NAG CL Interface
D02 (Ode)
Ordinary Differential EquationsIntegrators for Stiff Ordinary Differential Systems

Settings help

CL Name Style:

1 Scope of the Chapter

This chapter is concerned with the numerical solution of ordinary differential equations. There are two main types of problem: those in which all boundary conditions are specified at one point (initial value problems), and those in which the boundary conditions are distributed between two or more points (boundary value problems and eigenvalue problems). Functions are available for initial value problems and two-point boundary value problems.

2 Background to the Problems

For most of the functions in this chapter a system of ordinary differential equations must be written in the form
y1=f1(x,y1,y2,,yn), y2=f2(x,y1,y2,,yn), yn=fn(x,y1,y2,,yn),  
that is the system must be given in first-order form. The n dependent variables (also, the solution) y1,y2,,yn are functions of the independent variable x, and the differential equations give expressions for the first derivatives yi=dyi dx in terms of x and y1,y2,,yn. For a system of n first-order equations, n associated boundary conditions are usually required to define the solution.
A more general system may contain derivatives of higher order, but such systems can almost always be reduced to the first-order form by introducing new variables. For example, suppose we have the third-order equation
We write y1=z, y2=z, y3=z, and the third-order equation may then be written as the system of first-order equations
y1=y2 y2=y3 y3=-y1y3-k(l-y22).  
For this system n=3 and we require 3 boundary conditions in order to define the solution. These conditions must specify values of the dependent variables at certain points. For example, we have an initial value problem if the conditions are
y1=0.1  at ​x=0 y2=0.1  at ​x=0 y3=0.1  at ​x=0.  
These conditions would enable us to integrate the equations numerically from the point x=0 to some specified end point. We have a boundary value problem if the conditions are
y1=0  at ​x=0 y2=0  at ​x=0 y2=1  at ​x=10.  
These conditions would be sufficient to define a solution in the range 0x10, but the problem could not be solved by direct integration (see Section 2.2). More general boundary conditions are permitted in the boundary value case.

2.1 Initial Value Problems

To solve first-order systems, initial values of the dependent variables yi, for i=1,2,,n, must be supplied at a given point, a. Also a point, b, at which the values of the dependent variables are required, must be specified. The numerical solution is then obtained by a step-by-step calculation which approximates values of the variables yi, for i=1,2,,n, at finite intervals over the required range [a,b]. The functions in this chapter adjust the step length automatically to meet specified accuracy tolerances. Although the accuracy tests used are reliable over each step individually, in general an accuracy requirement cannot be guaranteed over a long range. For many problems there may be no serious accumulation of error, but for unstable systems small perturbations of the solution will often lead to rapid divergence of the calculated values from the true values. A simple check for stability is to carry out trial calculations with different tolerances; if the results differ appreciably the system is probably unstable. Over a short range, the difficulty may possibly be overcome by taking sufficiently small tolerances, but over a long range it may be better to try to reformulate the problem.
A special class of initial value problems are those for which the solutions contain rapidly decaying transient terms. Such problems are called stiff; an alternative way of describing them is to say that certain eigenvalues of the Jacobian matrix ( fi yj ) have large negative real parts when compared to others. These problems require special methods for efficient numerical solution; the methods designed for non-stiff problems when applied to stiff problems tend to be very slow, because they need small step lengths to avoid numerical instability. A full discussion is given in Hall and Watt (1976) and a discussion of the methods for stiff problems is given in Berzins et al. (1988).

2.2 Boundary Value Problems

In general, a system of nonlinear differential equations with boundary conditions at two or more points cannot be guaranteed to have a solution. The solution, if it exists, has to be determined iteratively. A comprehensive treatment of the numerical solution of boundary value problems can be found in Ascher et al. (1988) and Keller (1992). The methods for this chapter are discussed in Ascher et al. (1979), Ascher and Bader (1987) and Gladwell (1987).

