scholarly journals A DUAL ITERATIVE SUBSTRUCTURING METHOD WITH A SMALL PENALTY PARAMETER

2017 ◽  
Vol 54 (2) ◽  
pp. 461-477
Author(s):  
Chang-Ock Lee ◽  
Eun-Hee Park
2015 ◽  
Vol 32 (01) ◽  
pp. 1540006 ◽  
Author(s):  
Zhongwen Chen ◽  
Shicai Miao

In this paper, we propose a class of new penalty-free method, which does not use any penalty function or a filter, to solve nonlinear semidefinite programming (NSDP). So the choice of the penalty parameter and the storage of filter set are avoided. The new method adopts trust region framework to compute a trial step. The trial step is then either accepted or rejected based on the some acceptable criteria which depends on reductions attained in the nonlinear objective function and in the measure of constraint infeasibility. Under the suitable assumptions, we prove that the algorithm is well defined and globally convergent. Finally, the preliminary numerical results are reported.


2008 ◽  
Vol 46 (6) ◽  
pp. 2818-2842 ◽  
Author(s):  
Sven Beuchler ◽  
Tino Eibner ◽  
Ulrich Langer

2019 ◽  
Vol 97 (1) ◽  
pp. 1-13 ◽  
Author(s):  
Mubbashar Nazeer ◽  
N. Ali ◽  
Tariq Javed

The present article explores the effects of moving lid on the forced convection flow of micropolar fluid inside a right-angle triangular cavity saturated with porous medium. The base and hypotenuse or inclined sides of the cavity are maintained at constant temperatures, while the vertical side of the enclosure is adiabatic and moving with constant velocity in upward or downward direction. The flow equations are simulated by using the robust finite element numerical technique. The pressure term from the momentum equations is eliminated by using the penalty parameter. For a consistent solution, the value of the penalty parameter is selected as 107. The simulations are performed for the cases based on the direction of moving lid. The numerical outcomes are shown in terms of streamlines, temperature contours, and local and average Nusselt numbers for sundry parameters, such as micropolar parameter, Reynolds number, Richardson number, Darcy number, Hartmann number, and Prandtl number. It is observed that the shape of the inner circulating cell is elliptic when the lid is moving in the upward direction and fluid is clear (Newtonian fluid). It is also found that average Nusselt number in both cases increases with increasing Prandtl number, Richardson number, micropolar parameter, and Darcy number, whereas it decreases with increasing Hartmann number. Further, it achieves a maximum when the lid is moving in the downward direction, regardless of the choice of involved parameters. The numerical code is also validated with previous published results. The investigation of the current study is beneficial in porous heat exchangers, construction of triangular-shaped solar collectors, rigid crystal, polymeric fluid transport, etc.


2014 ◽  
Vol 543-547 ◽  
pp. 2045-2048
Author(s):  
Yuan Lv ◽  
Zhong Gan

In case of experimental data contaminated with errors and noise, the robust ε-support vector regression has good forecast accuracy and high generalization ability. However, it depends on the selection of system parameter. Firstly, this paper introduces the robust ε-support vector regression method. Secondly, as the experiments prove, the new method achieves high forecast accuracy by virtue of the optimal penalty parameter C. Finally, the optimal method of parameter C is presented in the last section.


2014 ◽  
Vol 675-677 ◽  
pp. 265-268 ◽  
Author(s):  
Yu Liu ◽  
Yu Xin Zhang

Creep landslide tunnel deformation is a diversity,changeability, less information, complicated nonlinear problem, it’s unable to establish accurate mathematical model.A creep landslide tunnel deformation prediction model based on SVM was constructed in this paper to enhance prediction accuracy, and penalty parameter c and Kernel function parameter g of SVM were optimized by genetic algorithms (GA).nine closely related factors in creep landslide tunnel deformation were selected as Input vector of SVM , creep landslide tunnel deformation measured value as a model target output. In Matlab 2011b simulation software,80 groups observation data from 2012 to 2013 of Laoyeling tunnel in Changchun to Hunchun highway of Jilin province as the sample data, 70 groups were used as training set, other 10 groups were used as testing set. The simulation result shows that testing value is very close to the true value in this method, the average relative error close to 2%. Effectiveness of the creep landslide tunnel deformation prediction based on GA_SVM model is verified by experiments.


Sign in / Sign up

Export Citation Format

Share Document