Abstract
We present two methods for the implicit integration of nonlinear stiff systems. Direct application of the Newton method to backward Euler discretization of such systems may diverge. We observe that the solution is recovered by smoothing out certain eigenvalues in the Jacobian matrix. To this end, we introduce a solution-dependent matrix-weighted combination of backward and forward Euler methods. The weight is tuned on each Newton iteration to reproduce the solution with an exponential integrator, whereby a weight function for smoothing eigenvalues is obtained. We apply the proposed techniques, namely quasi-Newton backward Euler and matrix-weighted Euler, to several stiff systems, including Lotka–Volterra, Van der Pol’s, and a blood coagulation cascade.
Funding source: Russian Science Foundation
Award Identifier / Grant number: 21-71-20024
Funding statement: This work was supported by the Russian Science Foundation through the grant 21-71-20024.
References
[1] R. Alexander, The modified Newton method in the solution of stiff ordinary differential equations, Math. Comp. 57 (1991), no. 196, 673–701. 10.1090/S0025-5718-1991-1094939-7Search in Google Scholar
[2] D. G. Anderson, Iterative procedures for nonlinear integral equations, J. Assoc. Comput. Mach. 12 (1965), 547–560. 10.1145/321296.321305Search in Google Scholar
[3] L. Armijo, Minimization of functions having Lipschitz continuous first partial derivatives, Pacific J. Math. 16 (1966), 1–3. 10.2140/pjm.1966.16.1Search in Google Scholar
[4] W. Auzinger and R. Frank, Asymptotic error expansions for stiff equations: An analysis for the implicit midpoint and trapezoidal rules in the strongly stiff case, Numer. Math. 56 (1989), no. 5, 469–499. 10.1007/BF01396649Search in Google Scholar
[5] W. Auzinger, R. Frank and G. Kirlinger, A note on convergence concepts for stiff problems, Computing 44 (1990), no. 3, 197–208. 10.1007/BF02262216Search in Google Scholar
[6] W. Auzinger, R. Frank and G. Kirlinger, Modern convergence theory for stiff initial value problems, J. Comput. Appl. Math. 45 (1993), no. 1–2, 5–16. 10.1016/0377-0427(93)90260-ISearch in Google Scholar
[7] A. Bouchnita, K. Terekhov, P. Nony, Y. Vassilevski and V. Volpert, A mathematical model to quantify the effects of platelet count, shear rate, and injury size on the initiation of blood coagulation under venous flow conditions, PloS one 15 (2020), no. 7, Article ID e0235392. 10.1371/journal.pone.0235392Search in Google Scholar PubMed PubMed Central
[8] P. N. Brown, A. C. Hindmarsh and H. F. Walker, Experiments with quasi-Newton methods in solving stiff ODE systems, SIAM J. Sci. Statist. Comput. 6 (1985), no. 2, 297–313. 10.1137/0906022Search in Google Scholar
[9] P. N. Brown and Y. Saad, Convergence theory of nonlinear Newton–Krylov algorithms, SIAM J. Optim. 4 (1994), no. 2, 297–330. 10.1137/0804017Search in Google Scholar
[10] P. R. Brune, M. G. Knepley, B. F. Smith and X. Tu, Composing scalable nonlinear algebraic solvers, SIAM Rev. 57 (2015), no. 4, 535–565. 10.1137/130936725Search in Google Scholar
[11] G. Dahlquist, A special stability problem for linear multistep methods, Nordisk Tidskr. Informationsbehandling (BIT) 3 (1963), 27–43. 10.1007/BF01963532Search in Google Scholar
[12] G. Dahlquist, On stability and error analysis for stiff non-linear problems part i, Technical report CM-P00069396, 1975. Search in Google Scholar
[13] G. Guennebaud, B. Jacob, Eigen v3, http://eigen.tuxfamily.org, 2010. Search in Google Scholar
[14] L. V. Kantorovič, On Newton’s method, Trudy Mat. Inst. Steklov. 28 (1949), 104–144. Search in Google Scholar
[15] M. Y. Liu, L. Zhang and C. F. Zhang, Study on banded implicit Runge–Kutta methods for solving stiff differential equations, Math. Probl. Eng. 2019 (2019), Article ID 4850872. 10.1155/2019/4850872Search in Google Scholar
[16] A. J. Lotka, Elements of Physical Biology, Williams & Wilkins, Baltimore, 1925. Search in Google Scholar
[17] P. K. Moore and L. R. Petzold, A stepsize control strategy for stiff systems of ordinary differential equations, Appl. Numer. Math. 15 (1994), no. 4, 449–463. 10.21236/ADA281551Search in Google Scholar
[18]
Y. E. Nesterov,
A method for solving the convex programming problem with convergence rate
[19] J. M. Ortega and W. C. Rheinboldt, Iterative Solution of Nonlinear Equations in Several Variables, Classics Appl. Math. 30, Society for Industrial and Applied Mathematics, Philadelphia, 2000. 10.1137/1.9780898719468Search in Google Scholar
[20] S. Schlenkrich, A. Walther and A. Griewank, Application of AD-based quasi-Newton methods to stiff ODEs, Automatic Differentiation: Applications, Theory, and Implementations, Lect. Notes Comput. Sci. Eng. 50, Springer, Berlin (2006), 89–98. 10.1007/3-540-28438-9_8Search in Google Scholar
[21] F. Shen, C. J. Kastrup, Y. Liu and R. F. Ismagilov, Threshold response of initiation of blood coagulation by tissue factor in patterned microfluidic capillaries is controlled by shear rate, Arteriosclerosis Thrombosis Vascular Biol. 28 (2008), no. 11, 2035–2041. 10.1161/ATVBAHA.108.173930Search in Google Scholar PubMed
[22] D. C. Sorensen, Newton’s method with a model trust region modification, SIAM J. Numer. Anal. 19 (1982), no. 2, 409–426. 10.2172/6836252Search in Google Scholar
[23] M. Takesaki, Theory of Operator Algebras. I. Operator Algebras and Non-commutative Geometry 5, Encyclopaedia Math. Sci. 124, Springer, Berlin, 2002. 10.1007/978-3-662-10453-8Search in Google Scholar
[24] Y. Vassilevski, K. Terekhov, K. Nikitin and I. Kapyrin, Parallel Finite Volume Computation on General Meshes, Springer, Cham, 2020. 10.1007/978-3-030-47232-0Search in Google Scholar
[25] P. Wolfe, Convergence conditions for ascent methods, SIAM Rev. 11 (1969), 226–235. 10.1137/1011036Search in Google Scholar
© 2022 Walter de Gruyter GmbH, Berlin/Boston
