Skip to main content
Log in

On explicit two-derivative two-step Runge–Kutta methods

  • Published:
Computational and Applied Mathematics Aims and scope Submit manuscript

Abstract

We introduce a class of methods for the numerical solution of ordinary differential equations. These methods called as two-derivative two-step Runge–Kutta methods are extension of the two-step Runge–Kutta methods in which the second derivative of the solution is included. These methods are a special class of second-derivative general linear methods studied by many authors Butcher et al. (Numer Algorithms 40:415–429, 2005), Abdi et al. (Numer Algorithms 57:149–167, 2011), Okuonghae and Ikhile (Numer Algorithms 67(3):637–654, 2014). The order conditions are derived based on the algebraic theory of Butcher (Mathe Comput 26:79–106, 1972) and the \(\mathcal {B}-\)series theory Hairer and Wanner (Computing 13:1–15, 1974), in a similar way to Chan and Chan (2006). In this study, special explicit two-derivative two-step Runge–Kutta methods that possess one evaluation of the first derivative and many evaluations of the second derivative per step are introduced. Methods with stages up to five and of order up to eight are presented. The numerical calculations have been performed on some non-stiff and mildly stiff problems and comparisons have been made with the accessible methods in the literature.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11

Similar content being viewed by others

References

  • Abdi A, Hojjati G (2011a) An extension of general linear methods. Numerical Algorithms 57:149–167

  • Abdi A, Hojjati G (2011b) Maximal order for second derivative general linear methods with Runge-Kutta stability. Applied Numerical Mathematics 61:1046–1058

  • Abdi A, Hojjati G (2015) Implementation of nordsieck second derivative methods for stiff odes. Applied Numerical Mathematics 94:241–253

    Article  MathSciNet  MATH  Google Scholar 

  • Akanbi MA, Okunuga SA, Sofoluwe AB (2012) Step size bounds for a class of multiderivative explicit Runge-Kutta methods, Modeling and Simulation in Engineering Economics and Management 188–197, Springer, Berlin

  • Albrecht P (1985) Numerical treatment of o.d.e.s: the theory of a-methods. Numerische Mathematik 47:59–87

    Article  MathSciNet  MATH  Google Scholar 

  • Albrecht P (1987) A new theoretical approach to Runge-Kutta methods. SIAM Journal on Numerical Analysis 24(2):391–406

    Article  MathSciNet  MATH  Google Scholar 

  • Bartoszewski Z, Jackiewicz Z (1998) Construction of two-step Runge-Kutta methods of high order for ordinary differential equations. Numerical Algorithms 18:51–70

    Article  MathSciNet  MATH  Google Scholar 

  • Burrage K (1988) Order properties of implicit multivalue methods for ordinary differential equations. IMA Journal of Numerical Analysis 8:43–69

    Article  MathSciNet  MATH  Google Scholar 

  • Butcher JC (1972) An algebraic theory of integration methods. Mathematics of Computation 26:79–106

    Article  MathSciNet  MATH  Google Scholar 

  • Butcher JC (1987) The Numerical Analysis of Ordinary Differential Equations. :Runge-Kutta and General Linear Methods. Wiley, New York

    MATH  Google Scholar 

  • Butcher JC (2008) Numerical Methods for Ordinary Differential Equations. John Wiley & Sons, Chichester

    Book  MATH  Google Scholar 

  • Butcher JC (2010) Trees and numerical methods for ordinary differential equations. Numerical Algorithms 53:153–170

    Article  MathSciNet  MATH  Google Scholar 

  • Butcher JC, Hojjati G (2005) Second derivative methods with RK stability. Numerical Algorithms 40:415–429

    Article  MathSciNet  MATH  Google Scholar 

  • Butcher JC, Tracogna S (1997) Order conditions for two-step Runge-Kutta methods. Applied Numerical Mathematics 24:351–364

    Article  MathSciNet  MATH  Google Scholar 

  • Byrne GD, Lambert RJ (1966) Pseudo Runge-Kutta methods involving two points. Journal of the ACM (JACM) 13:114–123

    Article  MathSciNet  MATH  Google Scholar 

  • Cash JR (1981) Second derivative extended backward differentiation formulas for the numerical integration of stiff systems. SIAM Journal on Numerical Analysis 18(1):21–36

    Article  MathSciNet  MATH  Google Scholar 

  • Chan TMH, Chan RPK (2006) A simplified approach to the order conditions of integration methods. Computing 77(3):237–252

    Article  MathSciNet  MATH  Google Scholar 

  • Chan RPK, Tsai AYJ (2010) On explicit two-derivative Runge-Kutta methods. Numerical Algorithms 53:171–194

    Article  MathSciNet  MATH  Google Scholar 

  • Chollom J, Jackiewicz Z (2003) Construction of two-step Runge-Kutta methods with large regions of absolute stability. Journal of Computational and Applied Mathematics 157:125–137

    Article  MathSciNet  MATH  Google Scholar 

  • D’Ambrosio R, Esposito E, Paternoster B (2012) Exponentially fitted two-step Runge-Kutta methods: construction and parameter selection. Applied Mathematics and Computation 218(14):7468–7480

    Article  MathSciNet  MATH  Google Scholar 

  • Enright WH (1974) Second derivative multistep methods for stiff ordinary differential equations. SIAM Journal on Numerical Analysis 11(2):321–331

    Article  MathSciNet  MATH  Google Scholar 

  • Ezzeddine AK, Hojjati G (2012) Third derivative multistep methods for stiff systems. International Journal of Nonlinear Science 14(4):443–450

    MathSciNet  MATH  Google Scholar 

  • Gear CW (1971) Numerical Initial Value Problems in Ordinary Differential Equations. Prentice-Hall, Englewood Cliffs, NJ

    MATH  Google Scholar 

  • Gekeler E, Widmann R (1986) On the order conditions of Runge-Kutta methods with higher derivatives. Numerische Mathematik 50(2):183–203

    Article  MathSciNet  MATH  Google Scholar 

  • Goeken D, Johnson O (1999) Fifth-order Runge-Kutta with higher order derivative approximations. Electronic Journal of Differential Equations Conference 02:1–9

    MathSciNet  MATH  Google Scholar 

  • Goeken D, Johnson O (2000) Runge-Kutta with higher order derivative approximations. Applied Numerical Mathematics 34(2–3):207–218

    Article  MathSciNet  MATH  Google Scholar 

  • Hairer E, Wanner G (1973) Multistep-multistage-multiderivative methods for ordinary differential equations. Computing 11(3):287–303

    Article  MathSciNet  MATH  Google Scholar 

  • Hairer E, Wanner G (1974) On the butcher group and general multi-value methods. Computing 13:1–15

    Article  MathSciNet  MATH  Google Scholar 

  • Hairer E, Wanner G (1996) Solving Ordinary Differential Equations II. Stiff and Differential-Algebraic Problems, 2nd edn., Springer Series in Computational Mathematics, vol. 14. Springer, Berlin

