nm

11438 Reputation

20 Badges

13 years, 74 days

MaplePrimes Activity


These are questions asked by nm

Question deleted since it tagged duplicate. Will go search for the duplicate. I did not know there was duplicate one.

update

Here is original question. If moderator thinks it is duplicate feel free to delete. 

I would like to duplicate this simplification done in Mathematica, but in Maple. Mathematica will cancel the exponential term automatically if told the domain is real, but in Maple it willl not.

My attempts in Maple which all fail

restart;


ode := diff(v(x), x, x)*exp(x^2) = 0;
simplify(ode);
simplify(ode) assuming real; #there is no such type
simplify(ode) assuming x::positive;
simplify(ode,symbolic);
use RealDomain in simplify(ode) end use;

How to cancel the exponential term from the above equation in Maple?

Any one knows if this is a new bug in int()? It happens only when kernelopts('assertlevel'=2): is on.

Using Maple 2023.2.1 on windows 10.

edit: Found another integral. #3 below. The difference now is that this third integral is solved completely when removing Physics update/lib from libname. While the first two are not solved. But all three now do not give exception with the updated libname. New attachment is below.

ps. just in case I also just send bug report to Maplesoft.  

restart;

201864

interface(version)

`Standard Worksheet Interface, Maple 2023.2, Windows 10, November 24 2023 Build ID 1762575`

Physics:-Version()

`The "Physics Updates" version in the MapleCloud is 1641. The version installed in this computer is 1637 created 2023, November 29, 17:28 hours Pacific Time, found in the directory C:\Users\Owner\maple\toolbox\2023\Physics Updates\lib\`

kernelopts('assertlevel'=2):

integrand:=(-b*x+a)^(4/3)*(b*x+a)^(8/3);
int(integrand,x,method=_RETURNVERBOSE);

(-b*x+a)^(4/3)*(b*x+a)^(8/3)

Error, (in IntegrationTools:-Indefinite:-ExpandAndMapOverSums) assertion failed, Invalid input for ExpandAndMapOverSums

integrand:=(-b*x+a)^(4/3)*(b*x+a)^(4/3);
int(integrand,x,method=_RETURNVERBOSE)

(-b*x+a)^(4/3)*(b*x+a)^(4/3)

Error, (in IntegrationTools:-Indefinite:-ExpandAndMapOverSums) assertion failed, Invalid input for ExpandAndMapOverSums

integrand:=(-b*x^2+a)^(3/2)*(b*x^2+a)^(3/2);
int(integrand,x,method=_RETURNVERBOSE)

(-b*x^2+a)^(3/2)*(b*x^2+a)^(3/2)

Error, (in IntegrationTools:-Indefinite:-ExpandAndMapOverSums) assertion failed, Invalid input for ExpandAndMapOverSums

libname

"C:\Users\Owner\maple\toolbox\2023\Physics Updates\lib", "C:\Program Files\Maple 2023\lib"

restart;

201864

libname:="C:/Program Files/Maple 2023/lib"

"C:/Program Files/Maple 2023/lib"

kernelopts('assertlevel'=2):
integrand:=(-b*x+a)^(4/3)*(b*x+a)^(8/3);
int(integrand,x,method=_RETURNVERBOSE);

(-b*x+a)^(4/3)*(b*x+a)^(8/3)

["risch" = -(1/1215)*(243*b^4*x^4+324*a*b^3*x^3-450*a^2*b^2*x^2-804*a^3*b*x+47*a^4)*(-b*x+a)^(1/3)*(b*x+a)^(2/3)*((-b*x+a)^2)^(1/3)/(b*((b*x-a)^2)^(1/3))+(int((256/729)*a^5/((b*x-a)^2*(b*x+a))^(1/3), x))*((-b*x+a)^2)^(1/3)*((b*x-a)^2*(b*x+a))^(1/3)/((-b*x+a)^(2/3)*((b*x-a)^2)^(1/3)*(b*x+a)^(1/3)), FAILS = ("gosper", "lookup", "derivativedivides", "default", "norman", "trager", "meijerg", "elliptic", "pseudoelliptic", "parallelrisch", "parts")]

integrand:=(-b*x+a)^(4/3)*(b*x+a)^(4/3);
int(integrand,x,method=_RETURNVERBOSE)

