Implementation of error control for solving plane problems in linear elasticity by mixed finite elements
DOI:
https://doi.org/10.7242/1999-6691/2014.7.1.8Keywords:
a posteriori error estimates, finite element method, mixed approximations, plane strain, ANSYSAbstract
The paper is devoted to investigation of one of the modern approaches to a posteriori error control for approximate solutions of boundary-value problems. Such type of approaches provides, for the given approximate solution and problem data, a direct computation of a quantitative error estimate for deviation from the unknown exact solution of a problem. In this work, we use the functional approach proposed by S. Repin and colleagues - see, for example, books by P. Neittaanmäki and S. Repin (Elsevier, 2004) and by S. Repin (de Gruyter, 2008) and references cited therein. It allows us to determine the reliable and sufficiently sharp upper bounds to errors for a wide class of approximate solutions. To demonstrate the potentials of the approach, the well-known software for engineering computations ANSYS is employed. Most of the standard approaches, including the technique implemented in the package, are strictly applicable only to a restricted set of approximate solutions with additional mathematical properties. The analysis of a series of plane strain problems shows that the functional approach with implementation based on the Arnold-Boffi-Falk mixed finite element (SIAM J. Numer. Anal., 2005, vol. 42, no. 6, pp. 2429-2451) significantly extends the capabilities of the standard methodology of the package. Numerical calculations are based on examples proposed in the paper by V. Manet (Composites Science and Technology, 1998. V. 58, N. 12. P. 1899-1905). The data thus obtained point to highly efficient robust estimates of the true error. At the same time, the standard procedure of the package may lead to a significant growth of overestimation of the error during the process of finite element mesh refinement.
Downloads
References
Prager W., Synge J.L. Approximations in elasticity based on the concept of function space // Quart. Appl. Math. - 1947. - V. 5, N. 3. - P. 241-269.
2. Synge J.L. The hypercircle method // Studies in numerical analysis / Ed. B.K.P. Scaife. - London: Academic Press, 1974. - P. 201-217.
3. Mihlin S.G. Variacionnye metody v matematiceskoj fizike. - M.: Nauka, 1970. - 512 s.
4. Verfurth R. A review of a posteriori error estimation and adaptive mesh-refinement techniques. - Chichester, Stuttgart: John Wiley & Teubner, 1996. - 127 p.
5. Ainsworth M., Oden J.T. A posteriori error estimation in finite element analysis. - New York: John Wiley & Sons, 2000. - 240 p.
6. Babuska I., Strouboulis T. The finite element method and its reliability. - New York: Oxford University Press, 2001. - 802 p.
7. Bangerth W., Rannacher R. Adaptive finite element methods for differential equations. Lectures in Mathematics. - ETH Zurich, Basel: Birkhauser, 2003. - 207 p.
8. Neittaanmaki P., Repin S. Reliable methods for computer simulation. Error control and a posteriori estimates. Studies in Mathematics and Its Applications. - New York: Elsevier, 2004. - 305 p.
9. Ladeveze P., Pelle J.-P. Mastering calculations in linear and nonlinear mechanics. Mechanical Engineering Series. - New York, NY: Springer, 2005. - 413 p. DOI
10. Repin S.I. A posteriori estimates for partial differential equations. Radon Series on Computational and Applied Mathematics 4. - Berlin: de Gruyter, 2008. - 316 p.
11. Szabo B., Babuska I. Introduction to finite element analysis: Formulation, verification and validation. - John Wiley & Sons, 2011. - 364 p. DOI
12. Mali O., Neittaanmaki P., Repin S. Accuracy verification methods. Theory and algorithms. Computational Methods in Applied Sciences 32. - Dordrecht: Springer, 2014. - 355 p. DOI
13. Alekseev A.K., Mahnev I.N. Ispol’zovanie lagranzevyh koefficientov pri aposteriornoj ocenke pogresnosti rasceta // Sib. zurn. vycisl. matem. - 2009. - T. 12, No 4. - S. 375-388. DOI
14. Bogolubov A.N., Malyh M.D., Panin A.A. Zavisimost’ effektivnosti aposteriornoj ocenki tocnosti priblizennogo resenia ellipticeskoj kraevoj zadaci ot vhodnyh dannyh i parametrov algoritma // Vestnik Moskovskogo Universiteta. Seria 3: Fizika. Astronomia. - 2009. - No 1. - S. 18-22. DOI
15. Frolov M.E., Curilova M.A. Adaptacia setok na osnove funkcional’nyh aposteriornyh ocenok s approksimaciej Rav’ara-Toma // ZVMMF. -2012. - T. 52, No 7. - C. 1277-1288.