  • Hairer E, Wanner G (1997) Order conditions for general two-step Runge-Kutta methods. SIAM Journal on Numerical Analysis 34(6):2087–2089

    Article  MathSciNet  MATH  Google Scholar 

  • Hairer E, Nørsett SP, Wanner G (1993) Solving Ordinary Differential Equations I: Nonstiff Problems, 2nd edn., Springer Series in Computational Mathematics, vol. 8. Springer, Berlin

  • Hojjati G, Rahimi Ardabili MY, Hosseini SM (2006) New second derivative multistep methods for stiff systems. Applied Mathematical Modelling 30(5):466–476

    Article  MATH  Google Scholar 

  • Jackiewicz Z (2009) General Linear Methods for Ordinary Differential Equations. John Wiley & Sons, Hoboken, New Jersey

    Book  MATH  Google Scholar 

  • Jackiewicz Z, Tracogna S (1995) A general class of two-step Runge-Kutta methods for ordinary differential equations. SIAM Journal on Numerical Analysis 32(5):1390–1427

    Article  MathSciNet  MATH  Google Scholar 

  • Jackiewicz Z, Verner JH (2002) Derivation and implementation of two-step Runge-Kutta pairs. Japan Journal of Industrial and Applied Mathematics 19:227–248

    Article  MathSciNet  MATH  Google Scholar 

  • Jackiewicz Z, Renaut R, Feldstein A (1991) Two-step Runge-Kutta methods. SIAM Journal on Numerical Analysis 28(4):1165–1182

    Article  MathSciNet  MATH  Google Scholar 

  • Jackiewicz Z, Renaut RA, Zennaro M (1995) Explicit two-step Runge-Kutta methods. Applications of Mathematics 40(6):433–456

    MathSciNet  MATH  Google Scholar 

  • Lambert JD (1973) Computational Methods in Ordinary Differential Equations. Wiley, New York

    MATH  Google Scholar 

  • Ökten Turacı M, Öziş T (2017) Derivation of three-derivative Runge-Kutta methods. Numerical Algorithms 74(1):247–265

    Article  MathSciNet  MATH  Google Scholar 

  • Okuonghae RI, Ikhile MNO (2014) Second derivative general linear methods. Numerical Algorithms 67(3):637–654

    Article  MathSciNet  MATH  Google Scholar 

  • Phohomsiri P, Udwadia FE (2004) Acceleration of Runge-Kutta integration schemes. Discrete Dynamics in Nature and Society 2004(2):307–314

    Article  MathSciNet  MATH  Google Scholar 

  • Rabiei F, Ismail F (2011) Third-order improved Runge-Kutta method for solving ordinary differential equation. International Journal of Applied Physics and Mathematics 1(3):191–194

    Article  Google Scholar 

  • Reynoso Gustavo Franco, Gottlieb Sigal,Grant Zachary J (2017) Strong stability preserving sixth order two-derivative Runge–Kutta methods, in: AIP Conference Proceedings, Vol. 1863, AIP Publishing, p. 560068

  • Rosenbrock HH (1963) Some general implicit processes for the numerical solution of differential equations. The Computer Journal 5:329–330

    Article  MathSciNet  MATH  Google Scholar 

  • Seal DC, Güçlü Y, Christlieb AJ (2014) High-order multiderivative time integrators for hyperbolic conservation laws. Journal of Scientific Computing 60:101–140

    Article  MathSciNet  MATH  Google Scholar 

  • Skvortsov LM (2010) Explicit two-step Runge-Kutta methods. Mathematical Models and Computer Simulations 2(2):222–231

    Article  MathSciNet  Google Scholar 

  • Süli E (2014) Numerical Solution of Ordinary Differential Equations. University of Oxford,

  • Tsai AYJ, Chan RPK, Wang S (2014) Two-derivative Runge-Kutta methods for PDEs using a novel discretization approach. Numerical Algorithms 65:687–703

    Article  MathSciNet  MATH  Google Scholar 

  • Udwadia FE, Farahani A (2008) Accelerated Runge-Kutta methods, Discrete Dynamics in Nature and Society 2008, 38 pages, Article ID 790619

  • Verner JH (1978) Explicit Runge-Kutta methods with estimates of the local truncation error. SIAM Journal on Numerical Analysis 15(4):772–790

    Article  MathSciNet  MATH  Google Scholar 

  • Watt JM (1967) The asymptotic discretization error of a class of methods for solving ordinary differential equations. Proc. Cambridge Philos. Soc. 63(2):461–472

    Article  MathSciNet  MATH  Google Scholar 

  • Wu X (2003) A class of Runge-Kutta formulae of order three and four with reduced evaluations of function. Applied Mathematics and Computation 146(2–3):417–432

    Article  MathSciNet  MATH  Google Scholar 

  • Wu X, Xia J (2006) Extended Runge-Kutta-like formulae. Applied Numerical Mathematics 56:1584–1605

    Article  MathSciNet  MATH  Google Scholar 

  • Wusu AS, Akanbi MA, Okunuga SA (2013) A three-stage multiderivative explicit Runge-Kutta method. American Journal of Computational Mathematics 3:121–126

    Article  Google Scholar 

Download references

Acknowledgements

The authors would like to thank the referees and the editor for their valuable comments and suggestions which improved the presentation of the paper.

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Mukaddes Ökten Turaci .

Additional information

Communicated by Jose Alberto Cuminato.

Appendix A

Appendix A

In this appendix, the values of \(\eta (t)\) and \(\eta _1(t)\) for trees up to order 5 and order conditions up to order 6 are given. We have assumed \(Ae=c\) and \(\hat{A}e=\hat{c}\) in the computation values of \(\eta (t)\) and \(\eta _1(t)\), which result in the stage values to be approximations of order at least one.

figure a
$$\begin{aligned} \eta (t)= & {} c,\; \eta _1(t)=c-e,\\ \alpha (t)= & {} \theta E^{-1}(t)+v^T\eta (\emptyset )+w^T\eta _1(\emptyset ).\\ \end{aligned}$$

Order condition:

\(-\theta +v^Te+w^Te=1.\)

figure b

Order condition:

$$\begin{aligned} \frac{1}{2}\theta +v^Tc+w^Tc-w^Te+\hat{v}^Te+\hat{w}^Te=\frac{1}{2}. \end{aligned}$$
figure c

Order condition:

\(-\frac{1}{3}\theta +v^Tc^2+w^Tc^2-2w^Tc+w^Te+2\hat{v}^Tc+2\hat{w}^Tc-2 \hat{w}^Te=\frac{1}{3}\).

figure d

Order condition:

\(-\frac{1}{6}\theta +v^TAc+v^T\hat{c}+\frac{1}{2}w^Te+w^TAc-w^Tc+w^T \hat{c}+\hat{v}^Tc+\hat{w}^Tc-\hat{w}^Te=\frac{1}{6}\).

