Use the Trapezoidal Rule, the Midpoint Rule, and Simpson's Rule to approximate the given integral with the specified value of n. (Round your answers to six decimal places.) 4 e2√t sin(2t) dt, n = 8 0

Respuesta :

It looks like the integral could be

[tex]\displaystyle\int_0^4\underbrace{e^{2\sqrt t}\sin(2t)}_{f(t)}\,\mathrm dt[/tex]

to be approximated by the three listed rules with [tex]n=8[/tex].

Splitting up the interval of integration into 8 subintervals gives us the partition,

[0, 1/2], [1/2, 1], [1, 3/2], ..., [7/2, 4]

where the left and right endpoints, respectively, are given by the sequences

[tex]\ell_i=\dfrac{i-1}2[/tex]

[tex]r_i=\dfrac i2[/tex]

where [tex]1\le i\le8[/tex]. The midpoints of each subinterval are given by

[tex]m_i=\dfrac{\ell_i+r_i}2=\dfrac{\frac{i-1}2+\frac i2}2=\dfrac{2i-1}4[/tex]

Each subinterval has length

[tex]\Delta t_i=r_i-\ell_i=\dfrac i2-\dfrac{i-1}2=\dfrac12[/tex]

  • Trapezoidal rule:

[tex]\displaystyle\sum_{i=1}^8\frac{f(\ell_i)+f(r_i)}2\Delta t_i\approx11.070970[/tex]

  • Midpoint rule:

[tex]\displaystyle\sum_{i=1}^8f(m_i)\Delta t_i\approx10.767065[/tex]

  • Simpson's rule:

First we interpolate [tex]f(t)[/tex] over each subinterval with its own quadratic polynomial, given by

[tex]p_i(t)=f(\ell_i)\dfrac{(t-m_i)(t-r_i)}{(\ell_i-m_i)(\ell_i-r_i)}+f(m_i)\dfrac{(t-\ell_i)(t-r_i)}{(m_i-\ell_i)(m_i-r_i)}+f(r_i)\dfrac{(t-\ell_i)(t-m_i)}{(r_i-\ell_i)(r_i-m_i)}[/tex]

The integral is then approximately equal to

[tex]\displaystyle\sum_{i=1}^8\int_{\ell_i}^{r_i}p_i(t)\,\mathrm dt[/tex]

It turns out that

[tex]\displaystyle\int_{\ell_i}^{r_i}p_i(t)\,\mathrm dt=\frac{r_i-\ell_i}6(f(\ell_i)+4f(m_i)+f(r_i))[/tex]

so the integral we want to approximate is about

[tex]\displaystyle\sum_{i=1}^8\frac{r_i-\ell_i}6(f(\ell_i)+4f(m_i)+f(r_i))\approx10.868366[/tex]

Compare these results to the actual value of the integral, which is about 10.873071.