PL
 |
EN
Nowa wersja platformy, zawierająca wyłącznie zasoby pełnotekstowe, jest już dostępna.
Przejdź na https://bibliotekanauki.pl
Preferencje help
Polish version English version Język
Widoczny [Schowaj] Abstrakt
Liczba wyników

Znaleziono wyników: 9

Liczba wyników na stronie
first rewind previous Strona / 1 next fast forward last
Wyniki wyszukiwania
Wyszukiwano:
w słowach kluczowych:  computational efficiency
help Sortuj według:

help Ogranicz wyniki do:
first rewind previous Strona / 1 next fast forward last
1
Content available remote Efficient simulation of unidirectional pulse propagation in high-contrast nonlinear nanowaveguides
100%
Andreasen J. , Kolesik M.
|
|
tom 2
157-165
EN
This work demonstrates an improved method to simulate long-distance femtosecond pulse propagation in highcontrast nanowaveguides. Different from typical beam propagation methods, the foundational tool here is capable of simulating strong spatiotemporal waveform reshaping and extreme spectral dynamics. Meanwhile, the ability to fully capture effects due to index contrast in the transverse direction is retained, without requiring a decomposition of the electric field in terms of waveguide modes. These simulations can be computationally expensive, however, so cost is reduced in the improved method by considering only the waveguide core. Fields in the cladding are then properly accounted for through a boundary condition suitable for the case of total internal reflection.
2
Content available Application of reanalysis methods in structural mechanics
86%
Delyová I. , Frankovský P. , Bocko J. , Sivák P. , Kurimský R.
|
2022
|
tom Vol. 27, no. 3
49--62
EN
When designing structures, it is often necessary to re-analyse a structure that is different in some parts from the original one. As real structures are often complex, their analysis is therefore very challenging. In such cases, reanalysis methods are advantageously used. The aim of this paper is to approach the problem of solving the constructions using reanalysis method in which the time taken in solving algebraic equations is reduced. In particular, the purpose of this work is to demonstrate on a chosen system the time savings and the advantages of the chosen direct efficient reanalysis method for a given design problem. A basic condition for meeting these criteria is the modernization of computational procedures in the mechanics of compliant solids.
3
Content available Numerical simulation of 3D flow in axial turbomachines
72%
Yershov S. , Rusanov A. V. , Gardzielewicz A. , Lampart P. , Świrdyczuk J.
TASK Quarterly : scientific bulletin of Academic Computer Centre in Gdansk
|
1998
|
tom Vol. 2, No 2
319-347
EN
The paper is intended to describe a method for the calculation of 3D viscous compressible (subsonic or supersonic) flow in axial turbomachines described in the form of thin-layer Reynolds-averaged Navier-Stokes equations. The method draws on Godunov-type upwind differencing and ENO reconstruction suggested by Harten (1987), so as to assure monotonicity preservation and high accuracy of computational results. The computational efficiency is achieved thanks to the implementation of a simplified H-type multi-grid approach and delta -form implicit step. Turbulent effects are simulated with the help of a modified algebraic model of Baldwin-Lomax (1978). This method was at the foundation of a computer code-a complex software package to calculate 3D flow in multi-stage turbomachines that allows us to obtain local characteristics, like temperature, pressure, density or velocity distributions, as well as global characteristics, such as flow rates, stage reaction, flow efficiency for the considered turbine/compressor stage. The paper also gives selected results of computation of a number of turbomachinery cascades, showing that these results agree reasonably well with the available experimental data.
4
Content available remote Evaluation of parallel efficiency in modeling of mechanisms using commercial multibody solvers
72%
Malczyk P. , Frączek J.
|
|
tom Vol. LVI, nr 3
237-249
EN
Parallel computers are becoming more available. The natural way to improve computational efficiency of multibody simulations seems to be parallel processing. Within this work we are trying to estimate the efficiency of parallel computations performed using one of the commercial multibody solver. First, the short theoretical outline is presented to give the overview of modeling issues in multibody dynamics. Next, the experimental part is demonstrated. The series of dynamics analyses are carried out. The test mechanisms with variable number of bodies are used to gather the performance results of the solver. The obtained data allow for estimating the number of bodies, which are sufficient to gain benefits from parallel computing as well as their level. The parallel processing profits are taken into account in the case of contact forces present in the system. The performance benefits are indicated, when the multilink belt chain is simulated, in which contact forces are included in the model.
PL
W związku z szybkim rozwojem komputerów równoległych naturalną drogą poprawienia wydajności obliczeń symulacyjnych staje się wykorzystywanie możliwości przetwarzania równoległego. W ramach niniejszej pracy podjęto próbę oceny efektywności obliczeń równoległych, w przypadku obliczeń jednym z pakietów komercyjnych do modelowania układów mechanicznych. Początkowo zaprezentowano podstawy teoretyczne metody układów wieloczłonowych. Następnie zademonstrowano wyniki eksperymentalne. W pracy zrealizowano obliczenia dynamiki testowych układów mechanicznych o zmiennej liczbie członów, podczas których zarejestrowano czasy obliczeń sekwencyjnych i równoległych. Uzyskane charakterystyki wydajnościowe pozwoliły oszacować liczbę ogniw, przy której otrzymuje się zyski z obliczeń równoległych, a także wskazać ich poziom. W zakresie badań uwzględniono potencjalne korzyści ze stosowania przetwarzania równoległego w przypadku sił kontaktu, obecnych w modelu, na przykładzie symulacji wieloczłonowego mechanizmu łańcuchowego.