figure e

Order condition:

\(\frac{1}{4}\theta +v^Tc^3+w^Tc^3-3w^Tc^2+3w^Tc-w^Te+3\hat{v}^Tc^2+3 \hat{w}^Tc^2-6\hat{w}^Tc+3\hat{w}^Te=\frac{1}{4}\).

figure f
figure g

Order condition:

$$\begin{aligned} \frac{1}{8}\theta&+v^TcAc+v^Tc\hat{c}+\frac{3}{2}w^Tc-\frac{1}{2} w^Te+w^T(cAc)-w^Tc^2-w^TAc+w^Tc\hat{c}\\&-w^T\hat{c}+\hat{v}^TAc+\hat{v}^T\hat{c}+\hat{v}^Tc^2+\frac{3}{2} \hat{w}^Te+\hat{w}^TAc-3\hat{w}^Tc+\hat{w}^T\hat{c}+\hat{w}^Tc^2=\frac{1}{8}. \end{aligned}$$
figure h

Order condition:

$$\begin{aligned} \frac{1}{12}\theta&+v^TAc^2+2v^T\hat{A}c-\frac{1}{3}w^Te+w^TAc^2 -2w^TAc+w^Tc+2w^T\hat{A}c-2w^T\hat{c}\\&+\hat{v}^Tc^2+\hat{w}^Tc^2-2\hat{w}^Tc+\hat{w}^Te=\frac{1}{12}. \end{aligned}$$
figure i

Order condition:

$$\begin{aligned} \frac{1}{24}\theta&+v^TA^2c+v^TA\hat{c}+v^T\hat{A}c-\frac{1}{6}w^Te +\frac{1}{2}w^Tc+w^TA^2c-w^TAc+w^TA\hat{c}+w^T\hat{A}c\\&-w^T\hat{c}+\hat{v}^TAc+\hat{v}^T\hat{c}+\frac{1}{2}\hat{w}^Te +\hat{w}^TAc-\hat{w}^Tc+\hat{w}^T\hat{c}=\frac{1}{24}. \end{aligned}$$
figure j

Order condition:

$$\begin{aligned} -\frac{1}{5}\theta&+v^Tc^4+w^Tc^4-4w^Tc^3+6w^Tc^2-4w^Tc+w^Te+4 \hat{v}^Tc^3+4\hat{w}^Tc^3-12\hat{w}^Tc^2\\&+12\hat{w}^Tc-4\hat{w}^Te=\frac{1}{5}. \end{aligned}$$
figure k

Order condition:

$$\begin{aligned} -\frac{1}{10}\theta&+v^Tc^2Ac+v^Tc^2\hat{c}+\frac{5}{2}w^Tc^2-2w^Tc +\frac{1}{2}w^Te+w^Tc^2Ac-2w^TcAc+w^TAc\\&-w^Tc^3+w^Tc^2\hat{c}-2w^Tc\hat{c}+w^T\hat{c}+2\hat{v}^TcAc+2\hat{v}^Tc \hat{c}+\hat{v}^Tc^3+6\hat{w}^Tc\\&-2\hat{w}^Te+2\hat{w}^TcAc-2\hat{w}^TAc-5\hat{w}^Tc^2+2\hat{w}^Tc \hat{c}-2\hat{w}^T\hat{c}+\hat{w}^Tc^3=\frac{1}{10}. \end{aligned}$$
figure l

Order condition:

$$\begin{aligned} -\frac{1}{15}\theta&+v^TcAc^2+2v^Tc\hat{A}c-\frac{4}{3}w^Tc +\frac{1}{3}w^Te+w^TcAc^2-2w^TcAc+w^Tc^2-w^TAc^2\\&+2w^TAc+2w^Tc\hat{A}c-2w^Tc\hat{c}-2w^T\hat{A}c+2w^T\hat{c}+\hat{v}^TAc^2 +2\hat{v}^T\hat{A}c+\hat{v}^Tc^3\\&-\frac{4}{3}\hat{w}^Te+\hat{w}^TAc^2-2\hat{w}^TAc+4\hat{w}^Tc+2\hat{w}^T \hat{A}c-2\hat{w}^T\hat{c}+\hat{w}^Tc^3-3\hat{w}^Tc^2=\frac{1}{15}.\\ \end{aligned}$$
figure m

Order condition:

$$\begin{aligned} -\frac{1}{30}\theta&+v^TcA^2c+v^TcA\hat{c}+v^Tc\hat{A}c-\frac{2}{3}w^Tc +\frac{1}{6}w^Te+\frac{1}{2}w^Tc^2+w^TcA^2c-w^TA^2c\\&-w^TcAc+w^TAc+w^TcA\hat{c}-w^TA\hat{c}+w^Tc\hat{A}c-w^T\hat{A}c -w^Tc\hat{c}+w^T\hat{c}\\&+\hat{v}^TA^2c+\hat{v}^TA\hat{c}+\hat{v}^T\hat{A}c+\hat{v}^TcAc +\hat{v}^Tc\hat{c}-\frac{2}{3}\hat{w}^Te+2\hat{w}^Tc+\hat{w}^TA^2c\\&-2\hat{w}^TAc+\hat{w}^TA\hat{c}+\hat{w}^T\hat{A}c-2\hat{w}^T\hat{c} +\hat{w}^TcAc-\hat{w}^Tc^2+\hat{w}^Tc\hat{c}=\frac{1}{30}. \end{aligned}$$
figure n

Order condition:

$$\begin{aligned} -\frac{1}{20}\theta&+v^T(Ac)^2+2v^T\hat{c}Ac+v^T\hat{c}^2+\frac{1}{4}w^Te +w^TAc+w^T(Ac)^2-w^Tc+w^T\hat{c}\\&-2w^TcAc+2w^T\hat{c}Ac+w^Tc^2-2w^Tc\hat{c}+w^T\hat{c}^2+2\hat{v}^TcAc +2\hat{v}^Tc\hat{c}\\&+3\hat{w}^Tc-\hat{w}^Te+2\hat{w}^TcAc-2\hat{w}^TAc-2\hat{w}^Tc^2 +2\hat{w}^Tc\hat{c}-2\hat{w}^T\hat{c}=\frac{1}{20}. \end{aligned}$$
figure o

Order condition:

$$\begin{aligned} -\frac{1}{20}\theta&+v^TAc^3+3v^T\hat{A}c^2+\frac{1}{4}w^Te+w^TAc^3 -3w^TAc^2+3w^TAc-w^Tc\\&+3w^T\hat{A}c^2-6w^T\hat{A}c+3w^T\hat{c}+\hat{v}^Tc^3+\hat{w}^Tc^3 -3\hat{w}^Tc^2+3\hat{w}^Tc-\hat{w}^Te=\frac{1}{20}. \end{aligned}$$
figure p

