In earlier four posts, we had introduced two major methods Finite Element Method

and Finite Difference Method. In this post we will discuss about various special
numerical procedures other than finite element methods applicable to particular
geotechnical situation. These methods are Method of Characteristics, Boundary
Integral Equation Method and Fast Fourier Transform.

A question arises in the mind of geotechnical engineer that which finite element
procedure should be adopted. We are already aware that the generality of
numerical procedure depends on its numerical characteristics and can be studied by
examining such properties such as convergence, stability, and consistency. These
properties have been rigorously studied for finite difference method. However,
numerical properties of the finite element method have not been established
adequately.

The engineer has adopted quantitative approach to assess these numerical

procedures and tried to use a given scheme to many problems and if it is found
satisfactory, it is considered acceptable. This pragmatic and practical approach is
often necessary, but it may not yield a general scheme. Use of stability and
convergence criteria is highly useful in selecting temporal and spatial mesh layouts
to achieve optimum economy and accuracy.

A number of recent studies have used von Neumann, Lax-Richtmyer, perturbation,

and other techniques for examining numerical accuracy and stability and for
deriving stability criteria for finite element procedures for parabolic equations like
consolidation and dynamic applications.

METHOD OF CHARACTERISTICS
In this method of characteristics, a set of partial differential equations is converted
into a set of ordinary differential equations and the latter is solved by using the
finite difference method.
In layman language, a characteristic can be explained as a path of transmission or
propagation followed by a quantity or a disturbance such as a shock wave. In a
sense, a characteristic is a line along which energy is propagated. The path followed
by a vehicle along road, the lines (or planes) along which an applied load is
transmitted in a geological mass, and lines along which a disturbance is caused by
an impact are some examples of characteristics.

Let us understand this concept by considering a simple example of propagation of a

wave and of displacement u with time t along a one dimensional medium (rod). The
paths of propagation of u in the xt plane are the characteristics. The propagation

can take place only in two directions left to right and right to left. Solution of the
problem involves finding values of u over the xt plane i.e. if we can find the
characteristics, we have the solution. The velocity of propagation v = du/dt also
propagates along the same characteristics as the displacement u.
A number of factors can influence the propagation of displacement. Main among
them are: initial conditions, the laws governing the problem often expressed by set
of differential equations, physical properties of the medium.
Another example that can be taken is propagation of longitudinal waves along a
one-dimensional soil medium. The problem can be assumed to be governed by two
differential equations called as wave equations and can be derived from the
equations of motion (conservation of momentum) and the equation of continuity
(conservation of mass). They show that both u and v velocity propagate along the
same path. Here E is the modulus of elasticity and is mass density of the medium.

The characteristics for linear differential equations form two sets of parallel lines.
The region with such parallel and straight characteristics is called as the region of
constant state. The characteristics for nonlinear equations are usually curved lines.

A general solution to the equation can be written as:

The method of characteristics follows the physical process of propagation. Since we
need to solve only ordinary differential equations, the necessity for solving large
system of equations can be avoided. Thus the method is found to be economical.
Though at present, it may be difficult to handle complexities like non-homogeneities
and nonlinearities; however, the method shows good promise for future research
and applications. Plasticity solutions have been provided using this approach.

BOUNDARY INTEGRAL EQUATION METHOD

The boundary-integral equation (BIE) method involves numerical solution of a set of
integral equations that connect the boundary, or surface, tractions to boundary
displacements. Unlike the finite element and finite difference methods, the BIE
method is based on solution of integral rather than differential equations and
consists in discretization of only the boundary or the surface of body into a number
of segments, or elements. The numerical solution is first obtained at the boundary,
and then the solutions at different points within the body is obtained from the
solutions at the boundary. As a consequence, the number of physical dimensions is
reduced by 1, resulting in saving of time and effort.

The integral equation may be regarded as an exact solution of the governing partial
differential equation. The boundary element method attempts to use the given

boundary conditions to fit boundary values into the integral equation, rather than
values throughout the space defined by a partial differential equation. Once this is
done, in the post-processing stage, the integral equation can then be used again to
calculate numerically the solution directly at any desired point in the interior of the
solution domain.
BIE is applicable to problems for which Green's functions can be calculated. These
usually involve fields in linear homogeneous media. This places considerable
restrictions on the range and generality of problems to which boundary elements
can usefully be applied. Nonlinearities can be included in the formulation, although
they will generally introduce volume integrals which then require the volume to be
discretized before solution can be attempted, removing one of the most often cited
advantages of BIE. A useful technique for treating the volume integral without
discretizing the volume is the dual-reciprocity method. The technique approximates
part of the integrand using radial basis functions (local interpolating functions) and
converts the volume integral into boundary integral after collocating at selected
points distributed throughout the volume domain (including the boundary). In the
dual-reciprocity BIE, although there is no need to discretize the volume into meshes,
unknowns at chosen points inside the solution domain are involved in the linear
algebraic equations approximating the problem being considered.
The BIE method has certain advantages over the finite element method and has
been successfully used in boundary based problems in geotechnical engineering.
Though it has been used in linear problems but holds good potential for further
research and applications.

THE FAST FOURIER TRANSFORM

The fast Fourier Transform (FFT) method is one of the most powerful developments
in numerical analysis in recent years. It has greatly affected the way computers are
used in dynamic analysis, signal processing and frequency domain methods applied
in geotechnical systems.
The importance of FFT lies in the fact that it reduces the number of operations by a
clever sequence of calculations.

A fast Fourier transform (FFT) algorithm computes the discrete Fourier transform
(DFT) of a sequence, or its inverse. Fourier analysis converts a signal from its
original domain (often time or space) to a representation in the frequency domain
and vice versa. A FFT rapidly computes such transformations by factorizing the DFT
matrix into a product of sparse (mostly zero) factors.[1] As a result, it manages to
reduce the complexity of computing the DFT from

, which arises if one simply

applies the definition of DFT, to , where is the data size.

In next post, we will discuss special topics relevant in geotechnical engineering.