Applied Mechanics and Materials Vols. 321-324

Abstract: The HVAC (Heating, Ventilation and Air-conditioner) technology is an efficient approach to achieve dynamic thermal comfort in the transitional spaces where connect the interior and exterior space. The paper describes an optimized method to thermal environmental parameters in the hot summer and warm winter zone is benefit to human comfort, health and energy saving. Based on the dynamic comfort equations and the advantage of artificial immune algorithm in tackling combinatorial optimization to engineering problems, the artificial immune algorithm has been applied to the optimization of thermal parameters for the HVAC design within a transitional space. Therefore, it is shown that the proposed algorithm here might be adopted in the control of the HVAC system.

Abstract: Fault feature extraction and recognition play crucial role in fault diagnosis. In this paper, a fault diagnosis method for three-phase fully-controlled bridge rectifier circuit based on Self-Organizing Map network is proposed. The method utilized the three phase AC input current as fault detection data. Then, perform spectrum analysis with the FFT, the fault data is trained through a Self-Organizing Map network for diagnosis. Simulation and relevant experiment verifying the proposed algorithm can classify various types of power electronics device faults accurately and rapidly.

Abstract: Based on variable fuzzy evaluation method and model, which is suitable for evaluation with the characteristics of multi-hierarchy, multi-indexes, dynamic variable and fuzzy though, this paper assesses the planning design schemes of urban traffic facilities to rich the evaluation research of this field, and that will provide references for more widespread application of variable fuzzy evaluation method and model in future. Key words: variable fuzzy, evaluation, urban traffic, planning design scheme

Abstract: The locality sensitive k-means clustering method has been presented recently. Although this approach can improve the clustering accuracies, it often gains the unstable clustering results because some random samples are employed for the initial centers. In this paper, an initialization method based on the core clusters is used for the locality sensitive k-means clustering. The core clusters can be formed by constructing the σ-neighborhood graph and their centers are regarded as the initial centers of the locality sensitive k-means clustering. To investigate the effectiveness of our approach, several experiments are done on three datasets. Experimental results show that our proposed method can improve the clustering performance compared to the previous locality sensitive k-means clustering.

Abstract: A clustering algorithm based on one-class support vector machine has been proposed recently. Because the kernel technique is used, this approach can appear preferable to the traditional k-means clustering. Clustering ensemble method can combine several divisions of all unlabeled data into a single clustering to gain the better clustering results. In this paper, the clustering ensemble method is applied to the clustering algorithm based one-class support vector machines. Several partitions of multiple runs with different random initial data sets are combined into a final clustering result. Experiments show that the new approach can improve the clustering performance.

Abstract: In the initialization of the traditional k-harmonic means clustering, the initial centers are generated randomly and its number is equal to the number of clusters. Although the k-harmonic means clustering is insensitive to the initial centers, this initialization method cannot improve clustering performance. In this paper, a novel k-harmonic means clustering based on multiple initial centers is proposed. The number of the initial centers is more than the number of clusters in this new method. The new method with multiple initial centers can divide the whole data set into multiple groups and combine these groups into the final solution. Experiments show that the presented algorithm can increase the better clustering accuracies than the traditional k-means and k-harmonic methods.

Abstract: The study of Concept Similarity is a very important aspect of Knowledge Representation and Information Retrieval in Artificial Intelligence, and it is also a bottleneck that hasn’t been well solved in the Ontology Research. In this article, we take every influencing factor into account, especially the area density, a new method of concept similarity based-on Domain Ontology is suggested. The experiment results show that: the new method we proposed in this article can more reasonably describe the concept similarity.

Abstract: This paper proposes an algorithm to compute the sensitivity of the Radial-Basis Function Neural Network (RBFNN) due to the errors of the inputs and others parameters of the net works. For simplicity and practicality, all inputs and weights are assumed to be independent and identically distributed (i.i.d) with uniform distribution. A number of simulations are conducted and the good agreement between the experimental results and the theoretical results verifies the reliability and feasibility of the proposed algorithm. The relationship between the sensitivity of RBFNN and input error and the perturbation of others parameters is given.

Abstract: A method of adaptive synchronization of one-dimensional discrete chaotic systems on complex networks is proposed. The nodes of complex networks are constructed by one-dimensional discrete chaotic systems, we consider a general drive-response synchronization model of one-dimensional discrete chaotic systems on complex dynamical networks. Based on the adaptive control technique, the parameter adaptive laws and property conversion laws are given to achieve synchronization and parameters identification simultaneously. Simulation results show that the arithmetic average and geometric mean of all the nodes states are equal, furthermore, the unknown node parameters can be successfully identified, all nodes are transformed to drive nodes. This indicates that chaos synchronization is reached in the whole networks.

Abstract: This study proposes a new modeling method for unknown systems. Through this method, the transfer functions can be identified. First, the input-output data pairs of the unidentified system should be collected. Then, the transfer function’s coefficients can be identified based on the errors via the derivative-free search methods such as GA etc. Here, a second-order transfer function is used in this study. For a second-order transfer function is difficult to approach each system, a plurality of transfer functions may be used depending on the precise requirement. Finally, following the previous steps, the other transfer function can be found in succession. In order to confirm the effectiveness of this proposed method, an electromagnetic flywheel (EF) system is used in this study. Such kinds of systems are always with many uncertainties as nonlinear electromechanical coupling and electromagnetic saturation, etc. They are difficult to modeling via traditional mathematic ways. In this study, the data pairs of EF system is collected by experiments. By assessing the results of the proposal and experimental data shows that this method is feasible to any unknown systems. system, achieve more saving energy and high efficiency control purposes.