Calculating Spin-Loss of a Particle Using Integral Form

Click For Summary
SUMMARY

The discussion focuses on calculating the spin-loss of a particle using an integral form that involves complex functions and numerical integration in Python. The integral is defined as $$\frac{s^2\nu^4}{(2\pi)^2}\int_{-1}^1 u(1-u^2)k_f^5[...]\, du$$, where various functions such as ##r_1, \chi_1, \chi_2## are defined in terms of the parameter u. The user encounters an error in the implementation of the function r1(u), specifically in the line involving complex multiplication and function calls. The discussion emphasizes debugging techniques and the importance of isolating the error through simplification.

PREREQUISITES
  • Understanding of complex functions in Python
  • Familiarity with numerical integration using SciPy
  • Knowledge of Python programming, particularly with NumPy and Matplotlib
  • Basic concepts of particle physics and spin-loss calculations
NEXT STEPS
  • Debugging Python code to identify errors in complex function implementations
  • Learn about numerical integration techniques in SciPy, specifically quad
  • Explore complex number handling in Python, focusing on cmath module
  • Study particle physics concepts related to spin-loss and integral forms
USEFUL FOR

Researchers in particle physics, Python developers working on scientific computing, and students studying numerical methods in physics will benefit from this discussion.

amjad-sh
Messages
240
Reaction score
13
TL;DR
I'm trying to integrate an imaginary function by using the package quad.
The function is a complex function and also a very complicated one.
This typing error is showing up to me after running the program: """unsupported operand type(s) for *: 'complex' and 'function' """".
I will show you the details below.
The integral has the form:
$$\frac{s^2\nu^4}{(2\pi)^2}\int_{-1}^1 u(1-u^2)k_f^5[|r_1\chi_1|^2+|r_1\chi_2|^2-|r_1|^2\chi_1^*\chi_2\cos(2k_f\sqrt{u^2-\nu^2}a)-|r_1|^2\chi_2^*\chi_1cos(2k_f\sqrt{u^2-\nu^2}a)]\, du$$
##r_1,\chi_1## and ##\chi_2## are also imaginary functions of u, because the form of these imaginary functions is so complicated, I also represented them interms of another imaginary functions like a,b,c,m1,m2 and f.
All the functions ##r_1,r_2,\chi_1,\chi_2,b,a## and ##c## are functions of u.
So I defined them as functions with one parameter u as below:

import numpy as np
from scipy.integrate import quad
import scipy
import matplotlib.pyplot as plt
import cmath as math

spinloss=[]
lnu=[]
a = 1.6*10**(-6)
nu = -0.01 \\defining some constants
kf = 4.2*1.6*10**(-19)

for i in range(0,101):
spinloss.append([])
for i in range(0,101):
lnu.append([])

def b(u):
q=u**2-nu**2
if q<0: \\defining the function b
q = -q
v = complex(0, math.sqrt(q))
else:
v = math.sqrt(u**2-nu**2)
return(kf*(u+v)*math.exp(-2j*kf*v*a)+kf*(u-v)*math.exp(2j*kf*v*a))def c(u):
q = u ** 2 - nu ** 2
if q < 0:
q = -q
v = complex(0, math.sqrt(q)) \\defining the function c
else:
v = math.sqrt(u ** 2 - nu ** 2)
return (kf**2*(u+v)**2*np.exp(-2j*kf*v*a)-kf**2*(u-v)**2*math.exp(2j*kf*v*a))

def a(u):
q = u ** 2 - nu ** 2
if q < 0:\\defining the function a
q = -q
v = complex(0, math.sqrt(q))
else:
v = math.sqrt(u ** 2 - nu ** 2)
return (((kf*b(u)*(u-v)+c(u))*math.exp(1j*kf*v*a)-2*kf**2*u*(u+v)*math.exp(-1j*kf*v*a))/((kf*b(u)*(u+v)+c(u))*math.exp(-1j*kf*v*a)-2*kf**2*u*(u-v)*math.exp(1j*kf*v*a)))def x1(u):
\\defining x1
return((a(u)*b(u)+2*kf*u)/c(u))