2.2.1 Collocation methods

In the collocation method, the solution components are approximated by piecewise polynomials on a mesh. The coefficients of the polynomials form the unknowns to be computed. The approximation to the solution must satisfy the boundary conditions and the differential equations at collocation points in each mesh sub-interval. A modified Newton method is used to solve the nonlinear equations. The mesh is refined by trying to equidistribute the estimated error over the whole interval. An initial estimate of the solution across the mesh is required.

2.2.2 Finite difference methods

If a boundary value problem seems insoluble by the above method and a good estimate for the solution of the problem is known at all points of the range then a finite difference method may be used. Finite difference equations are set up on a mesh of points and estimated values for the solution at the grid points are chosen. Using these estimated values as starting values a Newton iteration is used to solve the finite difference equations. The accuracy of the solution is then improved by deferred corrections or the addition of points to the mesh or a combination of both. Good initial estimates of the solution may be required in some cases and the method is unlikely to be successful when the solution varies very rapidly over short ranges. A discussion is given in Chapters 9 and 11 of Gladwell and Sayers (1980) and Chapter 4 of Gladwell (1979a).

3 Recommendations on Choice and Use of Available Functions

There are no functions which deal directly with complex equations. These may however be transformed to larger systems of real equations of the required form. Split each equation into its real and imaginary parts and solve for the real and imaginary parts of each component of the solution. Whilst this process doubles the size of the system and may not always be appropriate it does make available for use the full range of functions provided presently.

3.1 Initial Value Problems

In general, for non-stiff first-order systems, Runge–Kutta (RK) functions should be used. For the usual requirement of integrating across a range the appropriate functions are d02pec and d02pqc; d02pqc is a setup function for d02pec. For more complex tasks there are forward and reverse communication variants (Section 7 in How to Use the NAG Library) of single step functions with corresponding interpolator; for direct communication these are d02pfc and d02psc, while for reverse communication these are d02pgc, d02phc and d02pjc. There are also related utility functions d02prc, d02ptc and d02puc. When a system is to be integrated over a long range or with relatively high accuracy requirements the variable-order, variable-step Adams' codes may be more efficient. The appropriate function in this case is d02cjc. For more complex tasks using an Adams' code there are a further four related functions: d02qfc, d02qwc, d02qyc and d02qzc.
For stiff systems, that is those which usually contain rapidly decaying transient components, the Backward Differentiation Formula (BDF) variable-order, variable-step codes should be used. The appropriate functions in this case are: the simple driver d02ejc, or the more comprehensive d02nec, based on the DASSL implementation (see Brenan et al. (1996)), and its related functions d02mcc, d02mwc and d02npc.
If you are not sure how to classify a problem, you are advised to perform some preliminary calculations with d02pec, which can indicate whether the system is stiff. We also advise performing some trial calculations with d02pec (RK), d02cjc (Adams) and d02ejc (BDF) so as to determine which type of function is best applied to the problem. The conclusions should be based on the computer time used and the number of evaluations of the derivative function fi. See Gladwell (1979b) for more details.

3.1.1 Runge–Kutta functions

The basic RK functions are d02pfc (direct communication) and d02pgc (reverse communication) which take one integration step at a time. An alternative to d02pfc is d02pec, which provides output at user-specified points. The initialization of d02pec, d02pfc or d02pgc and the setting of optional inputs, including choice of method, is made by a call to the setup function d02pqc. Optional output information about error assessment, can be obtained by calls to the function d02puc while integration statistics are returned by the diagnostic function d02ptc. d02psc may be used to interpolate on information produced by d02pfc to give solution and derivative values between the integration points. Similarly d02phc may be used to setup an interpolator on information produced by d02pgc, and d02pjc can evaluate that interpolator to give solution and derivative values between integration points; these functions are recommended when a high-order RK method is specified in the setup function. d02prc may be used to reset the end of the integration range whilst integrating using d02pfc or d02pgc.