Order condition:

$$\begin{aligned} -\frac{1}{40}\theta&+v^TA(cAc)+v^TA(c\hat{c})+v^T\hat{A}Ac+v^T\hat{A} \hat{c}+v^T\hat{A}c^2+\frac{1}{8}w^Te+\frac{3}{2}w^TAc\\&-\frac{1}{2}w^Tc+w^TA(cAc)-w^TAc^2-w^TA^2c+w^TA(c\hat{c})-w^TA\hat{c} +\frac{3}{2}w^T\hat{c}\\&+w^T\hat{A}Ac-3w^T\hat{A}c+w^T\hat{A}\hat{c}+w^T\hat{A}c^2+\hat{v}^TcAc +\hat{v}^Tc\hat{c}+\frac{3}{2}\hat{w}^Tc-\frac{1}{2}\hat{w}^Te\\&+\hat{w}^TcAc-\hat{w}^TAc-\hat{w}^Tc^2+\hat{w}^Tc\hat{c} -\hat{w}^T\hat{c}=\frac{1}{40}. \end{aligned}$$
figure q

Order condition:

$$\begin{aligned} -\frac{1}{60}\theta&+v^TA^2c^2+2v^TA\hat{A}c+v^T\hat{A}c^2-\frac{1}{3}w^Tc +\frac{1}{12}w^Te+w^TA^2c^2-2w^TA^2c\\&+w^TAc+2w^TA\hat{A}c-2w^TA\hat{c}+w^T\hat{A}c^2-2w^T\hat{A}c+w^T\hat{c} +\hat{v}^TAc^2+2\hat{v}^T\hat{A}c\\&-\frac{1}{3}\hat{w}^Te+\hat{w}^TAc^2-2\hat{w}^TAc+\hat{w}^Tc +2\hat{w}^T\hat{A}c-2\hat{w}^T\hat{c}=\frac{1}{60}. \end{aligned}$$
figure r

Order condition:

$$\begin{aligned} -\frac{1}{120}\theta&+v^TA^3c+v^TA^2\hat{c}+v^TA\hat{A}c+v^T\hat{A} Ac+v^T\hat{A}\hat{c}+\frac{1}{24}w^Te-\frac{1}{6}w^Tc+\frac{1}{2}w^TAc\\&+w^TA^3c-w^TA^2c+w^TA^2\hat{c}+w^TA\hat{A}c-w^TA\hat{c}+\frac{1}{2}w^T \hat{c}+w^T\hat{A}Ac-w^T\hat{A}c\\&+w^T\hat{A}\hat{c}+\hat{v}^TA^2c+\hat{v}^TA\hat{c}+\hat{v}^T\hat{A}c -\frac{1}{6}\hat{w}^Te+\frac{1}{2}\hat{w}^Tc+\hat{w}^TA^2c-\hat{w}^TAc\\&+\hat{w}^TA\hat{c}+\hat{w}^T\hat{A}c-\hat{w}^T\hat{c}=\frac{1}{120}. \end{aligned}$$
figure s

Order condition:

$$\begin{aligned} \frac{1}{6}\theta&+v^Tc^5+w^Tc^5-5w^Tc^4+10w^Tc^3-10w^Tc^2+5w^Tc-w^Te+5\hat{v}^Tc^4\\&+5\hat{w}^Tc^4-20\hat{w}^Tc^3+30\hat{w}^Tc^2-20\hat{w}^Tc+5\hat{w}^Te=\frac{1}{6}.\\ \end{aligned}$$
figure t

Order condition:

$$\begin{aligned} \frac{1}{12}\theta&+v^Tc^3Ac+v^Tc^3\hat{c}+\frac{7}{2}w^Tc^3+w^Tc^3 Ac-w^Tc^4+w^Tc^3\hat{c}-\frac{9}{2}w^Tc^2-3w^Tc^2Ac\\&-3w^Tc^2\hat{c}+\frac{5}{2}w^Tc+3w^TcAc+3w^Tc\hat{c}-\frac{1}{2} w^Te-w^TAc-w^T\hat{c}+3\hat{v}^Tc^2Ac\\&+3\hat{v}^Tc^2\hat{c}+\hat{v}^Tc^4+\frac{27}{2}\hat{w}^Tc^2 +3\hat{w}^Tc^2Ac-7\hat{w}^Tc^3+3\hat{w}^Tc^2\hat{c}-10\hat{w}^Tc-6\hat{w}^TcAc\\&-6\hat{w}^Tc\hat{c}+\frac{5}{2}\hat{w}^Te+3\hat{w}^TAc+3 \hat{w}^T\hat{c}+\hat{w}^Tc^4=\frac{1}{12}. \end{aligned}$$
figure u

Order condition:

$$\begin{aligned} \frac{1}{30}\theta&+v^TAc^4+4v^T\hat{A}c^4-\frac{1}{5}w^Te+w^TAc^4 -4w^TAc^3+6w^TAc^2-4w^TAc+w^Tc\\&+4w^T\hat{A}c^3-12w^T\hat{A}c^2+12w^T\hat{A}c-4w^T\hat{c} +\hat{v}^Tc^4+\hat{w}^Tc^4-4\hat{w}^Tc^3+6\hat{w}^Tc^2\\&-4\hat{w}^Tc+\hat{w}^Te=\frac{1}{30}. \end{aligned}$$
figure v

Order condition:

$$\begin{aligned} \frac{1}{36}\theta&+v^Tc^2A^2c+v^Tc^2A\hat{c}+v^Tc^2\hat{A}c -\frac{7}{6}w^Tc^2+\frac{1}{2}w^Tc^3+w^Tc^2A^2c-w^Tc^2Ac\\&+w^Tc^2A\hat{c}+w^Tc^2\hat{A}c-w^Tc^2\hat{c}+\frac{5}{6}w^Tc -2w^TcA^2c+2w^TcAc-2w^TcA\hat{c}\\&-2w^Tc\hat{A}c+2w^Tc\hat{c}-\frac{1}{6}w^Te+w^TA^2c-w^TAc+w^TA \hat{c}+w^T\hat{A}c-w^T\hat{c}\\&+2\hat{v}^TcA^2c+2\hat{v}^TcA\hat{c}+2\hat{v}^Tc\hat{A}c +\hat{v}^Tc^2Ac+\hat{v}^Tc^2\hat{c}-\frac{10}{3}\hat{w}^Tc +\frac{7}{2}\hat{w}^Tc^2\\&+2\hat{w}^TcA^2c-4\hat{w}^TcAc+2\hat{w}^TcA\hat{c}+2\hat{w}^Tc \hat{A}c-4\hat{w}^Tc\hat{c}+\frac{5}{6}\hat{w}^Te-2\hat{w}^TA^2c\\&+3\hat{w}^TAc-2\hat{w}^TA\hat{c}-2\hat{w}^T\hat{A}c+3\hat{w}^T \hat{c}+\hat{w}^Tc^2Ac-\hat{w}^Tc^3+\hat{w}^Tc^2\hat{c}=\frac{1}{36}. \end{aligned}$$
figure w