(-b*x+a)^(4/3)*(b*x+a)^(4/3)

["risch" = (3/55)*x*(-5*b^2*x^2+13*a^2)*(-b*x+a)^(1/3)*(b*x+a)^(1/3)*((-b*x+a)^2)^(1/3)/((b*x-a)^2)^(1/3)+(int((16/55)*a^4/((b*x-a)^2*(b*x+a)^2)^(1/3), x))*((-b*x+a)^2)^(1/3)*((b*x-a)^2*(b*x+a)^2)^(1/3)/((-b*x+a)^(2/3)*(b*x+a)^(2/3)*((b*x-a)^2)^(1/3)), FAILS = ("gosper", "lookup", "derivativedivides", "default", "norman", "trager", "meijerg", "elliptic", "pseudoelliptic", "parallelrisch", "parts")]

integrand:=(-b*x^2+a)^(3/2)*(b*x^2+a)^(3/2);
int(integrand,x,method=_RETURNVERBOSE)

(-b*x^2+a)^(3/2)*(b*x^2+a)^(3/2)

["default" = (1/7)*(-b*x^2+a)^(1/2)*(b*x^2+a)^(1/2)*((b/a)^(1/2)*b^4*x^9-4*(b/a)^(1/2)*a^2*b^2*x^5+4*a^4*((-b*x^2+a)/a)^(1/2)*((b*x^2+a)/a)^(1/2)*EllipticF(x*(b/a)^(1/2), I)+3*(b/a)^(1/2)*a^4*x)/((-b^2*x^4+a^2)*(b/a)^(1/2)), "risch" = (1/7)*x*(-b^2*x^4+3*a^2)*(-b*x^2+a)^(1/2)*(b*x^2+a)^(1/2)+(4/7)*a^4*(1-b*x^2/a)^(1/2)*(1+b*x^2/a)^(1/2)*EllipticF(x*(b/a)^(1/2), I)*((-b*x^2+a)*(b*x^2+a))^(1/2)/((b/a)^(1/2)*(-b^2*x^4+a^2)^(1/2)*(-b*x^2+a)^(1/2)*(b*x^2+a)^(1/2)), "elliptic" = (-b*x^2+a)^(1/2)*(b*x^2+a)^(1/2)*(-(1/7)*b^2*x^5*(-b^2*x^4+a^2)^(1/2)+(3/7)*a^2*x*(-b^2*x^4+a^2)^(1/2)+(4/7)*a^4*(1-b*x^2/a)^(1/2)*(1+b*x^2/a)^(1/2)*EllipticF(x*(b/a)^(1/2), I)/((b/a)^(1/2)*(-b^2*x^4+a^2)^(1/2)))/(-b^2*x^4+a^2)^(1/2), FAILS = ("gosper", "lookup", "derivativedivides", "norman", "trager", "meijerg", "pseudoelliptic", "parallelrisch", "parts")]

Download jan_13_2024_integrationTools_expand.mw

I've reported this to Maplesoft 6 months ago.

I was wondering if someone with beta version of 2024 could check if these are fixed? (if one is allowed to do so). As these errors keep breaking my program. (not possible to trap).

436

interface(version);

`Standard Worksheet Interface, Maple 2023.2, Windows 10, November 24 2023 Build ID 1762575`

Physics:-Version();

`The "Physics Updates" version in the MapleCloud is 1637 and is the same as the version installed in this computer, created 2023, November 29, 17:28 hours Pacific Time.`

ode:=diff(y(x),x) = (x*y(x)+x^3+x*y(x)^2+y(x)^3)/x^2;
sol:=exp(3*sum(1/(9*_R^2-1)*ln((-_R*x+y(x)-1/3*x)/x),_R = RootOf(27*_Z^3-9*_Z+29)))-c__1*exp(x) = 0;
odetest(sol,ode);

diff(y(x), x) = (x*y(x)+x^3+x*y(x)^2+y(x)^3)/x^2

exp(3*(sum(ln((-_R*x+y(x)-(1/3)*x)/x)/(9*_R^2-1), _R = RootOf(27*_Z^3-9*_Z+29))))-c__1*exp(x) = 0

