Canonic bosons.py
From Werner KRAUTH
(Difference between revisions)
| Revision as of 00:17, 9 December 2016 Werner (Talk | contribs) ← Previous diff |
Current revision Werner (Talk | contribs) |
||
| Line 1: | Line 1: | ||
| - | This is the program Canonic_bosons.py, that is useful for Homework 13 | + | This is the python2 program Canonic_bosons.py useful for the homework session |
| - | + | of [[ICFP_Stat_Physics_2018|week 13 of my ICFP Lectures on statistical mechanics]]. | |
| + | Here, the density of state is 1,3,6,10,15, which corresponds to the three-dimensional isotropic harmonic trap. In the below program, we integrate in the complex plane from (-pi, lambda) to (pi, lambda) (neglecting the integral from (-pi, 0) to (-pi, lambda)), and always get the same result for the partition function. At T=1, for example, we get Z=17.373..., a result we can also obtain from the naive sum over the states, as implemented in Naive_bosons.py. | ||
| import math, cmath, numpy, pylab | import math, cmath, numpy, pylab | ||
| - | N0vec = [] | ||
| dos = [1, 3, 6, 10, 15] | dos = [1, 3, 6, 10, 15] | ||
| T = 1.0 | T = 1.0 | ||
| beta = 1.0 / T | beta = 1.0 / T | ||
| Zint = complex(0.0, 0.0) | Zint = complex(0.0, 0.0) | ||
| - | eps = 0.001 | ||
| dellambda = 0.01 | dellambda = 0.01 | ||
| oldl = complex(-math.pi, dellambda) | oldl = complex(-math.pi, dellambda) | ||
| complexi = complex(0.0, 1.0) | complexi = complex(0.0, 1.0) | ||
| for RL in numpy.arange(-math.pi, math.pi, 0.00001): | for RL in numpy.arange(-math.pi, math.pi, 0.00001): | ||
| - | newl = RL + eps * complexi | + | newl = RL + dellambda * complexi |
| integrand = cmath.exp(-complexi * 5 * newl) | integrand = cmath.exp(-complexi * 5 * newl) | ||
| for E in range(5): | for E in range(5): | ||
Current revision
This is the python2 program Canonic_bosons.py useful for the homework session of week 13 of my ICFP Lectures on statistical mechanics. Here, the density of state is 1,3,6,10,15, which corresponds to the three-dimensional isotropic harmonic trap. In the below program, we integrate in the complex plane from (-pi, lambda) to (pi, lambda) (neglecting the integral from (-pi, 0) to (-pi, lambda)), and always get the same result for the partition function. At T=1, for example, we get Z=17.373..., a result we can also obtain from the naive sum over the states, as implemented in Naive_bosons.py.
import math, cmath, numpy, pylab
dos = [1, 3, 6, 10, 15]
T = 1.0
beta = 1.0 / T
Zint = complex(0.0, 0.0)
dellambda = 0.01
oldl = complex(-math.pi, dellambda)
complexi = complex(0.0, 1.0)
for RL in numpy.arange(-math.pi, math.pi, 0.00001):
newl = RL + dellambda * complexi
integrand = cmath.exp(-complexi * 5 * newl)
for E in range(5):
integrand /= (1.0 - cmath.exp( - beta * E + complexi * newl)) ** dos[E]
Zint += integrand * (newl - oldl) / (2.0 * math.pi)
oldl = newl
print Zint