3.1.2 Adams' functions

The general Adams' variable-order variable-step function is d02qfc, which provides a choice of automatic error control and the option of a sophisticated root-finding technique. The initialization of d02qfc and the setting of optional inputs is made by a call to the setup function d02qwc. d02qzc may be used to interpolate on information produced by d02qfc to give solution and derivative values between the integration points.
There is a simple driving function d02cjc, which integrates a system over a range and, optionally, computes intermediate output and/or determines the position where a specified function of the solution is zero.

3.1.3 BDF functions

General functions for implicit ordinary differential equations with options for forms of numerical linear algebra are provided by a suite of functions using the DASSL implementation (see Brenan et al. (1996)) of the BDF. The main solver in this suite is d02nec is designed for solving systems of the form,
F (t,y,y) =0 .  
These formulations permits solution of differential/algebraic systems (DAEs). Additionally d02nec can be used to solve difficult algebraic problems by continuation; for example, the nonlinear algebraic problem
can be solved by integrating solutions of
f(x) + (1-t) g(x) =0  
where the solution to
f(x) + g(x) =0  
is known.
Options for the solver must be supplied as arguments in an initial call to the setup function d02mwc; the Jacobian of the system to be solved can be considered to have a banded structure by a call to d02npc prior to calling the solver function; and integration can be continued by a call to d02mcc between calls to the solver d02nec.
There is a simple driving function d02ejc, which integrates a system over a range and, optionally, computes intermediate output and/or determines the position where a specified function of the solution is zero. To solve the equations arising in the BDF method an approximation to the Jacobian ( fi yj ) is required. This approximation can be calculated internally but you may supply an analytic expression. In most cases supplying a correct analytic expression will reduce the amount of computer time used.

3.2 Boundary Value Problems

In general, for a nonlinear system of mixed order with separated boundary conditions, the collocation method (d02tlc and its associated functions) can be used. Problems of a more general nature can often be transformed into a suitable form for treatment by d02tlc, for example nonseparated boundary conditions or problems with unknown parameters (see Section 9 in d02tvc for details).
For simple boundary value problems with assigned boundary values you may prefer to use a code based on the finite difference method for which there is a function with simple calling sequence (d02gac).
For difficult boundary value problems, where you need to exercise some control over the calculation, and where the collocation method proves unsuccessful, you may wish to try the alternative method of finite differences (d02rac).
Note that it is not possible to make a fully automatic boundary value function, and you should be prepared to experiment with different starting values or a different function if the problem is at all difficult.

3.2.1 Shooting methods

Functions employing shooting methods are currently only available from the FL Interface, see Section 3.2.2 for details.

3.2.2 Finite difference methods

d02gac may be used for simple boundary value problems with assigned boundary values.
You may find that convergence is difficult to achieve using d02gac since only specifying the unknown boundary values and the position of the finite difference mesh is permitted. In such cases you may use d02rac, which permits specification of an initial estimate for the solution at all mesh points and allows the calculation to be influenced in other ways too. d02rac is designed to solve a general nonlinear two-point boundary value problem with nonlinear boundary conditions.
A function, d02gbc, is also supplied specifically for the general linear two-point boundary value problem written in a standard ‘textbook’ form.
You are advised to use interpolation functions from Chapter E01 to obtain solution values at points not on the final mesh.

3.2.3 Chebyshev integration method

The Chebyshev integration method is an implementation of the Chebyshev collocation method which is fully described and compared against other implementations in Muite (2010). d02uec solves a linear constant coefficient boundary value problem using the Chebyshev integration formulation on a Chebyshev Gauss–Lobatto grid and solving in the coefficient space. The required Chebyshev Gauss–Lobatto grid points on a given arbitrary interval [a,b] can first be generated using d02ucc. Then d02uac obtains the Chebyshev coefficients for the right-hand side (of system) function discretized on the obtained Chebyshev Gauss–Lobatto grid. d02uec then solves the problem in Chebyshev coefficient space using the integration formulation. Finally d02ubc evaluates the solution (or one of its lower order derivatives) from the set of Chebyshev coefficients returned by d02uec on the Chebyshev Gauss–Lobatto grid on [a,b]. The set of functions can be used to solve up to fourth order boundary value problems.