Error, (in simplify/RootOf) too many levels of recursion

ode:=diff(u(x),x)-1/2*(2*a*u(x)^3+u(x)+2*b)/x = 0;
sol:=2*sum(1/(6*_R^2*a+1)*ln(u(x)-_R),_R = RootOf(2*_Z^3*a+_Z+2*b))-1/2*ln(x)-_C1 = 0;
odetest(sol,ode);

diff(u(x), x)-(1/2)*(2*a*u(x)^3+u(x)+2*b)/x = 0

2*(sum(ln(u(x)-_R)/(6*_R^2*a+1), _R = RootOf(2*_Z^3*a+_Z+2*b)))-(1/2)*ln(x)-_C1 = 0

Error, (in simplify/RootOf) too many levels of recursion

 

Download in_simplify_rootof_too_many_level_of_recursion_jan_6_2024.mw

Maple dsolve allows one to specify the algorithm to use to solve the ode. But sometimes it is very tricky to figure the syntax,

This ode 

ode:=diff(y(x),x)*y(x)+a*x*y(x)+b*x^3=0;
DEtools:-odeadvisor(ode);

Gives

               [[_homogeneous, `class G`], _rational, [_Abel, `2nd type`, `class A`]]

I wanted now to call dsolve telling dsolve to use the first method above. But how? All the following syntax failed for me

sol:=dsolve(ode,['_homogeneous, `class G`']);
sol:=dsolve(ode,'[_homogeneous, `class G`]');

All return method not found .

I am sure I am using wrong syntax but do not know what the correct one should be.

infolevel[dsolve]:=5;
sol:=dsolve(ode);

gives

Methods for first order ODEs:
--- Trying classification methods ---
trying a quadrature
trying 1st order linear
trying Bernoulli
trying separable
trying inverse linear
trying homogeneous types:
trying homogeneous G
<- homogeneous successful

With long solution printed now OK. 

When using just '[homogeneous]' it works

sol:=dsolve(ode,'[homogeneous]');

It gives same solution as default case.

What is the correct syntax to tell dsolve to use specific method [_homogeneous, `class G`] ? i.e. I need to add class G

The reason I ask is becuase Maple have different kind of homogeneous method as described here

Maple 2023.2.1 on windows 10

Given equation 

We see this can be simplified to 

In Mathematica, I just need to tell it the denominator of the left side is not zero for it to do the simplification

The same thing in Maple did not work:

eq:= (y-2*x)^3/( (y-x)^2 * x ) = a/x;
the_denom:= denom(lhs(eq));
simplify(eq) assuming the_denom <>0

No change. 

I am doing this in code not by looking at the screen and simply wanted to eliminate common terms on both sides of equation. I can in code obtain the denominator of the LHS and RHS and add assumption. But since I do not know what if any common terms are on both sides, it is not easy to use elminate.

I see code at https://www.mapleprimes.com/questions/227043-How-To-Automatically-Cancel-Any-Common which actually worked on this and it did automatically elminate x from both sides.

But my question is: why Maple does not do it using simplify with the assumption given?  Tried simplify with size and no change.

I am looking for the simplist method to elminate common terms on both sides of equation. Is the link above the only way to do this in Maple? May be there is something simpler in recent Maple versions?


#code from https://www.mapleprimes.com/questions/227043-How-To-Automatically-Cancel-Any-Common

restart;

eq:= (y-2*x)^3/( (y-x)^2*x) = a/x;
TT := simplify(expand((eq)),size):
if lhs(TT)::`*` and rhs(TT)::`*` then
  TTT := map[2](map,freeze,TT);
  comm := `*`(op({op(lhs(TTT))} intersect {op(rhs(TTT))}));
  new := simplify(thaw(lhs(TTT)/comm=rhs(TTT)/comm));
end if:
new

May be this should be part of Maple build in functions and used by simplify? 

Update 

I just hit on a way to do this automatically with no assumption of anything. The idea is to simply rewrite the equation, like this

restart;

eq:= (y-2*x)^3/( (y-x)^2 * x ) = a/x;
numer(lhs(eq))*denom(rhs(eq)) /  (denom(lhs(eq)) *numer(rhs(eq)))=1;

You see, the common term on both sides is automatically gone!   I need to test this more. 

To moderator: if you think this question is duplicate, feel free to delete it.

First 40 41 42 43 44 45 46 Last Page 42 of 201