Applied Mechanics and Materials Vols. 513-517

Abstract: This paper proposed a discrete harmony search algorithm, named DHS, for solving the no-wait flow shop scheduling problem with the objective to minimize total flowtime. Firstly, the total flowtime is shown. Secondly, a harmony is represented as a discrete job permutation and the well-known NEH method is proposed to initialize the harmony memory. Thirdly, Extensive computational experiments are carried out based on a set of well-known benchmark instances. Computational results show the effectiveness of the DHS algorithm in solving the no-wait flow shop scheduling problem.

Abstract: In this study, on the basis of pixel-based classification, object-oriented classification method was used to extract information from high resolution satellite imagery. Select the Binhai New Area as the study area,World View-2 data was selected as data sources, the rule sets of information extraction developing were established firstly, then the parameters of imagery segmentation and classification were tested repeatedly to achieve building hierarchies and map elements. The results showed that object-oriented information extraction method was feasible, and the extracted information was used to produce thematic map on the ArcGIS platform.

Abstract: Based on metadata, this paper introduced an equipment maintenance support information description mechanism to store, manage and apply mass maintenance support data efficiently. The metadata was acquiesced by manual logging and auto extracting. The core content was the metadata generated process, in which metadata content, summary and core metadata were defined based on actual equipment maintenance support resource information. The unitized description laid a strong foundation for equipment maintenance support resource information integration.

Abstract: In order to improve evaluation of e-learning systems, this paper introduces a model of learning content recommendation based on grey systems theory and gray relational analysis algorithms. This work analyses e-learners behaviours of e-learning and log files of the platform, find the most resemble e-learner, use his/her learning trail to recommend e-learning resources. The gray system theory can solve the problem of lacking historical data. And this model reach the purpose of selecting individual teaching contents intelligently.

Abstract: Today's society is a society of information explosion, the popularity of the Internet and development bring a lot of convenience to people, people can easily get a lot of information on the network, however, facing so many information, people prone to the problems of "information overload" and "resources disorientation. Therefore, the recommended system came into being, the recommendation system can provide people with the most in need and most concern to avoid the time of the search and comparison. This article intends to use the very mature recommendation system in the field of electronic commerce to distance education system and promotes personalized learning, shifting the traditional "what teachers teach, what students receive" to "what the students need, what the system provides, which is consistent of constructivism study philosophy. The analysis of users interested as the basis of the recommendation system, users clustering is very important, the objective classification of fuzzy clustering analysis can recommend for users to enjoy high-quality service.

Abstract: A comparison of nonlinear autoregression with exogenous inputs (NARX) neural network and back-propagation (BP) neural network in short-term prediction of building cooling load is presented in this dissertation. Both predictive models have been applied in a group of commercial buildings and analysis of prediction errors has been highlighted. Training and testing data for both prediction models have been generated from DeST (Designers Simulation Toolkits) with climate data of Shanghai. The simulation results indicate that NARX method can achieve better accuracy and generalization ability than traditional method of BP neural network. This work provides a key support in smooth and optimizing control in air-conditioning system.

Abstract: Image interpolation is one of the key technologies in image processing. A special distance correction algorithm based on membership function is proposed in this paper to complete image interpolation. Fuzzy logic is used to get the membership function with the local characteristics of the gradient and phase angle. The first step is to correct the special distance of interpolated pixels along one dimension in the basis of local asymmetry features and the membership function, then convert the corrected distance of one dimension into two dimension, applying the corrected distance to conventional image interpolation algorithms. Experimental results demonstrate that this algorithm can produce better results in regard to the signal-to-nosie ratio and succeeds in preserving interpolation image edges of various directions.

Abstract: An adaptive clock variance calculation algorithm is designed to improve the real-time characteristics and complexity of general method. This algorithm use the latency and smoothing characteristics of exponential smoothing to achieve real-time calculation online. Using this algorithm, synchronous system can response to environmental changes in a short time. The test result shows that the real-time characteristics and output robustness can be improved obviously.

Abstract: This paper addresses a novel global appearance-based approach to recognize objects in images by using multi-scale Gaussian derivatives (GDs). Because the GDs distributions of filtered images are almost picked, this obstacles obtaining discriminative binned distributions for each image. For this reason, we execute k-means clustering on each scale of pooled Gaussian derivative set of the instances come from all classes to yield k-cluster centroids for partitioning feature space, thus generating normalized binned marginal distributions for all training and testing samples, which are holistically adaptive to underlying distributions. On similarity matching, we identify each image with a point of product multinomial manifold with boundary, and use the direct sum of geodesic distance metric for sets of binned marginal densities. The promising experimental results on Zurich buildings database (ZuBuD) validate the feasibility and effectiveness of our approach.

Abstract: Weight function neural network is a new kind of neural network developed in recent years, which has many advantages, such as finding globe minima directly, good performance of generalization, extracting some useful information inherent in the problems and so on. In this article, we apply orthogonal weight function neural network to the speech recognition. The result indicates that the weight function neural network has a good efficiency in speech recognition.