[feature] Create conversion_functions.py for functions that can be used in several contexts

For now it contains conversion from cartesian to spherical coordinates. Takes numpy arrays as input, of dimensions (3,) or (N, 3) and automatically adapts.

MODULES/ENVIRONMENT/light.py

@@ -12,6 +12,7 @@ import pvlib.solarposition as pvlibSP
 import matplotlib.pyplot as plt
 import os
 
+import MODULES.conversion_functions as cf
 
 
 def fibonacci_half_sphere(samples=18):

MODULES/conversion_functions.py

+import numpy as np
+
+def cart_to_sph(x, y, z):
+    """
+
+            Args:
+                cart_coords:
+
+            Returns:
+                zenith_angle: zenith angle [rad]
+            """
+    r = np.sqrt(x**2 + y**2 + z**2)
+
+    if z > 0:
+        zenith = np.arctan((x**2+y**2)/z)
+    elif z < 0:
+        raise NotImplementedError('Not done yet')
+    elif (z == 0) and (np.sqrt(x**2+y**2) != 0):
+        zenith = np.pi/2
+
+    elev = np.pi/2 - zenith
+
+
+    if x > 0:
+        az = np.arctan(y/x)
+    elif (x < 0) and (y >= 0):
+        az = np.arctan(y/x) + np.pi
+    elif (x < 0) and (y < 0):
+        az = np.arctan(y/x) - np.pi
+    elif (x==0) and (y > 0):
+        az = np.pi/2
+    elif (x==0) and (y < 0):
+        az = -np.pi/2
+    elif (x==0) and (y==0):
+        az = np.nan
+
+    return az, zenith
+
+def get_zenith_angle_from_cart(cart_coords):
+    """
+
+    Args:
+        cart_coords: cartesian coordinates (x, y and z) to convert to azimuth, zenith
+        cart_coords type: Numpy array of shape (N, 3)
+
+    Returns:
+        azimuth_angle: zenith angle [rad]
+        zenith_angle: zenith angle [rad]
+    """
+    cart_to_sph_vect = np.vectorize(cart_to_sph)
+
+    x_vect = np.atleast_2d(cart_coords)[:, 0]
+    y_vect = np.atleast_2d(cart_coords)[:, 1]
+    z_vect = np.atleast_2d(cart_coords)[:, 2]
+
+    azimuth, zenith_angle = cart_to_sph_vect(x_vect, y_vect, z_vect)
+
+    return azimuth, zenith_angle