The motivation for this paper is the practical problem of interpreting bi-exponential rate-of-rise curves that are observed during many field piezometric tests. In the authors' previous study, a laboratory W-tube ...
详细信息
The motivation for this paper is the practical problem of interpreting bi-exponential rate-of-rise curves that are observed during many field piezometric tests. In the authors' previous study, a laboratory W-tube system of water flow through two samples of sand was introduced with an adequate mathematical model. The desired bi-exponential character of water flow was obtained by placing two different samples of sand in separate but connected columns. In the present paper, a so-called inverse problem is solved. The optimization procedure is applied in order to jointly estimate a pair of hydraulic conductivity values based on experimentally recorded bi-exponential rate-of-rise curves. The obtained values of hydraulic conductivities are presented and compared to the values determined from independent constant-head permeability tests conducted for the analyzed sands. The results of this identification procedure varied in accuracy. The mean percent errors between the hydraulic conductivity values measured jointly and independently for the analyzed experimental series were in the range of 8.8 to 37.2%. The discussion presents the restrictions of this interpretational method and suggests further modeling plans.
X rays focusing telescopes, composed of several Wolter I nested mirror shells reflecting at grazing incidence, are commonly employed only in the soft x-rays band (below 10 keV). At higher energies only direct view det...
详细信息
ISBN:
(纸本)0819454745
X rays focusing telescopes, composed of several Wolter I nested mirror shells reflecting at grazing incidence, are commonly employed only in the soft x-rays band (below 10 keV). At higher energies only direct view detectors (with low or not at all imaging capabilities) have been operative so far. The use of multilayer coatings on Wolter I mirrors can be a suitable way to realize focusing optics operating at higher energies (10 - 100 keV). A procedure based on an iterated simplex numerical method has been developed in order to get a global optimization of the multilayer reflectors sequences that will be applied for this kind of hard X-ray focusing telescopes. In particular, the specific case of the multilayer optimization for the optics of some future X-ray missions has been investigated.
The issue of variable stepsize in the backpropagation training algorithm has been widely investigated and several techniques employing heuristic factors have been suggested to improve training time and reduce converge...
详细信息
The issue of variable stepsize in the backpropagation training algorithm has been widely investigated and several techniques employing heuristic factors have been suggested to improve training time and reduce convergence to local minima. In this contribution, backpropagation training is based on a modified steepest descent method which allows variable stepsize. It is computationally efficient and possesses interesting convergence properties utilizing estimates of the Lipschitz constant without any additional computational cost. The algorithm has been implemented and tested on several problems and the results have been very satisfactory. numerical evidence shows that the method is robust with good average performance on many classes of problems. Copyright (C) 1996 Elsevier Science Ltd.
We develop a new training algorithm for feedforward supervised neural networks based on an affine scaling quadratic optimization method. Specifically, we approximate the error function by a quadratic convex function a...
详细信息
We develop a new training algorithm for feedforward supervised neural networks based on an affine scaling quadratic optimization method. Specifically, we approximate the error function by a quadratic convex function and then optimize the new error function by an affine scaling method, which replaces the steepest descent method in the backpropagation algorithm. Comparative numerical simulations for medical diagnosis problems show significant reductions in learning time with respect to a quasi-Newton algorithm. Copyright (C) 1996 Elsevier Science Ltd
This paper considers the problems that lie in the construction of curves and surfaces. In particular, it is concerned with the modelling of curves and surfaces using spline curves and tensor product spline surfaces. I...
详细信息
This paper considers the problems that lie in the construction of curves and surfaces. In particular, it is concerned with the modelling of curves and surfaces using spline curves and tensor product spline surfaces. It aims to show that many of the problems in this area (which are described in the paper) can be handled by the use of B-spline representation of splines and by the employment of numericaloptimization and linear algebra techniques. Examples are presented, using the techniques discussed to solve some problems.
暂无评论