Order condition:

$$\begin{aligned} \frac{1}{120}\theta&+v^TA^2c^3+3v^TA\hat{A}c^2+v^T\hat{A}c^3 -\frac{1}{20}w^Te+\frac{1}{4}w^Tc+w^TA^2c^3-3w^TA^2c^2\\&+3w^TA^2c-w^TAc+3w^TA\hat{A}c^2-6w^TA\hat{A}c+3w^TA\hat{c} +w^T\hat{A}c^3-3w^T\hat{A}c^2\\&+3w^T\hat{A}c-w^T\hat{c}+\hat{v}^TAc^3+3\hat{v}^T\hat{A}c^2 +\frac{1}{4}\hat{w}^Te+\hat{w}^TAc^3-3\hat{w}^TAc^2+3\hat{w}^TAc\\&-\hat{w}^Tc+3\hat{w}^T\hat{A}c^2-6\hat{w}^T\hat{A}c+3\hat{w}^T\hat{c}=\frac{1}{120}. \end{aligned}$$
figure x

Order condition:

$$\begin{aligned} \frac{1}{60}\theta&+v^TA(c^2Ac)+v^TA(c^2\hat{c})+2v^T\hat{A}(cAc) +2v^T\hat{A}(c\hat{c})+v^T\hat{A}c^3-\frac{1}{10}w^Te\\&+\frac{5}{2}w^TAc^2+w^TA(c^2Ac)-w^TAc^3+w^TA(c^2\hat{c})-2w^TAc-2w^TA(cAc)\\&-2w^TA(c\hat{c})+\frac{1}{2}w^Tc+w^TA^2c+w^TA\hat{c}+6w^T\hat{A}c +2w^T\hat{A}(cAc)-5w^T\hat{A}c^2\\&+2w^T\hat{A}(c\hat{c})-2w^T\hat{c}-2w^T\hat{A}Ac-2w^T\hat{A}\hat{c} +w^T\hat{A}c^3+\hat{v}^Tc^2Ac+\hat{v}^Tc^2\hat{c}\\&+\frac{5}{2}\hat{w}^Tc^2+\hat{w}^Tc^2Ac-\hat{w}^Tc^3+\hat{w}^T c^2\hat{c}-2\hat{w}^Tc-2\hat{w}^TcAc-2\hat{w}^Tc\hat{c}+\frac{1}{2}\hat{w}^Te\\&+\hat{w}^TAc+\hat{w}^T\hat{c}=\frac{1}{60}. \end{aligned}$$
figure y

Order condition:

$$\begin{aligned} \frac{1}{24}\theta&+v^Tc(Ac)^2+2v^Tc(\hat{c}Ac)+v^Tc\hat{c}^2+\frac{5}{4} w^Tc+3w^TcAc+w^Tc(Ac)^2-2w^Tc^2\\&+3w^Tc\hat{c}-2w^Tc^2Ac+2w^Tc(\hat{c}Ac)+w^Tc^3-2w^Tc^2\hat{c}+w^T c\hat{c}^2-\frac{1}{4}w^Te\\&-w^TAc-w^T(Ac)^2-w^T\hat{c}-2w^T\hat{c}Ac-w^T\hat{c}^2+2\hat{v}^Tc^2 Ac+2\hat{v}^Tc^2\hat{c}\\&+\hat{v}^T(Ac)^2+2\hat{v}^T\hat{c}Ac+\hat{v}^T\hat{c}^2+6\hat{w}^Tc^2 +2\hat{w}^Tc^2Ac-2\hat{w}^Tc^3+2\hat{w}^Tc^2\hat{c}\\&-5\hat{w}^Tc-6\hat{w}^TcAc-6\hat{w}^Tc\hat{c}+\frac{5}{4}\hat{w}^Te+3 \hat{w}^TAc+3\hat{w}^T\hat{c}+\hat{w}^T(Ac)^2\\&+2\hat{w}^T\hat{c}Ac+\hat{w}^T\hat{c}^2=\frac{1}{24}. \end{aligned}$$
figure z

Order condition:

$$\begin{aligned} \frac{1}{18}\theta&+v^Tc^2Ac^2+2v^Tc^2\hat{A}c-\frac{7}{3}w^Tc^2+w^Tc^2A c^2-2w^Tc^2Ac+w^Tc^3+2w^Tc^2\hat{A}c\\&-2w^Tc^2\hat{c}+\frac{5}{3}w^Tc-2w^TcAc^2+4w^TcAc-4w^Tc\hat{A}c+4w^T c\hat{c}-\frac{1}{3}w^Te\\&+w^TAc^2-2w^TAc+2w^T\hat{A}c-2w^T\hat{c}+2\hat{v}^TcAc^2+4\hat{v}^Tc \hat{A}c+\hat{v}^Tc^4-\frac{20}{3}\hat{w}^Tc\\&+2\hat{w}^TcAc^2-4\hat{w}^TcAc+8\hat{w}^Tc^2+4\hat{w}^Tc\hat{A}c-4 \hat{w}^Tc\hat{c}+\frac{5}{3}\hat{w}^Te-2\hat{w}^TAc^2\\&+4\hat{w}^TAc-4\hat{w}^T\hat{A}c+4\hat{w}^T\hat{c}+\hat{w}^Tc^4-4 \hat{w}^Tc^3=\frac{1}{18} \end{aligned}$$
figure aa

Order condition:

$$\begin{aligned} \frac{1}{24}\theta&+v^TcAc^3+3v^Tc\hat{A}c^2+\frac{5}{4}w^Tc+w^TcAc^3-3 w^TcAc^2+3w^TcAc-w^Tc^2\\&+3w^Tc\hat{A}c^2-6w^Tc\hat{A}c+3w^Tc\hat{c}-\frac{1}{4}w^Te-w^TAc^3+3 w^TAc^2-3w^TAc \\&-3w^T\hat{A}c^2+6w^T\hat{A}c-3w^T\hat{c}+\hat{v}^TAc^3+3\hat{v}^T \hat{A}c^2+\hat{v}^Tc^4+\frac{5}{4}\hat{w}^Te+\hat{w}^TAc^3\\&-3\hat{w}^TAc^2+3\hat{w}^TAc-5\hat{w}^Tc+3\hat{w}^T\hat{A}c^2-6 \hat{w}^T\hat{A}c+3\hat{w}^T\hat{c}+\hat{w}^Tc^4-4\hat{w}^Tc^3\\&+6\hat{w}^Tc^2=\frac{1}{24}. \end{aligned}$$
figure ab