16. Curilova M.A. Primenenie funkcional’nogo podhoda k adaptivnomu reseniu ellipticeskih zadac // Naucno-tehniceskie vedomosti SPbGPU. - 2012. - No 158. - S. 64-69.
17. Korneev V.G. Kontrol’ pogresnosti cislennyh resenij kraevyh zadac mehaniki splosnoj sredy // Naucno-tehniceskie vedomosti SPbGPU. - 2009. - T. 4, No 88. - S. 31-43.
18. Korneev V.G. Prostye algoritmy vycislenia klassiceskih aposteriornyh ocenok pogresnosti cislennyh resenij ellipticeskih uravnenij // Ucen. zap. Kazan. un-ta. Ser. Fiz.-matem. nauki. - 2011. - T. 153, No 4. - S. 11-27.
19. Zolotareva N.D., Nikolaev E.S. Metod postroenia setok, adaptiruusihsa k reseniu kraevyh zadac dla obyknovennyh differencial’nyh uravnenij vtorogo i cetvertogo poradkov // Differencial’nye uravnenia. - 2009. - T. 45, No 8. - S. 1165-1178. DOI
20. Zolotareva N.D., Nikolaev E.S. Adaptivnaa p-versia metoda konecnyh elementov resenia kraevyh zadac dla obyknovennyh differencial’nyh uravnenij vtorogo poradka // Differencial’nye uravnenia. - 2013. - T. 49, No 7. - S. 863-876. DOI
21. Bagaev B.M., Karepova E.D., Sajdurov V.V. Setocnye metody resenia zadac s pogranicnym sloem. - Novosibirsk: Nauka, 2001. - C. 2. - 224 s.
22. Karavaev A.S., Kopysov S.P. Perestroenie nestrukturirovannyh cetyrehugol’nyh i smesannyh setok // Vestnik Udmurtskogo universiteta. Matematika. Mehanika. Komp’uternye nauki. - 2013. - No 4. - S. 62-78.
23. Karavaev A.S., Kopysov S.P., Ponomarev A.B. Algoritmy postroenia i perestroenia nestrukturirovannyh cetyrehugol’nyh setok v mnogosvaznyh oblastah // Vycisl. meh. splos. sred. - 2012. - T. 5, No 2. - S. 144-150. DOI
24. Kopysov S.P., Novikov A.K. Metod dekompozicii dla parallel’nogo adaptivnogo konecno-elementnogo algoritma // Vestnik Udmurtskogo universiteta. Matematika. Mehanika. Komp’uternye nauki. - 2010. - No 3. - S. 141-154.
25. Frolov M.E. Primenenie funkcional’nyh ocenok pogresnosti so smesannymi approksimaciami k ploskim zadacam linejnoj teorii uprugosti // ZVMMF. - 2013. - T. 53, No 7. - S. 1178-1191. DOI
26. Arnold D.N., Boffi D., Falk R.S. Quadrilateral H(div) finite elements // SIAM J. Numer. Anal. - 2005. - V. 42, N. 6. - P. 2429-2451. DOI
27. Muzalevsky A.V., Repin S.I. On two-sided error estimates for approximate solutions of problems in the linear theory of elasticity // Russ. J. Numer. Anal. M. - 2003. - V. 18, N. 1. - P. 65-85. DOI
28. Duran R.G. Mixed finite element methods. Mixed finite elements, compatibility conditions, and applications // Lecture Notes in Mathematics. - 2008. - V. 1939. - P. 1-44. DOI
29. Zienkiewicz O.C., Zhu J.Z. A simple error estimator and adaptive procedure for practical engineering analysis // Int. J. Numer. Meth. Eng. - 1987. - V. 24, N. 2. - P. 337-357. DOI
30. Oganesan L.A., Ruhovec L.A. Issledovanie skorosti shodimosti variacionno-raznostnyh shem dla ellipticeskih uravnenij vtorogo poradka v dvumernoj oblasti s gladkoj granicej // ZVMMF. - 1969. - T. 9. - S. 1102-1120. DOI
31. Manet V. The use of ANSYS to calculate the behaviour of sandwich structures // Compos. Sci. Technol. - 1998. - V. 58, N. 12. - P. 1899-1905. DOI
Downloads
Published
Issue
Section
License
Copyright (c) 2014 Computational Continuum Mechanics
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.