2
$\begingroup$

I have a list of data (acceleration) and I want to compute the displacement (so I have to integrate twice). A small part of the data is shown below:

{{0.38, 0.04}, {0.4, 0.105}, {0.42, 0.114}, {0.44, 0.065}, {0.46, 0.044}, 
{0.48, 0.023}, {0.5, 0.021}, {0.52, 0.019}, {0.54, -0.039}, {0.56, -0.074}, 
{0.58, -0.011}, {0.6, 0.058}, {0.62, 0.028}, {0.64, -0.102}, {0.66, -0.113}, 
{0.68, -0.047}, {0.7, 0.049}, {0.72, 0.01}, {0.74, -0.04}, {0.76, -0.025}, 
{0.78, 0.058}, {0.8, 0.072}, {0.82, -0.065}, {0.84, -0.107}, {0.86,0.032}, 
{0.88, 0.169}, {0.9, 0.099}, {0.92, -0.048}, {0.94, -0.065}, {0.96, 0.036}, 
{0.98, 0.045}}

Lets say the data is the discrete form of a function $f(t)$ and the displacements data that I want to obtain are the discrete equivalent of $F(t)$.

The data can't be fitted as you can see into a a typical function so I have to use Interpolation. I want to evaluate the integral at certain points, basically I want to find out the value of $F(t)$ over certain values $t$, the same $t$ that I have in the original list of data. Using Integrate will only allow me to compute the area under $f(t)$ over a range, Integrate with Interpolation will not give me the definite integral of the function. What should I use in this case and what can be done?

Thank you.

$\endgroup$
4
$\begingroup$

Assuming the initial velocity is zero:

accdata = {{0.38, 0.04}, {0.4, 0.105}, {0.42, 0.114}, {0.44, 0.065},
  {0.46, 0.044}, {0.48, 0.023}, {0.5, 0.021}, {0.52, 0.019}, {0.54, -0.039},
  {0.56, -0.074}, {0.58, -0.011}, {0.6, 0.058}, {0.62, 0.028}, {0.64, -0.102},
  {0.66, -0.113}, {0.68, -0.047}, {0.7, 0.049}, {0.72, 0.01}, {0.74, -0.04},
  {0.76, -0.025}, {0.78, 0.058}, {0.8, 0.072}, {0.82, -0.065}, {0.84, -0.107},
  {0.86, 0.032}, {0.88, 0.169}, {0.9, 0.099}, {0.92, -0.048}, {0.94, -0.065},
  {0.96, 0.036}, {0.98, 0.045}};

pos = Derivative[-2][Interpolation[accdata, InterpolationOrder -> 1]][t] // Head
pos["ValuesOnGrid"]
(*
  InterpolatingFunction[{{0.38, 0.98}}, <>]
  {0., 0.0000123333, 0.0000629333, 0.000155267, 0.000275467, 
   0.000413267, 0.000561533, 0.0007182, 0.000878733, 0.0010252, 
   0.0011486, 0.001268, 0.001404, 0.00154453, 0.0016522, 0.0017198, 
   0.0017706, 0.001832, 0.00189667, 0.00194967, 0.0019972, 0.00206333, 
   0.0021482, 0.0022134, 0.00224787, 0.002295, 0.00239593, 0.00253133, 
   0.0026562, 0.00276293, 0.00287793}
*)

There is also the problem of whether the time strictly runs between the end points of the interval 0.38 <= t <= 0.98 and how much weight the endpoints should get in the integration.

$\endgroup$
3
$\begingroup$
data = {{0.38, 0.04}, {0.4, 0.105}, {0.42, 0.114}, {0.44, 0.065}, 
    {0.46, 0.044}, {0.48, 0.023}, {0.5, 0.021}, {0.52, 
    0.019}, {0.54, -0.039}, {0.56, -0.074}, {0.58, -0.011}, {0.6, 
    0.058}, {0.62, 
    0.028}, {0.64, -0.102}, {0.66, -0.113}, {0.68, -0.047}, {0.7, 
    0.049}, {0.72, 0.01}, {0.74, -0.04}, {0.76, -0.025}, {0.78, 
    0.058}, {0.8, 0.072}, {0.82, -0.065}, {0.84, -0.107}, {0.86, 
    0.032}, {0.88, 0.169}, {0.9, 
    0.099}, {0.92, -0.048}, {0.94, -0.065}, {0.96, 0.036}, {0.98, 
    0.045}};

f = Interpolation[data];

times = data[[All, 1]];

{tmin, tmax} = MinMax[times];

fi = Interpolation[{#, Integrate[f[t], {t, tmin, #}]} & /@ times];

fii = Interpolation[{#, Integrate[fp[t], {t, tmin, #}]} & /@ times];

Plot[{fi[t], fii[t]}, {t, tmin, tmax},
 PlotLegends -> Placed[{"Velocity", "Displacement"}, {.65, .7}]]

enter image description here

$\endgroup$
2
$\begingroup$

How about:

Transpose[{data[[;; , 1]], Accumulate[data[[;; , 2]]]}]

Yielding:

{{0.38, 0.04}, {0.4, 0.145}, {0.42, 0.259}, {0.44, 0.324}, {0.46, 
0.368}, {0.48, 0.391}, {0.5, 0.412}, {0.52, 0.431}, {0.54, 
0.392}, {0.56, 0.318}, {0.58, 0.307}, {0.6, 0.365}, {0.62, 
0.393}, {0.64, 0.291}, {0.66, 0.178}, {0.68, 0.131}, {0.7, 
0.18}, {0.72, 0.19}, {0.74, 0.15}, {0.76, 0.125}, {0.78, 
0.183}, {0.8, 0.255}, {0.82, 0.19}, {0.84, 0.083}, {0.86, 
0.115}, {0.88, 0.284}, {0.9, 0.383}, {0.92, 0.335}, {0.94, 
0.27}, {0.96, 0.306}, {0.98, 0.351}}
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.