We study the efficiency of individual stations of the CRL seismic network in recording the seismic activity in the western Gulf of Corinth, Greece. The stations are located on both the northern and southern coast of the Gulf. The study is based on 5027 earthquakes recorded in 2001, separated into three groups, the southern, central and the northern one. The events were located using the HYPO71PC algorithm. It is shown that the stations significantly differ in their monitoring ability., Jaromír Janský, Jiří Zahradník and Vladimír Plicka., and Obsahuje bibliografii
Two exogenous NO donors were used to act as substitutes for impaired endogenous nitric oxide (NO) production due to inhibition of NO synthase in rats. Six weeks' lasting inhibition of NO synthase by NG-nitro-L-arginine methyl ester (L-NAME) induced stabilized hypertension. Simultaneously administered isosorbide-5-mononitrate did not prevent the development of hypertension. Molsidomine, administered concomitantly with L-NAME, significantly attenuated the BP increase. However, BP was still found to be moderately increased compared to the initial values. Remarkable alterations in the geometry of the aorta, carotid and coronary artery found in NO-deficient hypertension were prevented in rats administered L-NAME plus molsidomine at the same time. In spite of 6 weeks' lasting inhibition of NOS, the NOS activators acetylcholine and bradykinin induced BP decrease; the maximum hypotensive value did not differ from the values recorded in the controls or in animals treated with L-NAME plus molsidomine. Notably enough, the hypotension was similar to that found in rats administered L-NAME alone for six weeks. After NO synthase inhibition, Isosorbide-5-mononitrate does not substitute and molsidomine substitute only partially the impaired endogenous NO production., M. Gerová, F. Kristek., and Obsahuje bibliografii
The periodic orbits in circular restricted 3-body problem are calculated by different numerical as well as analytical methods. The efficiency of both kinds are compared in this contribution. The improvement of analytical methods can be achieved by an artificial splitting of perturbation term. The analytical approximations are thus sufficiently accurate even for large values of mass ratio μ. The use of these approximations as a zero-order approximation In numerical codes for search for periodic orbits improves their efficiency also.
In this paper, we propose a novel algorithm for solving an optimal boundary control problem of the Burgers' equation. The solving method is based on the transformation of the original problem into a homogeneous boundary conditions problem. This transforms the original problem into an optimal distributed control problem. The modal expansion technique is applied to the distributed control problem of the Burgers' equation to generate a low-dimensional dynamical system. The control parametrization method is formulated for approximating the time-varying control by a finite term of the orthogonal functions with unknown coefficients determined through an optimization process. The minimization of the objective functional is performed by using a conjugate gradient method. The accuracy and convergent rate of this hybrid method are shown by some numerical examples.
We introduce a new efficient way of computation of partial differential equations using a hybrid method composed from FEM in space and FDM in time domain. The overall computational scheme is explicit in time. The key idea of the suggested way is based on a transformation of standard basis functions into new basis functions. The results of this matrix transformation are e-invariants (effective invariants) with such suitable properties which save the number of arithmetical operations needed for a problem solution. The application of this procedure and its effectiveness for 2D problem was the first time published in \cite{halabala}. Now we describe the generalization of this procedure for 3D problem. In order to present the main principle of our process and its advantage, we first explain the main idea of our approach on a simple 1D example and then the application of the e-invariants on an elastodynamics equation using hexahedral elements in 3D is described. Finally, the efficiency of the suggested method in both cases from the point of the required number of arithmetical operations is analyzed. The result of this analysis confirms computational efficiency the suggested method and the usefulness of e-invariants which save only the essential information needed for the computation. Moreover, the method can be used for various types of elements and equations.
This paper is devoted to analysis of block multi-indexed higher-order covariance matrices, which can be used for the least-squares estimation problem. The formulation of linear and nonlinear least squares estimation problems is proposed, showing that their statements and solutions lead to generalized `normal equations', employing covariance matrices of the underlying processes. Then, we provide a class of efficient algorithms to estimate higher-order statistics (generalized multi-indexed covariance matrices), which are necessary taking in mind practical aspects of the nonlinear treatment of the least-squares estimation problem. The algorithms are examined for different higher-order and non-Gaussian processes (time-series) and an impact of signal properties on covariance matrices is analysed.
In a mirror server environment, clients request services from servers. Therefore, the system must have an intelligent algorithm to select the most suitable server to fulfill a coming request. Choosing such a server for a particular client may be very difficult. Evolutionary techniques can be utilized to determine the server best suited to a particular client request based on parameters such as processing and reply times. Usage of genetic algorithms in server selection is researched in this paper taking into consideration various probabilities for mutation and crossover.
This paper focuses on gradient-based backpropagation algorithms that use either a common adaptive learning rate for all weights or a separate adaptive learning rate for each weight. The learning-rate adaptation is based on descent techniques and estimates of the local constants that are obtained without additional error function and gradient evaluations. This paper proposes three algorithms to improve the different versions of backpropagation training in terms of both convergence rate and convergence characteristics, such as stable learning and robustness to oscillations. The new modification consists of a simple change in the error signal function. Experiments are conducted to compare and evaluate the convergence behavior of these gradient-based training algorithms with three training problems: XOR, encoding problem and character recognition, which are popular training problems.
The egg shell of Huffmanela huffmani Moravec, 1987 forms three main layers: an outer vitelline layer, a middle chitinous layer, and an inner lipid layer. The vitelline layer, forming the superficial projections of the egg shell, comprises two parts: an outer electron-dense, and an inner electron-lucid part. The chitinous layer is differentiated into three parts: an outer homogenous electron-dense part, a lamellated part, and an inner electron-dense net-like part. The lipid layer comprises an outer net-like electron-lucid part, and an inner homogenous electron-lucid part. The polar plugs are formed by electron-lucid material with fine electron-dense fibrils.