Abstract. This paper is concerned with the implementation and numerical study of a discrete negative norm least-squares method for the Navier-Stokes equations proposed in [2] and [3]. The main focus of the paper is on the algorithmic development and comput
Electronic Transactions on Numerical Analysis.Volume 6, pp. 44-62, December 1997.Copyright©1997, Kent State University.ISSN 1068-9613.
Kent State University etna@mcs.kent.edu
EXPERIENCESWITHNEGATIVENORMLEAST–SQUAREMETHODSFOR
THENAVIER–STOKESEQUATIONS
P.BOCHEV
Abstract.Thispaperisconcernedwiththeimplementationandnumericalstudyofadiscretenegativenormleast-squaresmethodfortheNavier-Stokesequationsproposedin[2]and[3].Themainfocusofthepaperisonthealgorithmicdevelopmentandcomputationalanalysisofthismethod,includingdesignofef cientpreconditioners,numericalestimatesofconvergencerates,etc.Ourexperimentsindicatethatthenegativenormmethodyieldsresultsthatareinagreementwiththetheoreticalerrorestimatesof[3]andcomparefavorablywiththebenchmarkstudiesof[11].
Keywords.Navier-Stokesequations,least-squaresprinciple, niteelementmethods.AMSsubjectclassi cations.76D05,76D07,65F10,65F30.
1.Introduction.Inthispaperweexaminealgorithmicandcomputationalissuesre-latedtoanegativenormleast-squaresmethodforthenumericalsolutionofthestationary,incompressibleNavier-Stokesequations.Intherecentyearsmethodsofleast-squarestypefor uid owproblemshavebeenreceivingincreasingattention;seee.g.,[1]-[9],[15],[16],[17],and[18]amongothers.Thisinterestislargelycausedbytheattractiveanalyticandcomputationalfeaturesofleast-squaresmethodsthatarenotpresentinotherdiscretizationschemes,suchasmixedGalerkinmethods.Mostofthesefeaturesstemfromthefactthatweakvariationalproblemsinleast-squaresmethodsrepresentnecessaryminimumconditionsforproblem-dependentfunctionalswhicharede nedbysummingupresidualnormsofthedifferentialequations.Theguidingprincipleinthechoiceofthenormsistoobtainnorm-equivalentfunctionals.Then,correspondingweakproblemsareingeneralcoerciveandtheirdiscretizationleadstosymmetricandpositivede nitealgebraicsystems.Speci cally,inthecontextoftheNavier-Stokesequationsapplicationofleast-squaresvariationalprinciplesof-fersthefollowingadvantages:
methodsarenotsubjecttotheinf-sup(LBB)stabilitycondition;see[12]and[14]; usedinconjunctionwithNewtonlinearizationleast-squaresleadtosymmetric,pos-itivede nitelinearsystems,atleastinaneighborhoodofthesolution; essentialboundaryconditionscanbeenforcedinaweak,variationalsense.Asaresult,
asingleapproximatingspacecanbeusedforboththevelocityandthepressureleadingtosimpli edandmoreef cientalgorithmicdesign;
solutionofthelinearizedproblemscanbeaccomplishedbyrobustandef cientiter-ativemethodswithoutassemblingthediscretizationmatrix;
approximatingspacesarenotsubjecttotheessentialboundaryconditions.
However,withoutathoroughexaminationofthesettingsfortheleast-squaresmethodmanyoftheseadvantagesmaybelostorutilizedincompletely.Forexample,aleast-squaresmethodbasedontheprimitivevariableNavier-Stokesequationsmayrequireconformingdiscretiza-tionsbycontinuouslydifferentiable niteelementspaces,i.e.,itwillbeimpractical.Further-more,iftheleast-squaresfunctionalisnotnorm-equivalentthenresultingmethodsmayfailtobeoptimallyaccurate.
Abstract. This paper develops least-squares methods for the solution of linear elastic problems in both two and three dimensions. Our main approach is ...
Block SOR methods for rank-deficient least-squares problems_理学_高等教育_...norm solution, we shall derive a new system for the minimum 2-norm ...
t using methods of linear least squares. The m-function to be used with lm.m is simply: 1 2 function y_hat = lm_func (t ,p , c ) y_hat ...
暂无评价 19页 免费 On the least squares app... 21页 免费...of the norm on F . Using (1.2) it follows
An Algorithm for Least-Squares Estimation of Nonlinear Parameters_专业资料。... Regularization methods... 15页 免费 Optimal rates for regu... 暂无评价 ...
least squares (IRLS) method (Lee et al. 2006;...1 (called negative). We can see from (4) ...Extensive comparison with many existing methods ...
Least squares and least norm in Matlab_理学_高等教育_教育专区。math---matlabEE263 Prof. S. Boyd Least squares and least norm in Matlab Least squares...
least squares approximate solution of an overdetermined system ? least norm ...standard methods for computing P ?1 (m + 1) require O(n3 ) operations...
which consists of the combination of SVD and EVD methods, has several advantages in solving the normal equations of a least-squares problem: (1) The ...
热门文档