I assume you want to do some discrete time modeling. Known formulas require analytical input (see the Green function). If at some point in time you have a power table, typical analytical formulas will not help you.
The idea is this: for each point in time t0, the oscillator has some given acceleration, speed, etc. Now a force acts on it - according to the table that you gave - which will change its acceleration (F = m * a). For the next step t1, we assume that the acceleration remains at this constant, so we can apply the simple Newton equations (v = a * dt) with dt = (t1-t0) for this time interval. Iterate until the desired time range is simulated.
The most important parameter of this simulation is dt, that is, as a fine-grained calculation. For example, you can have 10 steps per second, but it depends entirely on your input parameters. What we are doing here is essentially the Eulerian integration of equations.
This, of course, is not all there is - such simulations can be quite complex, especially. in cases not very good when extreme accelerations, etc. In these cases, you need to perform numerical tests of operability inside the frame, because something "extreme" happens in one frame. Some numerical integration may also be required, for example. Runge-Kutta algorithm . I guess this leads far away at this stage, however.
EDIT: right after I posted this, someone posted a comment on the original question, pointing to “ Verlet Algorithm ”, which is basically an implementation of what I described above.
source share