have been reported on formal approaches for this area. Bigraph was proposed as a mathematical structure that is consist of two graphs, a place graph and a link graph, intended for modeling distributed, mobile systems, etc. Bigraphs and its corresponding Bigraphical Reactive Systems (BRSs) present a modern, graphical structure, algebra calculus for the description of the syntax and semantics of systems in accordance with the orthogonal notions of connectivity and locality. Bigraphs and BRSs have been applied in various areas. Bigraph is used to formalize context-aware models in this paper. Then, we present an approach to check the formal context-aware models. The formalisms and checking are explicitly depicted through a simple but non-trivial example.

Abstract: Based on the American NSWC-10 criterion, we analyzed the mechanical standard components failure mechanisms and failure modes, through which inducted their common methods and calculation models of reliability prediction according to their specific characteristics. Furthermore, they have been transformed into the metric system standards used in China. And also the development of the softwares interface, algorithm, connection and data preservation were accomplished utilizing C# development language in the integrated development environment of Visual Studio to realize the reliability prediction of mechanical standard components.

Abstract: Aiming at the present situation of research institutions attaching great importance to air pollution control but air quality data acquisition is not convenient, solutions based on Python and Django to provide the service of the air quality data downloading such as PM2.5 is proposed along with their detailed implementation process in Web system design and development. The results indicate that, these techniques enhanced the developing efficiency, the system's service of data downloading can provide comprehensive monitoring data which can be used directly by researchers and improve their efficiency. Accumulated historical data storage is also very important to the study of air quality changing and the pollution prevention.

Abstract: Detection of design patterns in software design phase possibly ensures the non-functional requirements, regarding performance features, before investing the implementation. We formalize the structural UML class diagram using graph. By applying graph matching technique, we propose an alternative of subgraph matching algorithm to extract the local properties of the UML class diagrams and perform the detecting of subgraph of possible design patterns found in the target software design model.

Abstract: To ensure the correctness of Web applications, the dynamically generated Web pages must be tested. However, this can be impractical as dynamic pages can be infinite depending on user inputs and internal states of Web applications. This paper presents an approach to classify the JSP dynamic pages for reducing the number of pages to be tested. Specifically, the approach identifies the HTML statement slices inside a JSP script and classifies the dynamic pages generated by the script according to the compositions of the HTML statement slices associated with the script's linearly independent paths. With the approach, the JSP dynamic pages containing similar testing artifacts can be classified into the same category. Thus, the number of JSP dynamic pages to be tested can be greatly reduced.

Abstract: Different refrigerants have different performance in same compression cycle, as TSCE+TG has two compressors, the refrigerants must be compressed between high pressure compressors and the lower. With high pressure and intermediate pressure increasing, the COP shows a downtrend. When evaporating temperature goes up, the COP decreases.

Abstract: Nowadays, a variety of database systems coexist and the application softwares access and operation for database is often custom-written, which can lead to more complicated cross-database access. Through analysis of the operational process of the database middleware, adding the generic database query middleware eQueryMW which can shield the difference of the data source to the traditional three-tier architecture as the middleware level can realize the access of relational heterogeneity database, providing a unified database query interface and cross-database integration method for the users.

Abstract: Combine enterprise business process operation knowledge base, unified management knowledge agent and business Process Model, proposed a complete conceptual framework and implementation model that can fully complete business process intelligent modeling, business process automatic control and monitoring and business process automatic optimization. Then build a software platform system on SOA and web service technology. The proposed scheme can fully complete the fast intelligent modeling, management and control, simulation and optimization of business processes.

Abstract: A perfect spam filter would avoid ham misclassification. This paper describes the development and application of spam filter. Specifically, the evaluation methodology was designed on-line open source spam filters. Finally, ASP-based anti-spam management system was created to combine the results of multiple filters. We finally find that the filters of ASP-based anti-spam system also make mistakes but can be used in conjunction with users to minimize errors.

Abstract: This paper introduces parallel feature of the GPU, which will help GPU parallel computation methods to achieve the parallelization of PSO parallel path search process; and reduce the increasingly high problem of PSO (PSO: Particle Swarm Optimization) in time and space complexity. The experimental results show: comparing with CPU mode, GPU platform calculation improves the search rate and shortens the calculation time.