Message Boards Message Boards

0
|
6484 Views
|
4 Replies
|
1 Total Likes
View groups...
Share
Share this post:

[?] Solve a path for the flight of a ball?

Posted 7 years ago

Hello all,

I've run into a problem while I was trying to solve a path for the flight of a ball with lift force and drag. Sadly, even after searching the forums and reading the most common beginner mistakes post I still haven't found the issue. I fear it has to do with my defining of functions and use of DSolve, but I find the issue. Here is my code:

ClearGlobal[] := (ClearAll["Global`*"]; Clear[Derivative];);
ClearGlobal[]
(* First I start off by defining a few initial values. I hope the correct way to define these is with the ":=". *)
cl := 0.3
m := 0.050
rball := 0.02
g := 9.81
vi := 60
cd := 0.3
\[Rho] := 1.225
g := 9.81

(* Here are some numerical calculations that have to be done. Again, I think the ":=" Is needed. *)
area := Pi*rball^2
\[Alpha]d := (1/2)*\[Rho]*area*cd
\[Alpha]l := (1/2)*\[Rho]*area*cl

(* Here I define the position of the ball as 'r'. It's position as the time-derivative of r.
And finally a vector that is dependant on v's direction in order to orient the lifting force due to spindrift.
 *)
r[x_, y_] := {x, y}
v := D[r, t]
l := Normalize[Rotate[v, Pi/2]]

(* Here are the supplied equations for the problem. The drag and lift forces are proportional to a constant and the velocity, with a direction given by l and normalizing v. *)
fd[v_] :=-\[Alpha]d*Abs[v]^2*Normalize[v]
fl[v_] := \[Alpha]l*Abs[v]^2*l
fl[v_] := -m*g*{0, 1}

(* This is the total force on the ball in flight. *)
f[t_] := fd[v]+fd[v]+fd[v]

(* This is a fuction for the acceleration, which i want to solve using the acceleration given by taking the second derivative of r. *)
a[t_] := f[t]/m

(* Here I try to solve for x and y using these initial values. *)
DSolve[{D[r, {t, 2}] == a[t], x[0]==0, y[0]==0, v[0]==60}, x[t], {t,0,10}]

(* And now, as you can see it doesn't recognize the functions, but just outputs that it is true. I'm quite sure there is a problem with my defining of functions or the setup of my Dsolve.  *)

Any help for this newbie would be greatly appreciated, as I couldn't find it myself.

POSTED BY: Puck Planje
4 Replies
Posted 7 years ago

Thank you all for the help! I was finally able to solve the problem. For any future reader who is encountering the same problem I will leave my code below.

ClearGlobal[] := (ClearAll["Global`*"]; Clear[Derivatives];);
ClearGlobal[]

cl := 0.3
m := 0.050
rball := 0.02
cd := 0.3
\[Rho] := 1.225
g := 9.81
v0 := 60

area := Pi*rball^2
\[Alpha]d := (1/2)*\[Rho]*area*cd
\[Alpha]l := (1/2)*\[Rho]*area*cl
\[Theta] := Pi/3

v1 := v0*{Cos[\[Theta]], Sin[\[Theta]]} 
v[t_] := r'[t]

fd[t_] := -\[Alpha]d*Norm[v[t]]^2*Normalize[v[t]]
fl[t_] := \[Alpha]l*Norm[v[t]]^2*RotationMatrix[Pi/2].Normalize[v[t]]
fg[t_] := -m*g*{0, 1}
f[t_] := fd[t]+fl[t]+fg[t]

a[t_] := f[t]/m

sol = NDSolve[{r''[t] == a[t], r[0]=={0,0}, v[0]==v1}, r[t], {t,0,10}];

ParametricPlot[Evaluate[r[t] /. First[%], {t, 0, 10}]]
POSTED BY: Puck Planje

I am not quite sure what exactly you want to do. But notice that "Rotate" does not rotate your vector but produces a Graphics Primitive. So using this you won't get a meaningful equation. See the very end of the notebook attached.

And you define fl[ v_ ] twice. I am sure you meant to give a third function fg for the gravity - term.

I tried DSolve, but that, at least in my version (Mma 7) failed. So I used NDSolve for a numerical integration to obtain the path of the ball.

Attachments:
POSTED BY: Hans Dolhaine

@Puck Planje please make sure you know the rules: https://wolfr.am/READ-1ST

Next time please make your titles more informative and follow code formatting standards described in the rules above.

POSTED BY: Moderation Team
Posted 7 years ago

Let's start with

fd[v_] :=-?d*Abs[v]^2*Normalize[v]
fl[v_] := ?l*Abs[v]^2*l
fl[v_] := -m*g*{0, 1}
f[t_] := fd[v]+fd[v]+fd[v]

Do you perhaps mean something more like

fd[v_] :=-?d*Abs[v]^2*Normalize[v]
fl[v_] := ?l*Abs[v]^2*l
fg[v_] := -m*g*{0, 1}
f[t_] := fd[v]+fl[v]+fg[v]

But that definition of f[t] doesn't seem right either. Maybe you mean

f[v_]:=fd[v]+fl[v]+fg[v]

but I'm not sure that is right either.

Next let's look at a[t]

In[20]:= a[t_] := f[t]/m
         a[t]

Out[20]= {0., -9.81}

and that value is unchanged if I try a[0] or a[5] or a[1000].

Next let's look at what you are going to give to DSolve

In[21]:= {D[r, {t, 2}] == a[t], x[0] == 0, y[0] == 0, v[0] == 60}

Out[21]= {0 == {0., -9.81}, x[0] == 0, y[0] == 0, 0[0] == 60}

That doesn't look like a differential equation.

Can you use this to try to fix some problems before you try to give it to DSolve?

POSTED BY: Bill Simpson
Reply to this discussion
Community posts can be styled and formatted using the Markdown syntax.
Reply Preview
Attachments
Remove
or Discard

Group Abstract Group Abstract