That the Bernoulli numbers has z/(exp(z)-1)) as an exponentional generating functions is of course very well known. I am wondering why the gfun package doesn't seem to be able to work this out? Naive approaches like

gfun[guessgf]([seq(bernoulli(n),n=0..100)],z);

give FAIL. I would expect gfun to be able to transform the given sequence corresponding to replacing A(z) by 1/A(z), easily recognize (exp(z)-1))/z, and transform back. Perhaps the package is in need of maintenance?

Any help would be much appreciated.

Best,

Soren

Hi

Is there any idea using Maple  to compute the leg lengths  in Stewart-Gough platform ( see the following figure)

I would like to make a code using cross-product and the unit Normal N 

·       P is of length 13 and displaced in the Y direction by 10 degrees from the vertical (Z axis)

·       N is displaced in the X direction by 18 degrees from the vertical (Z axis)

·       L_B is position [7 5] from the bottom plate centre in the XY plane

·       L_T is in position [3.5 4.2] from the top-plate centre in the AB plane

Many thanks for any help

L := sum( 1/ln(k), k=2..n ) * ln(n)/n;
        
limit(L, n=infinity);
                               0
# Should be 1

Just curious: in Maple 2017, is it OK?

Hi, i don't know what happened but every time that I run the simulation appears this: 

Dear sir,

in the program boundary conditions D(f)(0)=0 doesn't showing result but when use d(f)(0)=1 it will execute, why is this can you explain this ?program.mw
 

So, I'm trying to delelop an algorithm for the method of multiple scales. Starting with a simple ODE:

After scaling, it should be written in the form:

A proto-algorithm would be:

restart;
ode := diff(x(t), `$`(t, 2))+x(t) = 0;
i_ini := 1; i_fin := 3; j_ini := 0; j_fin := 2;
PDEtools:-dchange({t = sum(epsilon^j*T[j], j = j_ini .. j_fin), x(t) = sum(epsilon^i*X[i](T[1]), i = i_ini .. i_fin)}, ode, [{T[0], T[1], T[2]}, {X[1], X[2], X[3]}])

It is not working, though. Could anyone help me out?

Thanks in advance.

hello everyone. i wanna ask. how to do a coding of stability region on diagonally implicit Runge-Kutta for order 3 and order 4? Thank you in advanced :) 

Hello,

I am trying to solve analytically a simple system of partial differential equations with boundary conditions and I am not able to do it. Even in the very simple case of

pdsolve([diff(u(x, y, t), y, y) = 0, diff(p(x, y, t), y) = 0, u(x, 0, t) = 1, (D[2](u))(x, 1, t) = 0, p(x, 1, t) = 2], [p(x, y, t), u(x, y, t)]);

I don't get any answer.  However if I remove the boundary conditions I get the right answer

pdsolve([diff(u(x, y, t), y, y) = 0, diff(p(x, y, t), y) = 0], [{p(x, y, t), u(x, y, t)}]);
 {p(x, y, t) = _F3(x, t), u(x, y, t) = _F1(x, t) y + _F2(x, t)}

Can maple 2015 solve analytically systems of partial differential equations with boundary conditions? I have not been able to find any example anywhere.

Thanks a lot for your help.

Javier

Hi Mapleprimes,

We know that '' rsolve '' is a recurrence equation solver.  It is more than an expression simplifier.

Congratulations to the Maple computer algebra team for creating such a great computer tool.  simply want to know more.

rsolve_on_May_16_2017.pdf

Surely there are many steps to determine the values to place.

Regards,

Matt

I have an important question regarding the symbolic toolbox from maple for matlab

I have declared three symbolic variables:

syms t;

x = sym('x(t)', 'real');

y = sym('y(t)', 'real');

z = sym('z(t)', 'real'); 

f = x + y + z;

Now i want to  calculate the derivation of f regarding the variable t:

df = diff (f, t);

The result should be 

df = dx(t) / dt + dy(t) / dt + dz(t) / dt

but  instead i get the result df = 0. If i use the command diff (f) i get the result

df = dx(t) / dt 

Does anyone know how to fix this problem?

Hi Guys,

got my last Problem solved :) Now i do have a new one...

restart; with(RealDomain); with(CodeGeneration); with(ExcelTools); with(plots);
dx1:=133;
dy1:=132;

n := 1; 
for i to 256 do 
for j to 256 do 
r := evalf(sqrt((i-dx1)^2+(j-dy1)^2)); 
Ints := R0[i, j]; 
IntsR[n] := [r, Ints]; 
n := n+1 
end do; 
end do; 
IntsR := [seq(IntsR[i], i = 1 .. n-1)]

The list IntsR consists of unsorted values doublets. Now it would be nice to get some sort of mean value of my Ints over r.  Didnt really find a solution for it until now... do you have any hint?

I saw a question earlier today regarding 3 particles in a gravitational field.  It has since disappeared. 

I see no rhyme or reason why it was deleted by someone, whoever did.  It was Math related and it was Maple related hence no reason for it's deletion.  This happened just before Mapleprimes and Maplesoft went offline for a couple of hours. 

Can someone explain?  Did anyone else see it?

If there is a list

[[1,2],[2,2],[3,3]...

how to use correlate function?

assume [1,2] and [2,1] count as 2 

when find correlation between 1and 2

and how about 

if [1,2] and [2,1] are different

first position as key

means 1 relate with 2 count as 1 case for 1 only

[2,1] count as 1 case for 2 only

If input a group of outputs which are binary numbers

can it simplify to give a logic that can output these outputs