Search

Abstract: The requirements of device under the multi-mission of monitoring, protecting, controlling and communicating is discussed in this paper to bring forward a design of dual CPU structure with ARM & DSP for microprocessor protection. Via the comparative analysis of several common algorithms, the algorithm of FFT, which can make the controlling of calculation accuracy easier and has a filter feature inherently, is selected to analyze the sampling data of protector and calculates relative electrical parameters. Relevant program is written for distance protection in 110kV transmissions, then debugging on software and hardware is made. The results show that the design of DSP & ARM based-microprocessor protection presented in this paper is feasible; such a device has simple and reliable circuit structure and is expandable in both software and hardware. [Copyright &y& Elsevier]

Abstract: Large solar thermal systems (LSTS) are a promising market segment for solar energy. However, the realization of LSTS is more challenging compared to smaller plants, in both technical and economical terms. Permanent monitoring, data evaluation and fault detection during the operation of LSTS has shown by Fink et al. [3] and Peuser et al. [13] to be crucial for ensuring optimal performance. The only cost-effective way for permanent surveillance of LSTS operation is to make use of a computer-aided tool that performs as many steps as possible in an automated mode. The R&D project ‘IP-Solar’ developed the scientific basis and the technical fundamentals for such a system. This paper presents the current state of development emphasizing the methodology of the operation diagnostics. In particular, the algorithm-based approach, its implementation and testing are described in detail. [Copyright &y& Elsevier]

Abstract: Protein-protein interaction (PPI) network is essential to understand the fundamental processes governing cell biology. It is well known that document classification systems are potential to accelerate the curation process by retrieving PPI-related documents. However, it is usually a case that there is an imbalanced two class data set on text classification. Learning from imbalanced data sets is an important challenge to the machine learning community. In this paper, we compare the performance of several document classifiers on one PPI document set and vary the size of the number of positives and the ratio of the number of positives to the number of negatives on training set and testing set in the experiment. Through the experiment, we try to find what kind of classification algorithm is suitable for imbalanced PPI document classification. [Copyright &y& Elsevier]