Order condition:

$$\begin{aligned} \begin{aligned} \frac{1}{144}\theta&+v^TcA^3c+v^TcA^2\hat{c}+v^Tc(A\hat{A}c)+v^T c(\hat{A}Ac)+v^Tc\hat{A}\hat{c}+\frac{5}{24}w^Tc-\frac{1}{6}w^Tc^2\\&+\frac{1}{2}w^TcAc+w^TcA^3c-w^TcA^2c+w^TcA^2\hat{c}+w^Tc(A\hat{A}c)-w^T cA\hat{c}+\frac{1}{2}w^Tc\hat{c}\\&+w^Tc(\hat{A}Ac)-w^Tc\hat{A}c+w^Tc\hat{A}\hat{c}-\frac{1}{24}w^Te -\frac{1}{2}w^TAc-w^TA^3c+w^TA^2c\\&-w^TA^2\hat{c}-w^TA\hat{A}c+w^TA\hat{c}-\frac{1}{2}w^T\hat{c}-w^T\hat{A} Ac+w^T\hat{A}c-w^T\hat{A}\hat{c}+\hat{v}^TA^3c\\&+\hat{v}^TA^2\hat{c}+\hat{v}^TA\hat{A}c+\hat{v}^T\hat{A}Ac+\hat{v}^T \hat{A}\hat{c}+\hat{v}^TcA^2c+\hat{v}^TcA\hat{c}+\hat{v}^Tc\hat{A}c +\frac{5}{24}\hat{w}^Te\\&-\frac{5}{6}\hat{w}^Tc+\frac{3}{2}\hat{w}^TAc+\hat{w}^TA^3c-2\hat{w}^T A^2c+\hat{w}^TA^2\hat{c}+\hat{w}^TA\hat{A}c-2\hat{w}^TA\hat{c}\\&+\frac{3}{2}\hat{w}^T\hat{c}+\hat{w}^T\hat{A}Ac-2\hat{w}^T\hat{A}c +\hat{w}^T\hat{A}\hat{c}+\frac{1}{2}\hat{w}^Tc^2+\hat{w}^TcA^2c-\hat{w}^TcAc\\&+\hat{w}^TcA\hat{c}+\hat{w}^Tc\hat{A}c-\hat{w}^Tc\hat{c}=\frac{1}{144}. \end{aligned} \end{aligned}$$
figure ac

Order condition:

$$\begin{aligned} \begin{aligned} \frac{1}{360}\theta&+v^TA^3c^2+2v^TA^2\hat{A}c+v^TA\hat{A}c^2+v^T\hat{A} Ac^2+2v^T\hat{A}^2c-\frac{1}{60}w^Te+\frac{1}{12}w^Tc\\&-\frac{1}{3}w^TAc+w^TA^3c^2-2w^TA^3c+w^TA^2c+2w^TA^2\hat{A}c-2w^TA^2 \hat{c}+w^TA\hat{A}c^2\\&-2w^TA\hat{A}c+w^TA\hat{c}-\frac{1}{3}w^T\hat{c}+w^T\hat{A}Ac^2-2w^T \hat{A}Ac+w^T\hat{A}c+2w^T\hat{A}^2c\\&-2w^T\hat{A}\hat{c}+\hat{v}^TA^2c^2+2\hat{v}^TA\hat{A}c+\hat{v}^T \hat{A}c^2+\frac{1}{12}\hat{w}^Te-\frac{1}{3}\hat{w}^Tc+\hat{w}^TA^2c^2\\&-2\hat{w}^TA^2c+\hat{w}^TAc+2\hat{w}^TA\hat{A}c-2\hat{w}^TA\hat{c} +\hat{w}^T\hat{A}c^2-2\hat{w}^T\hat{A}c+\hat{w}^T\hat{c}=\frac{1}{360}. \end{aligned} \end{aligned}$$
figure ad

Order condition:

$$\begin{aligned} \frac{1}{240}\theta&+v^TA^2(cAc)+v^TA^2(c\hat{c})+v^TA\hat{A}Ac +v^TA\hat{A}\hat{c}+v^TA\hat{A}c^2+v^T\hat{A}(cAc)\\&+v^T\hat{A}(c\hat{c})-\frac{1}{40}w^Te+\frac{1}{8}w^Tc+\frac{3}{2} w^TA^2c-\frac{1}{2}w^TAc+w^TA^2(cAc)\\&-w^TA^2c^2-w^TA^3c+w^TA^2(c\hat{c})-w^TA^2\hat{c}+\frac{3}{2}w^TA \hat{c}+w^TA\hat{A}Ac\\&-3w^TA\hat{A}c+w^TA\hat{A}\hat{c}+w^TA\hat{A}c^2+\frac{3}{2}w^T \hat{A}c+w^T\hat{A}(cAc)-w^T\hat{A}c^2\\&+w^T\hat{A}(c\hat{c})-\frac{1}{2}w^T\hat{c}-w^T\hat{A}Ac-w^T \hat{A}\hat{c}+\hat{v}^TA(cAc)+\hat{v}^TA(c\hat{c})\\&+\hat{v}^T\hat{A}Ac+\hat{v}^T\hat{A}\hat{c}+\hat{v}^T\hat{A}c^2 +\frac{1}{8}\hat{w}^Te+\frac{3}{2}\hat{w}^TAc-\frac{1}{2}\hat{w}^Tc+\hat{w}^TA(cAc)\\&-\hat{w}^TAc^2-\hat{w}^TA^2c+\hat{w}^TA(c\hat{c})-\hat{w}^TA\hat{c} +\frac{3}{2}\hat{w}^T\hat{c}+\hat{w}^T\hat{A}Ac\\&-3\hat{w}^T\hat{A}c+\hat{w}^T\hat{A}\hat{c}+\hat{w}^T\hat{A}c^2=\dfrac{1}{240}. \end{aligned}$$
figure ae

Order condition:

$$\begin{aligned} \frac{1}{120}\theta&+v^TA(Ac)^2+2v^TA(\hat{c}Ac)+v^TA\hat{c}^2+2v^T \hat{A}(cAc)+2v^T\hat{A}(c\hat{c})-\frac{1}{20}w^Te\\&+\frac{1}{4}w^Tc+w^TA(Ac)^2+w^TAc^2+w^TA\hat{c}^2+w^TA^2c-w^T Ac+w^TA\hat{c}\\&-2w^TA(cAc)+2w^TA(\hat{c}Ac)-2w^TA(c\hat{c})+3w^T\hat{A}c+2w^T\hat{A}(cAc)\\&-2w^T\hat{A}c^2+2w^T\hat{A}(c\hat{c})-w^T\hat{c}-2w^T\hat{A}Ac-2w^T \hat{A}\hat{c}+\hat{v}^T(Ac)^2 \\&+2\hat{v}^T\hat{c}Ac+\hat{v}^T\hat{c}^2+\frac{1}{4}\hat{w}^Te+\hat{w}^T (Ac)^2+\hat{w}^Tc^2+\hat{w}^T\hat{c}^2+\hat{w}^TAc\\&-\hat{w}^Tc+\hat{w}^T\hat{c}-2\hat{w}^TcAc+2\hat{w}^T\hat{c}Ac-2 \hat{w}^Tc\hat{c}=\frac{1}{120}. \end{aligned}$$
figure af

