Many mathematical models describe quantities that vary continuously over time or space. A computer, however, generally performs calculations using a finite collection of numerical values. Discretization bridges this difference by replacing a continuous domain with discrete points, intervals, cells or elements. For example, a continuous time interval may be divided into equally spaced time steps, while a physical region may be divided into a grid. Derivatives can then be approximated using finite differences, and integrals can be approximated using numerical quadrature. Discretization introduces approximation error, so the choice of grid spacing or step size is important.
Discretization of Continuous Problems
Discretization converts a continuous mathematical problem into a finite or countable set of numerical quantities that can be processed computationally.
xᵢ = a + ih
Explanation
Variables
| Symbol | Meaning |
|---|---|
xᵢ | The ith discrete grid point. |
a | The starting point of the interval. |
i | A non-negative integer grid index. |
h | Grid spacing or step size. |
Example or application
A continuous interval from 0 to 1 can be divided into five equal subintervals using the points 0, 0.2, 0.4, 0.6, 0.8 and 1.
Definition
Discretization is the process of transforming a continuous mathematical domain, variable or equation into a discrete representation suitable for numerical computation.
Purpose
To convert continuous mathematical problems into discrete computational structures that can be handled by numerical algorithms.
Importance
Discretization is fundamental to numerical solutions of differential equations, numerical integration, computational physics, engineering simulation and many other scientific applications.
Learning Objectives
Define discretization.
Explain the difference between continuous and discrete representations.
Calculate grid points using a fixed step size.
Understand the role of mesh and grid spacing.
Recognize discretization error in numerical computation.
Key Concepts
Continuous domain
Discrete domain
Grid point
Step size
Mesh
Finite difference
Discretization error
Working Principle
A continuous interval or spatial region is divided into discrete computational units. The numerical method evaluates the mathematical model at selected points or over selected elements to approximate the behaviour of the original continuous system.
Process
Identify the continuous domain.
Choose the number of intervals or computational elements.
Determine the grid spacing.
Generate the discrete grid points.
Apply numerical approximations at those points.
Evaluate the effect of grid refinement on accuracy.
Applications
Finite-difference methods
Finite-element analysis
Computational fluid dynamics
Numerical solutions of differential equations
Digital image processing
Scientific simulation
Worked Example
Problem
Discretize the interval [0, 1] using a step size h = 0.25.
Steps
Use the grid formula xᵢ = a + ih.
Here a = 0 and h = 0.25.
For i = 0, x₀ = 0.
For i = 1, x₁ = 0.25.
For i = 2, x₂ = 0.50.
For i = 3, x₃ = 0.75.
For i = 4, x₄ = 1.00.
Answer
The grid points are 0, 0.25, 0.50, 0.75 and 1.00.
Common Misconceptions
Discretization does not mean randomly selecting a few values; it follows a defined mathematical representation.
A smaller step size often improves approximation accuracy but may increase computational cost.
Discretization error is different from rounding error.
A discrete approximation is not identical to the original continuous model.
Practice Questions
| Question | Answer |
|---|---|
What is discretization? | It is the conversion of a continuous mathematical domain or problem into a discrete computational representation. |
Find the grid points on [0, 2] with h = 0.5. | 0, 0.5, 1.0, 1.5 and 2.0. |
What is one disadvantage of using a very small grid spacing? | It can increase computational time and memory requirements. |
Key Takeaway
Discretization makes continuous mathematical problems computationally manageable by representing them through discrete points or elements.
Prerequisites
SCI-MAT-014-001
SCI-MAT-014-006
SCI-MAT-014-009
Functions and intervals
Reference
LeVeque, R. J. Finite Difference Methods for Ordinary and Partial Differential Equations.
Chapra, S. C., and Canale, R. P. Numerical Methods for Engineers.
Related Topics
SCI-MAT-014-009
SCI-MAT-014-011
SCI-MAT-014-013
SCI-MAT-014-276
Sources and verification
Sources are shown as provided in the project data. Not every source has yet been independently verified. Please report errors or missing citations.
- Source verification pending; check an authoritative textbook, university, standards body, or peer-reviewed source.