5
Content available remote Wykorzystanie metody CMS do budowy efektywnego numerycznie modelu MES
72%
Danielczyk P.
|
|
tom R. 89, nr 7
674--675
PL
W pracy przedstawiono efektywny numerycznie model obliczeniowy do rozwiązania zadania optymalizacji parametrycznej z wykorzystaniem metody elementów skończonych. Do budowy modelu obliczeniowego użyto superelementów, a do redukcji rozmiaru zadania wykorzystano metodę Component Mode Synthesis. Pokazano korzyści wynikające z przyjętej metodyki modelowania w odniesieniu do czasu potrzebnego na znalezienie rozwiązania optymalnego oraz wielkości plików generowanych podczas analizy.
EN
The paper presents a numerically efficient computational model for solving a parametric optimization problem with the use of the Finite Element Method. Building a computational model involved applying superelements and the CMS method was applied for model reduction. The advantages of the adopted modelling methodology were shown with reference to the time required to find the optimal solution and the size of the files generated during the analysis.
6
Content available remote Optimal design of control law under uncertainty by dynamic programming
72%
Siomak M. , Malinowski K.
Archives of Control Sciences
|
1998
|
tom Vol. 7, no. 3/4
202-226
EN
In this paper various aspects of the realization of the dynamic programming algorithm are presented. Particular attention is focused on computational efficiency, which is obtained by presented discretization, approximation and parallelization techniques. Numerical solutions of some examples are briefly described and discussed. The last part presents a particular approach to handle stochastic constraints with illustrative examples. All numerical results were obtained on the CRAY 6416 super-server installed in Computing Center of the Warsaw University of Technology.
7
Content available remote Computing Duals of Finite Gödel Algebras
72%
Codara P. , Maurina G. , Valota D.
Annals of Computer Science and Information Systems
|
2020
|
tom Vol. 21
31--34
EN
We introduce an algorithm that computes and counts the duals of finite G\"odel-Dummett algebras of k ≥ 1 elements. The computational cost of our algorithm depends on the factorization of k, nevertheless a Python implementation is sufficiently fast to compute the results for very large values of k.
8
Content available remote Generalisation and improvement of the compact gravity inversion method
58%
Zhu W. , Peng J. , Luo S. , Meng X. , Liu J. , Zhu C.
|
|
tom Vol. 68, no. 6
1667--1677
EN
Compact gravity inversion (CGI) is widely used to invert gravity data following the principle of minimising the volume of the causative body due to its simplicity, high efciency, and sharp-boundary inversion results. In this study, the compactness weighting function is generalised and the depth weighting function is introduced to CGI to obtain the reweighted CGI (RCGI) method. Although RCGI exhibits better fexibility than CGI, selecting an appropriate compactness factor α and depth weighting function β is difcult, and we design a parameter selection rule to search the proper α and β quantitively. Furthermore, we improve RCGI for boasting superior computational efciency by gradually eliminating the model blocks that reach the designated boundaries in the iterative algorithm of inversion. This approach is termed the reweighted and element-elimination CGI (REECGI) method. The inversion results show that both RCGI and REECGI result in better inversion accuracy than CGI, and REECGI has higher computational efciency than RCGI and CGI, which increases with the number of iterations.
9
Content available remote A chance-constraint approach for optimizing Social Engagement-based services
58%
Bierlaire M. , Fadda E. , Fotio Tiotsop L. , Manerba D.
Annals of Computer Science and Information Systems
|
2022
|
tom Vol. 30
301--304
EN
Social Engagement is a novel business model whose goal is transforming final users of a service from passive components into active ones. In this framework, people are contacted by the decision-maker (generally a company) and they are asked to perform tasks in exchange for a reward. This paves the way to the interesting optimization problem of allocating the different types of workforce so as to minimize costs. Despite this problem has been investigated within the operations research community, there is no model that allows to solve it by explicitly and appropriately modeling the behavior of contacted candidates through consolidated concepts from utility theory. This work aims at filling this gap. We propose a stochastic optimization model including a chance constraint that puts in relation, under probabilistic terms, the candidate willingness to accept a task and the reward actually offered by the decision-maker. The proposed model aims at optimally deciding which user to contact, the amount of the reward proposed, and how many employees to use in order to minimize the total expected costs of the operations. A solution approach is proposed to address the formulated stochastic optimization problem and its computational efficiency and effectiveness are investigated through an extensive set of computational experiments.
first rewind previous Strona / 1 next fast forward last
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.