Order condition:

$$\begin{aligned} \frac{1}{180}\theta&+v^TA(cA^2c)+v^TA(cA\hat{c})+v^TA(c\hat{A}c)+v^T\hat{A} A^2c+v^T\hat{A}A\hat{c}+v^T\hat{A}^2c\\&+v^T\hat{A}(cAc)+v^T\hat{A}(c\hat{c})-\frac{1}{30}w^Te-\frac{2}{3}w^TAc +\frac{1}{2}w^TAc^2+w^TA(cA^2c)\\&-w^TA(cAc)+w^TA(cA\hat{c})+w^TA(c\hat{A}c)-w^TA(c\hat{c})+\frac{1}{6}w^Tc-w^TA^3c\\&+w^TA^2c-w^TA^2\hat{c}-w^TA\hat{A}c+w^TA\hat{c}-\frac{2}{3}w^T\hat{c} +2w^T\hat{A}c+w^T\hat{A}A^2c\\&-2w^T\hat{A}Ac+w^T\hat{A}A\hat{c}+w^T\hat{A}^2c-2w^T\hat{A}\hat{c}+w^T \hat{A}(cAc)-w^T\hat{A}c^2\\&+w^T\hat{A}(c\hat{c})+\hat{v}^TcA^2c+\hat{v}^TcA\hat{c}+\hat{v}^Tc\hat{A}c -\frac{2}{3}\hat{w}^Tc+\frac{1}{2}\hat{w}^Tc^2+\hat{w}^TcA^2c\\&-\hat{w}^TcAc+\hat{w}^TcA\hat{c}+\hat{w}^Tc\hat{A}c-\hat{w}^Tc\hat{c} +\frac{1}{6}\hat{w}^Te-\hat{w}^TA^2c+\hat{w}^TAc\\&-\hat{w}^TA\hat{c}-\hat{w}^T\hat{A}c+\hat{w}^T\hat{c}=\frac{1}{180}. \end{aligned}$$
figure ag

Order condition:

$$\begin{aligned} \frac{1}{72}\theta&+v^T(Ac)(A^2c)+v^T(Ac)(A\hat{c})+v^T(Ac)(\hat{A}c) +v^T\hat{c}A^2c+v^T\hat{c}A\hat{c}+v^T\hat{c}\hat{A}c\\&-\frac{1}{12}w^Te+\frac{5}{12}w^Tc+\frac{1}{2}w^TA^2c-\frac{2}{3}w^TAc +\frac{1}{2}w^TA\hat{c}+\frac{1}{2}w^T\hat{A}c-\frac{2}{3}w^T\hat{c}\\&+\frac{3}{2}w^TcAc+w^T(Ac)(A^2c)-w^T(Ac)^2+w^T(Ac)(A\hat{c})+w^T(Ac)(\hat{A}c)\\&-2w^T\hat{c}Ac-\frac{1}{2}w^Tc^2-w^TcA^2c-w^TcA\hat{c}-w^Tc\hat{A}c +\frac{3}{2}w^Tc\hat{c}+w^T\hat{c}A^2c\\&+w^T\hat{c}A\hat{c}+w^T\hat{c}\hat{A}c-w^T\hat{c}^2+\hat{v}^TcA^2c +\hat{v}^TcA\hat{c}+\hat{v}^Tc\hat{A}c+\hat{v}^T(Ac)^2\\&+2\hat{v}^T\hat{c}Ac+\hat{v}^T\hat{c}^2-\frac{5}{3}\hat{w}^Tc +\frac{3}{2}\hat{w}^Tc^2+\hat{w}^TcA^2c-3\hat{w}^TcAc+\hat{w}^TcA\hat{c}\\&+\hat{w}^Tc\hat{A}c-3\hat{w}^Tc\hat{c}+\frac{5}{12}\hat{w}^Te -\hat{w}^TA^2c+2\hat{w}^TAc-\hat{w}^TA\hat{c}-\hat{w}^T\hat{A}c+2\hat{w}^T\hat{c}\\&+\hat{w}^T(Ac)^2+\hat{w}^T\hat{c}^2+2\hat{w}^T\hat{c}Ac=\frac{1}{72}. \end{aligned}$$
figure ah

Order condition:

$$\begin{aligned} \frac{1}{48}\theta&+v^TcA(cAc)+v^TcA(c\hat{c})+v^Tc(\hat{A}Ac)+v^T c\hat{A}\hat{c}+v^Tc\hat{A}c^2+\frac{5}{8}w^Tc+\frac{3}{2}w^TcAc\\&-\frac{1}{2}w^Tc^2+w^TcA(cAc)-w^TcAc^2-w^TcA^2c+w^TcA(c\hat{c})-w^TcA\hat{c} +\frac{3}{2}w^Tc\hat{c}\\&+w^Tc(\hat{A}Ac)-3w^Tc\hat{A}c+w^Tc\hat{A}\hat{c}+w^Tc\hat{A}c^2 -\frac{1}{8}w^Te-\frac{3}{2}w^TAc-w^TA(cAc)\\&+w^TAc^2+w^TA^2c-w^TA(c\hat{c})+w^TA\hat{c}-\frac{3}{2}w^T\hat{c} -w^T\hat{A}Ac+3w^T\hat{A}c-w^T\hat{A}\hat{c}\\&-w^T\hat{A}c^2+\hat{v}^TA(cAc)+\hat{v}^TA(c\hat{c})+\hat{v}^T\hat{A}A c+\hat{v}^T\hat{A}\hat{c}+\hat{v}^T\hat{A}c^2+\hat{v}^Tc^2Ac\\&+\hat{v}^Tc^2\hat{c}+\frac{5}{8}\hat{w}^Te+\frac{5}{2}\hat{w}^TAc -\frac{5}{2}\hat{w}^Tc+\hat{w}^TA(cAc)-\hat{w}^TAc^2-\hat{w}^TA^2c\\&+\hat{w}^TA(c\hat{c})-\hat{w}^TA\hat{c}+\frac{5}{2}\hat{w}^T\hat{c} +\hat{w}^T\hat{A}Ac-3\hat{w}^T\hat{A}c+\hat{w}^T\hat{A}\hat{c} +\hat{w}^T\hat{A}c^2\\&+\frac{5}{2}\hat{w}^Tc^2+\hat{w}^Tc^2Ac-\hat{w}^Tc^3+\hat{w}^Tc^2 \hat{c}-2\hat{w}^TcAc-2\hat{w}^Tc\hat{c}=\frac{1}{48}. \end{aligned}$$
figure ai

