Equationsolving with Mathematica
in [Math]
|
Prev: proper analysis of Euclid's Infinitude of Primes rather than the botched job of Hardy/Woodgold and Mathematical Intelligencer #609 Correcting Math
Next: video
From: Jan W. Coenen on 28 Jun 2010 08:22 Hi guys, id like to find out how to find out a numerical solution to the following equation with mathematica: qpar/2\[LongEqual]1.96763*10^-8 T^4+(2.65932*10^10 E^(-99104.3 (1/ T-1/5823)))/Sqrt[T] It seems i am not able to fine a function or values T(qpar) Jan
From: Alois Steindl on 28 Jun 2010 08:46 "Jan W. Coenen" <jan.w.coenen(a)googlemail.com> writes: > Hi guys, > > id like to find out how to find out a numerical solution to the > following equation with mathematica: > > qpar/2\[LongEqual]1.96763*10^-8 T^4+(2.65932*10^10 E^(-99104.3 (1/ > T-1/5823)))/Sqrt[T] > > It seems i am not able to fine a function or values T(qpar) > > Hello, to obtain reasonable results, you should in any case scale your variables and equations. Since you have huge entries in your equation, it is likely that any numerical method will fail otherwise. (I don't know how skilled you are in doing such kind of preparation, but in your case it seems indispensible!) In any case, if you try to solve any equation of kind f(T) = q you can plot f(T) first, to learn about possible solutions and guess good starting points. Alois
From: Axel Vogt on 28 Jun 2010 08:55 Jan W. Coenen wrote: > Hi guys, > > id like to find out how to find out a numerical solution to the > following equation with mathematica: > > qpar/2\[LongEqual]1.96763*10^-8 T^4+(2.65932*10^10 E^(-99104.3 (1/ > T-1/5823)))/Sqrt[T] > > It seems i am not able to fine a function or values T(qpar) > > > Jan You may wish to write your question in ASCII format in a readable line and ask a concrete question, for me it is not understandable that way.
From: Robert Israel on 28 Jun 2010 13:25 > Hi guys, > > id like to find out how to find out a numerical solution to the > following equation with mathematica: > > qpar/2\[LongEqual]1.96763*10^-8 T^4+(2.65932*10^10 E^(-99104.3 (1/ > T-1/5823)))/Sqrt[T] > > It seems i am not able to fine a function or values T(qpar) > > > Jan I don't know about Mathematica, but Maple has little trouble with solving it numerically. Of course, you can't expect a symbolic solution. > eq:= qpar/2 = 1.96763*10^(-8)*T^4 + (2.65932*10^10 *exp(-99104.3* (1/ T-1/5823)))/sqrt(T); For example, with qpar=1: > fsolve(eval(eq, qpar=1)); 70.99972101 -- Robert Israel israel(a)math.MyUniversitysInitials.ca Department of Mathematics http://www.math.ubc.ca/~israel University of British Columbia Vancouver, BC, Canada
From: Rob Johnson on 28 Jun 2010 21:31
In article <e12316bc-8381-4045-9270-df3618e7d8e3(a)t10g2000yqg.googlegroups.com>, "Jan W. Coenen" <jan.w.coenen(a)googlemail.com> wrote: >id like to find out how to find out a numerical solution to the >following equation with mathematica: > >qpar/2\[LongEqual]1.96763*10^-8 T^4+(2.65932*10^10 E^(-99104.3 (1/ >T-1/5823)))/Sqrt[T] > >It seems i am not able to fine a function or values T(qpar) I extended the precision in your formula to 20 places to get a bit more precision in the answer. If you know more precision to the constants in your formula, use it. If not, remove the "`20"s I added to your constants. qpar[T_] := 2 (1.96763`20*10^-8 T^4 + (2.65932`20*10^10 E^(-99104.3`20 (1/T - 1/5823)))/Sqrt[T]) iqpar[x_] := FindRoot[qpar[T] == x, {T, 1}, WorkingPrecision -> 20] Rob Johnson <rob(a)trash.whim.org> take out the trash before replying to view any ASCII art, display article in a monospaced font |