Numerical integration

Numerical integration. In mathematics the study of integrals is fundamental since it solves many problems of daily life, this branch is studied in Infinitesimal Calculus as a definite integral function f (x) in an interval [a, b] as the result of a process the limit of a finite sum.

 

Summary

[ hide ]

  • 1 Applications in the different branches
  • 2 Trapezius method
  • 3 Simpson’s method
  • 4 Source

Applications in the different branches

The reader surely remembers the important applications that integrals have in the different branches of Geometry , Physics , Chemistry , Economic Sciences and, practically, in all branches of knowledge. Assuming that f (x) is continuous on [a, b], the integral can be calculated using the Barrow rule (also called Newton-Leibniz), where F (x) is any primitive of f (x), that is , a function whose derivative of a function is f (x).

The weak point of this analytical procedure to evaluate an integral is obtaining a primitive function. For many simple functions primitives are obtained with more or less difficulty, but in many cases integrals are presented for which there are no primitives that can be expressed in terms of elementary functions.

The worst thing is that, on many occasions, they are simple integrands (that is, formed by elementary functions). Of course, if you do not have a primitive expressed in terms of elementary functions, it is not possible to evaluate it at the limits of integration, and the Barrow rule becomes inapplicable.

Trapezoidal method

Although the most elementary method to calculate the definite integral of a function is to use the same definition of the definite integral but disregarding the limit process of a function, that is:

the result obtained is very inaccurate, unless a very large number n of subintervals is taken. The trapezium method below requires a number of calculations similar to expression (1) and produces much better results.

The trapezoidal method is based on the idea of ​​dividing the integration interval into n subintervals of amplitude h by means of a set of points {a = x0, x1, x2, …, xn = b} and decomposing the integral into n integrals, each of which has a small integration interval of length h, as illustrated in the figure. The parameter h is called the step and plays an important role in the accuracy of the result obtained.

Simpson’s method

Simpson’s method is an improvement to that of the trapezoids since the approximation is made with Polnomial interpolators of degree two. Consider the integral:

where f (x) is assumed to be continuous on [a, b]. Let the integration interval be divided into an even number of intervals of equal amplitude h by the points {a = x0, x1, x2, …, xn = b}. Figures 1 and 2 show the geometric idea of the Simpson method : Figure 1 shows that the region between the horizontal axis and the graph of f (x) has been divided into vertical stripes using lines that determine the points of the partition of [a, b]. In figure 2 the function f (x) has been replaced by a set of polynomials of degree 2, that is, parabolas with a vertical axis.

As in the trapezoidal method, the notation will be simplified by calling yi = f (xi) for i = 0, 1, 2, …, n.

 

Leave a Comment

Scroll to Top