Search

Automatic generation control systems are designed to adjust electric power outputs of multiple generators simultaneously in accordance with the current load. However, the control instruction and the main steam pressure have significant impacts on the resulting active power generation of a conventional thermal generator, and the impacts may be associated with nonlinear characteristics. As a result, the control instruction requires an accurate modeling of the relationship between these three variables for a satisfactory control performance. This paper proposes a method to build a piecewise linear model for the nonlinear relationship from steady-state data hidden in historical data samples. The proposed method is composed by two main steps of steady-state interval detection and steady-state data segmentation. Historical data samples are grouped using the k-means clustering algorithm, and the time domains of each cluster are merged in a specific way to obtain the steady-state intervals. The steady-state data are taken as the samples means of data in the steady-state intervals. A bottom-up algorithm is utilized to partition the steady-state data into numbers of sets iteratively, and the parameters of the piecewise linear model for each data set are estimated by the least squares algorithm. The effectiveness of the proposed method is illustrated via industrial applications to two thermal power generation units.

ABSTRACT Decomposition of structural domains is an essential task in classifying protein structures, predicting protein function, and many other proteomics problems. As the number of known protein structures in PDB grows exponentially, the need for accurate automatic domain decomposition methods becomes more essential. In this article, we introduce a bottom-up algorithm for assigning protein domains using a graph theoretical approach. This algorithm is based on a center-based clustering approach. For constructing initial clusters, members of an independent dominating set for the graph representation of a protein are considered as the centers. A distance matrix is then defined for these clusters. To obtain final domains, these clusters are merged using the compactness principle of domains and a method similar to the neighbor-joining algorithm considering some thresholds. The thresholds are computed using a training set consisting of 50 protein chains. The algorithm is implemented using C++ language and is named ProDomAs. To assess the performance of ProDomAs, its results are compared with seven automatic methods, against five publicly available benchmarks. The results show that ProDomAs outperforms other methods applied on the mentioned benchmarks. The performance of ProDomAs is also evaluated against 6342 chains obtained from ASTRAL SCOP 1.71. ProDomAs is freely available at . Proteins 2014; 82:1937-1946. © 2014 Wiley Periodicals, Inc. [ABSTRACT FROM AUTHOR]

Abstract: We analyze two bottom-up reduction algorithms over binary trees that represent replaceable data within a certain system, assuming the binary search tree (BST) probabilistic model. These reductions are based on idempotent and nilpotent operators, respectively. In both cases, the average size of the reduced tree, as well as the cost to obtain it, is asymptotically linear with respect to the size of the original tree. Additionally, the limiting distributions of the size of the trees obtained by means of these reductions satisfy a central limit law of Gaussian type. [Copyright &y& Elsevier]