def x2(u):
\\defining x2
return((2*a(u)*kf*u+b(u))/c(u))def m1(u):
q = u ** 2 - nu ** 2
if q < 0:
q = -q
v = complex(0, math.sqrt(q)) \\defining m1
else:
v = math.sqrt(u ** 2 - nu ** 2)
return(kf*v*(1-a(u)))def m2(u):
q = u ** 2 - nu ** 2
if q < 0:
q = -q \\defining m2
v = complex(0, math.sqrt(q))
else:
v = math.sqrt(u ** 2 - nu ** 2)
return((-kf*v*(1-a(u))))def f(u):
q = u ** 2 - nu ** 2
if q < 0:
q = -q
v = complex(0, math.sqrt(q)) \\defining f
else:
v = math.sqrt(u ** 2 - nu ** 2)
return((a(u)+1)*math.exp(-1j*math.sqrt(kf*a))*math.cos(kf*v*a))def r1(u):
q = u ** 2 - nu ** 2
if q < 0:
q = -q \\defining r1
v = complex(0, math.sqrt(q))
else:
v = math.sqrt(u ** 2 - nu ** 2)
return((2*math.exp(-2j*kf*u*a))/(2*f(u)-((1/(kf*u))*(m1(u)*math.exp(1j*(kf*v-kf*u)*a)-m2(u)*math.exp(-1j*(kf*v+kf*u)*a))))) \\the error started from this linedef r0(u):
q = u ** 2 - nu ** 2
if q < 0:
q = -q \\defining r0
v = complex(0, math.sqrt(q))
else:
v = math.sqrt(u ** 2 - nu ** 2)
return(math.exp(-2j*kf*u*a)-(r1(u)/(kf*u))*(2*kf*v*math.exp(-1j*kf*(v+u)*a)-(2*kf*v*math.exp(1j*kf*(v-u)*a))))def t0(u):
q = u ** 2 - nu ** 2
if q < 0:
q = -q \\defining t0
v = complex(0, math.sqrt(q))
else:
v = math.sqrt(u ** 2 - nu ** 2)
return(((r1(u))/(u))*v*a(u)*(math.exp(-1j*kf*(v+u)*a)-math.exp(1j*kf*(v-u)*a)))
def ls(u):
q = u ** 2 - nu ** 2
if q < 0: \\defining ls
q = -q
v = complex(0, math.sqrt(q))
else:
v = math.sqrt(u ** 2 - nu ** 2)
return(u*(1-u**2)*kf**5*((abs(r1(u)*x1(u)))**2+(abs(r1(u)*x2(u)))**2-(abs(r1(u)))**2)*np.conjugate(x1(u))*x2(u)*math.cos(2*kf*v*a)-(abs(r1(u)))**2*np.conjugate(x2(u))*x1(u)*math.cos(2*kf*v*a))\\ it showed also here.(the error)
def jx(u):
return((((-kf)**2)/(8*(math.pi)**2))*u*(abs(r0(u)))**2) \\defining jx
def complex_quadrature(func, a, b):
def real_func(u):
return scipy.real(func(u))
\\for integration
real_integral = quad(real_func, a, b)

return (real_integral[0])

for i in range(0, 101):
nu = nu + 0.01
if nu == 1: \\ nu represents ##\nu## and it must be between zero and 1.
spinloss = 0
lnu = nu

if nu != 1:
spinloss = ((complex_quadrature(ls, -1, 1)) * (nu ** 4) /(2*math.pi)**2) / (complex_quadrature(jx, -1, 1)) \\ this gives us the result of the integration of ls from -1 to 1 divided by the integration of jx from -1 to 1.
lnu = nu
plt.plot(lnu,spinloss,color='b')
plt.show()

If some python expert can point out for me the reason why this error is appearing, I will be so appreciated.
 
Technology news on Phys.org
It would help to know which multiplication exactly causes the error. The debugger might show that, alternatively simplify this equation step by step until you find a minimal expression that causes the error:
amjad-sh said:
return((2*math.exp(-2j*kf*u*a))/(2*f(u)-((1/(kf*u))*(m1(u)*math.exp(1j*(kf*v-kf*u)*a)-m2(u)*math.exp(-1j*(kf*v+kf*u)*a))))) \\the error started from this line
 
  • Like
Likes   Reactions: amjad-sh

Similar threads

  • · Replies 8 ·
Replies
8
Views
2K
  • · Replies 50 ·
2
Replies
50
Views
6K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 6 ·
Replies
6
Views
2K
  • · Replies 2 ·
Replies
2
Views
3K
  • · Replies 5 ·
Replies
5
Views
3K
Replies
0
Views
2K
  • · Replies 3 ·
Replies
3
Views
4K
  • · Replies 1 ·
Replies
1
Views
3K
  • · Replies 3 ·
Replies
3
Views
2K