# Finding Hamiltonian from a perturbative Action

I am given a perturbative action $$\frac{S}{\mathcal{T}}=\int dt\sum _{n=0,1} (\dot{c_n}{}^2-c_n^2 \omega _n^2)+7.11 c_0^3+35.3 c_0 c_1^2+4.66 c_0 \dot{c_0}{}^2+1.32 c_0 \dot{c_1}{}^2-7.57 \dot{c_0} c_1 \dot{c_1}$$ ($$\omega _0^2=-1.4$$ and $$\omega _1^2=7.57$$).


\[Omega]sq[0] = -1.4; \[Omega]sq[1] = 7.57;
lagrangian =
Sum[c[n]'[t]^2 - c[n][t]^2 \[Omega]sq[n], {n, {0, 1}}] +
7.11 c[0][t]^3 + 35.3 c[0][t] c[1][t]^2 +
4.66 c[0][t] c[0]'[t]^2 + 1.32 c[0][t] c[1]'[t]^2 -
7.57 c[0]'[t] c[1][t] c[1]'[t];



I tried finding the hamiltonian by hand and found out it to be equal to:


H[t_] = c[0]'[t]^2 + c[1]'[t]^2 + 4.66 c[0][t] c[0]'[t]^2 +
1.32 c[0][t] c[1]'[t]^2 - 7.57 c[0]'[t] c[1][t] c[1]'[t] -
1.40 c[0][t]^2 + 7.57 c[1][t]^2 - 7.11 c[0][t]^3 -
35.3 c[0][t] c[1][t]^2;


How shall I use Mathematica to verify the above result?

• (1) Is this a Mathematica question? (2) If so please include code and define all symbols (3) Do you seek to codify an existing algorithm or are you searching for a new algorithm? If so then can this question can be asked more suitably at the Math SE?
– Syed
Commented Apr 26, 2022 at 14:05
• I don't understand what you mean by easily. Isn't the Lagrangian the integrand of the action? Hence, a sum of derivatives w.r.t to the dotted coordinates of the iintegrand times themselves minus the integrand? @Syed the code is there. Is it possible you missed the edit?
– bmf
Commented Apr 26, 2022 at 14:05
• @bmf yes basically I wish to find the hamiltonian from the lagrangian. Since I am working with a complicated problem where the numerical values keep changing, I am trying to find a suitable code to work around the problem, instead of doing it repetitively on a notebook. I have checked out this question: mathematica.stackexchange.com/questions/113017/… but don't understand how to implement it in my problem. Commented Apr 26, 2022 at 14:15

The procedure is basically what you would do by hand, $$\mathcal{H} = \pi \dot{\phi} - \mathcal{L}$$ So just define the momentum $$\pi$$ as usual

momentum[n_] := D[lagrangian, c[n]'[t]]


and then

hamiltonian = Sum[momentum[n] c[n]'[t], {n, {0, 1}}] - lagrangian


gives the result you found by hand (if you Expand). If you need a symbol with a passable t, you can add

h[t_] = hamiltonian

• Thank you for the explanation! The results match perfectly :) Commented Apr 26, 2022 at 15:03