Order condition:

$$\begin{aligned} \frac{1}{90}\theta&+v^TA(cAc^2)+2v^TA(c\hat{A}c)+v^T\hat{A}Ac^2+2v^T \hat{A}^2c+v^T\hat{A}c^3-\frac{1}{15}w^Te\\&-\frac{4}{3}w^TAc+w^TA(cAc^2)-2w^TA(cAc)+w^TAc^2+2w^TA(c\hat{A}c)-2 w^TA(c\hat{c})\\&+\frac{1}{3}w^Tc-w^TA^2c^2+2w^TA^2c-2w^TA\hat{A}c+2w^TA\hat{c} -\frac{4}{3}w^T\hat{c}+w^T\hat{A}Ac^2\\&-2w^T\hat{A}Ac+4w^T\hat{A}c+2w^T\hat{A}^2c-2w^T\hat{A}\hat{c}+w^T \hat{A}c^3-3w^T\hat{A}c^2+\hat{v}^TcAc^2\\&+2\hat{v}^Tc\hat{A}c-\frac{4}{3}\hat{w}^Tc+\hat{w}^TcAc^2-2\hat{w}^TcAc +\hat{w}^Tc^2+2\hat{w}^Tc\hat{A}c-2\hat{w}^Tc\hat{c}\\&+\frac{1}{3}\hat{w}^Te-\hat{w}^TAc^2+2\hat{w}^TAc-2\hat{w}^T\hat{A}c+2 \hat{w}^T\hat{c}=\frac{1}{90}. \end{aligned}$$
figure aj

Order condition:

$$\begin{aligned} \frac{1}{72}\theta&+v^TcA^2c^2+2v^Tc(A\hat{A}c)+v^Tc\hat{A}c^2 +\frac{5}{12}w^Tc-\frac{1}{3}w^Tc^2+w^TcA^2c^2\\&-2w^TcA^2c+w^TcAc+2w^Tc(A\hat{A}c)-2w^TcA\hat{c}+w^Tc\hat{A}c^2-2w^Tc\hat{A}c\\&+w^Tc\hat{c}-\frac{1}{12}w^Te-w^TA^2c^2+2w^TA^2c-w^TAc-2w^T A\hat{A}c+2w^TA\hat{c}\\&-w^T\hat{A}c^2+2w^T\hat{A}c-w^T\hat{c}+\hat{v}^TA^2c^2 +2\hat{v}^TA\hat{A}c+\hat{v}^T\hat{A}c^2+\hat{v}^TcAc^2\\&+2\hat{v}^Tc\hat{A}c+\frac{5}{12}\hat{w}^Te-\frac{5}{3} \hat{w}^Tc+\hat{w}^TA^2c^2-2\hat{w}^TA^2c+3\hat{w}^TAc\\&+2\hat{w}^TA\hat{A}c-2\hat{w}^TA\hat{c}+\hat{w}^T\hat{A} c^2-4\hat{w}^T\hat{A}c+3\hat{w}^T\hat{c}+\hat{w}^TcAc^2\\&-2\hat{w}^TcAc+\hat{w}^Tc^2+2\hat{w}^Tc\hat{A}c-2\hat{w}^T c\hat{c}-\hat{w}^TAc^2=\frac{1}{72}. \end{aligned}$$
figure ak

Order condition:

$$\begin{aligned} \frac{1}{36}\theta&+v^T(Ac)(Ac^2)+2v^T(Ac)(\hat{A}c)+v^T\hat{c} Ac^2+2v^T\hat{c}\hat{A}c-\frac{1}{6}w^Te+\frac{1}{2}w^TAc^2\\&-\frac{4}{3}w^TAc+\frac{5}{6}w^Tc+w^T\hat{A}c-\frac{4}{3}w^T \hat{c}+w^T(Ac)(Ac^2)-2w^T(Ac)^2\\&+3w^TcAc+2w^T(Ac)(\hat{A}c)-4w^T\hat{c}Ac-w^TcAc^2-w^Tc^2-2 w^Tc\hat{A}c\\&+3w^Tc\hat{c}+w^T\hat{c}Ac^2+2w^T\hat{c}\hat{A}c-2w^T\hat{c}^2 +\hat{v}^TcAc^2+2\hat{v}^Tc\hat{A}c+\hat{v}^Tc^2Ac\\&+\hat{v}^Tc^2\hat{c}-\frac{10}{3}\hat{w}^Tc+\hat{w}^TcAc^2-4 \hat{w}^TcAc+\frac{7}{2}\hat{w}^Tc^2+2\hat{w}^Tc\hat{A}c-4\hat{w}^Tc\hat{c}\\&+\frac{5}{6}\hat{w}^Te-\hat{w}^TAc^2+3\hat{w}^TAc-2\hat{w}^T \hat{A}c+3\hat{w}^T\hat{c}+\hat{w}^Tc^2Ac-\hat{w}^Tc^3\\&+\hat{w}^Tc^2\hat{c}=\frac{1}{36}. \end{aligned}$$
figure al

Order condition:

$$\begin{aligned} \frac{1}{720}\theta&+v^TA^4c+v^TA^3\hat{c}+v^TA^2\hat{A}c+v^TA \hat{A}Ac+v^TA\hat{A}\hat{c}+v^T\hat{A}A^2c\\&+v^T\hat{A}A\hat{c}+v^T\hat{A}^2c-\frac{1}{120}w^Te+\frac{1}{24}w^T c-\frac{1}{6}w^TAc+\frac{1}{2}w^TA^2c\\&+w^TA^4c-w^TA^3c+w^TA^3\hat{c}+w^TA^2\hat{A}c-w^TA^2\hat{c} +\frac{1}{2}w^TA\hat{c}\\&+w^TA\hat{A}Ac-w^TA\hat{A}c+w^TA\hat{A}\hat{c}-\frac{1}{6}w^T \hat{c}+\frac{1}{2}w^T\hat{A}c+w^T\hat{A}A^2c\\&-w^T\hat{A}Ac+w^T\hat{A}A\hat{c}+w^T\hat{A}^2c-w^T\hat{A}\hat{c} +\hat{v}^TA^3c+\hat{v}^TA^2\hat{c}+\hat{v}^TA\hat{A}c\\&+\hat{v}^T\hat{A}Ac+\hat{v}^T\hat{A}\hat{c}+\frac{1}{24}\hat{w}^Te -\frac{1}{6}\hat{w}^Tc+\frac{1}{2}\hat{w}^TAc+\hat{w}^TA^3c-\hat{w}^TA^2c\\&+\hat{w}^TA^2\hat{c}+\hat{w}^TA\hat{A}c-\hat{w}^TA\hat{c} +\frac{1}{2}\hat{w}^T\hat{c}+\hat{w}^T\hat{A}Ac-\hat{w}^T\hat{A}c+\hat{w}^T \hat{A}\hat{c}=\frac{1}{720}. \end{aligned}$$

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Ökten Turaci , M., Öziş, T. On explicit two-derivative two-step Runge–Kutta methods. Comp. Appl. Math. 37, 6920–6954 (2018). https://doi.org/10.1007/s40314-018-0719-y

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s40314-018-0719-y

Keywords

Mathematics Subject Classification

Navigation