3.3 Summary of Recommended Functions

Problem Function
RK Method Adams' Method BDF Method
Initial Value Problems
Driver Functions
Integration over a range with optional intermediate output and optional determination of position where a function of the solution becomes zero d02cjc d02ejc
Integration of a range with intermediate output d02cjc d02ejc
Integration of a range until function of solution becomes zero d02cjc d02ejc
Comprehensive Integration Functions d02pec, d02pfc, d02pgc, d02phc, d02pjc, d02pqc, d02prc, d02psc, d02ptc and d02puc d02qfc, d02qwc and d02qzc d02mcc, d02mwc, d02nec and d02npc
Package for Solving Second-order Systems of Special Form
Boundary Value Problems
Collocation Method, Mixed Order
d02tlc, d02tvc, d02txc, d02tyc and d02tzc
Boundary Value Problems
Finite Difference Method
linear problem d02gbc
full nonlinear problem d02rac
Boundary Value Problems
Chebyshev Collocation, Integration Formulation
single linear equation d02uec with d02uac, d02ubc, d02ucc

4 Decision Trees

Tree 1: Initial Value Problems

If the problem is not first order convert to first order.
Is the problem known to be stiff?
  Is the system of the form y=f(x,y)?   d02ejc
  no   no
d02mcc, d02mwc, d02nec and d02npc
Backward Difference Formula: d02ejc.
Adams' method with driver function: d02cjc
Adams' method with comprehensive suite: d02qfc, d02qwc, d02qyc and d02qzc
BDF with comprehensive suite: d02mcc, d02mwc, d02nec and d02npc
Runge–Kutta method: d02pec, d02pfc, d02pgc, d02phc, d02pjc, d02pqc, d02prc, d02psc, d02ptc and d02puc

Tree 2: Boundary Value Problems

Is the problem linear, expressible as
y = A(x)y = g(x) ?
  Are boundary conditions linearly dependent:
Cy(a) + Dy(b) = E ?
  no   no
d02jaf, d02jbf, d02tgf and d02uec
Are boundary values or parameter values to be determined?   Are only boundary values to be determined?   d02gac
  no   no
d02hbf and d02saf
Are boundary conditions simple, e.g., y(a) = ya ?   d02haf
d02rac and d02tlc

5 Functionality Index

Differentiation of a function discretized on Chebyshev Gauss–Lobatto points   d02udc
Linear constant coefficient boundary value problem,  
Chebyshev spectral integration method,  
Chebyshev coefficients generator for a function discretized on Chebyshev Gauss–Lobatto grid   d02uac
Chebyshev coefficients to function values on Chebyshev Gauss–Lobatto grid   d02ubc
Chebyshev Gauss–Lobatto grid generator   d02ucc
Chebyshev integration solver for linear constant coefficient boundary value problem   d02uec
Clenshaw–Curtis quadrature weights generator at Chebyshev Gauss–Lobatto points   d02uyc
Evaluation on uniform grid of function by Barycentric Lagrange interpolation   d02uwc
value of kth Chebyshev polynomial   d02uzc
System of first-order ordinary differential equations, initial value problems,  
comprehensive integrator functions for stiff systems,  
continuation to call d02nec   d02mcc
implicit ordinary differential equations coupled with algebraic equations,  
banded Jacobian selector for DASSL integrator   d02npc
DASSL integrator   d02nec
integrator setup for DASSL   d02mwc
comprehensive integrator functions using Adams' method with root-finding option,  
diagnostic function for root-finding   d02qyc
direct communication   d02qfc
interpolant   d02qzc
setup function   d02qwc
comprehensive integrator functions using Runge–Kutta methods,  
diagnostic function   d02ptc
diagnostic function for global error assessment   d02puc
interpolant, reverse communication   d02phc
interpolant and interpolation, direct communication   d02psc
interpolation, reverse communication   d02pjc
over a range with intermediate output   d02pec
over a step, direct communication   d02pfc
over a step, reverse communication   d02pgc
reset end of range   d02prc
setup function   d02pqc
simple driver functions,  
variable-order variable-step Adams' method,  
until (optionally) a function of the solution is zero, with optional intermediate output   d02cjc
variable-order variable-step backward differentiation formulae method for stiff systems,  
until (optionally) a function of the solution is zero, with optional intermediate output   d02ejc
System of ordinary differential equations, boundary value problems,  
comprehensive functions using a collocation technique,  
continuation function   d02txc
diagnostic function   d02tzc
general nonlinear problem solver   d02tlc
interpolation function   d02tyc
setup function   d02tvc
finite difference technique with deferred correction,  
general linear problem   d02gbc
general nonlinear problem, with continuation facility   d02rac
simple nonlinear problem   d02gac

