Integration in Calculus

Last modified: December 22, 2024

This article is written in: 🇺🇸

Integration is another fundamental concept in calculus. It is used to calculate the area under a curve, the total change given a rate of change, or to solve differential equations.
The integral of a function can be interpreted as the area under the graph of the function.
Like differentiation, the exact mathematical function may not be known, or it may be too complex to integrate analytically.

Numerical Integration (or Quadrature)

Numerical integration, often referred to as quadrature, is a method used to approximate the integral of a function using finite sums.
It is often used when it may be impossible to integrate some functions analytically, or when the formula for the function is unknown, for instance, when data from an experiment is available.
The function is evaluated at a set of points and these values are used to estimate the integral's value.
The numerical approaches for integration divide the large interval into smaller intervals and add the results from each smaller piece, according to the equation: $$ \int_{a}^{b} f(x) dx = \int_{a}^{c} f(x) dx + \int_{c}^{b} f(x)dx $$ where $c$ is a point between $a$ and $b$.

Rectangle Rule (or Midpoint Rule): This method approximates the area under the curve as a series of rectangles. For example, if we have a function $f(x) = x^2$, and we want to approximate the integral over the interval $[1, 2]$ with a step size $h = 0.1$, using the rectangle rule, we get:

$$\int_{1}^{2} x^2 dx \approx \sum_{i=1}^{10} h \cdot f(1 + 0.1i)$$

Trapezoidal Rule: This method approximates the area under the curve as a series of trapezoids, which provides a better approximation compared to the rectangle rule. An example application would be the same function $f(x) = x^2$, the integral can be approximated as:

$$\int_{1}^{2} x^2 dx \approx \sum_{i=1}^{10} \frac{h}{2} (f(1 + 0.1(i-1)) + f(1 + 0.1i))$$

Simpson's Rule: This method approximates the area under the curve as a series of parabolic segments. It is typically more accurate than both the rectangle and trapezoidal rules. Let's consider an example where we want to approximate the same integral, using Simpson's rule, it would look like:

$$\int_{1}^{2} x^2 dx \approx \frac{h}{3} (f(1) + 4\sum_{i=1}^{5} f(1 + 0.2(i-1)) + 2\sum_{i=1}^{4} f(1 + 0.2i) + f(2))$$

Monte Carlo Integration: This method uses random sampling to estimate the integral, which can be especially useful for higher-dimensional integrals. An example of this could be estimating the integral of a complex multi-variable function over a specific region.

Each of the methods for numerical integration has its own rate of convergence, which determines how quickly the numerical approximation approaches the exact value. For instance, the trapezoidal rule converges as $O(h^2)$, Simpson's rule converges as $O(h^4)$, etc., where $h$ is the step size.

Numerical integration can handle a wide range of functions, including ones where analytical integration is difficult or impossible.
These methods can also handle functions that are only known at certain discrete points, such as data obtained from experiments or observations.

Numerical integration is only an approximation and can introduce errors. The approximation is more accurate for a larger number of points, but it may require more computational resources.
The accuracy of the approximation also depends on the behavior of the function. If the function is rapidly changing, discontinuous, or noisy, the approximation may not be very accurate.
Numerical methods require the function to be known or evaluated at certain points. If the function is expensive to evaluate, or if it is only known at a few points, numerical integration may not be practical.
Different methods have different strengths and weaknesses. While Simpson's Rule is generally more accurate than the Trapezoid Rule for smooth functions, it may perform worse for functions with discontinuities or sharp turns.