6 Auxiliary Functions Associated with Library Function Arguments


7 Withdrawn or Deprecated Functions

The following lists all those functions that have been withdrawn since Mark 24 of the Library or are in the Library, but deprecated.
Function Status Replacement Function(s)
d02pcc Withdrawn at Mark 26 d02pec and associated d02p functions
d02pdc Withdrawn at Mark 26 d02pfc and associated d02p functions
d02ppc Withdrawn at Mark 26 No replacement required
d02pvc Withdrawn at Mark 26 d02pqc
d02pwc Withdrawn at Mark 26 d02prc
d02pxc Withdrawn at Mark 26 d02psc
d02pzc Withdrawn at Mark 26 d02puc

8 References

Ascher U M and Bader G (1987) A new basis implementation for a mixed order boundary value ODE solver SIAM J. Sci. Stat. Comput. 8 483–500
Ascher U M, Christiansen J and Russell R D (1979) A collocation solver for mixed order systems of boundary value problems Math. Comput. 33 659–679
Ascher U M, Mattheij R M M and Russell R D (1988) Numerical Solution of Boundary Value Problems for Ordinary Differential Equations Prentice–Hall
Berzins M, Brankin R W and Gladwell I (1988) Design of the stiff integrators in the NAG Library SIGNUM Newsl. 23 16–23
Brenan K, Campbell S and Petzold L (1996) Numerical Solution of Initial-Value Problems in Differential-Algebraic Equations SIAM, Philadelphia
Gladwell I (1979a) The development of the boundary value codes in the ordinary differential equations chapter of the NAG Library Codes for Boundary Value Problems in Ordinary Differential Equations. Lecture Notes in Computer Science (eds B Childs, M Scott, J W Daniel, E Denman and P Nelson) 76 Springer–Verlag
Gladwell I (1979b) Initial value routines in the NAG Library ACM Trans. Math. Software 5 386–400
Gladwell I (1987) The NAG Library boundary value codes Numerical Analysis Report 134 Manchester University
Gladwell I and Sayers D K (ed.) (1980) Computational Techniques for Ordinary Differential Equations Academic Press
Hall G and Watt J M (ed.) (1976) Modern Numerical Methods for Ordinary Differential Equations Clarendon Press, Oxford
Keller H B (1992) Numerical Methods for Two-point Boundary-value Problems Dover, New York
Muite B K (2010) A numerical comparison of Chebyshev methods for solving fourth-order semilinear initial boundary value problems Journal of Computational and Applied Mathematics 234(2) 317–342
Pryce J D (1986) Error estimation for phase-function shooting methods for Sturm–Liouville problems IMA J. Numer. Anal. 6 103–123