Merge pull request #145 from AngelFP/bug/fix_envelope_phase

Implement 1D unwrapping of laser envelope phase

wake_t/physics_models/laser/envelope_solver.py

@@ -11,6 +11,7 @@
 
 from wake_t.utilities.numba import njit_serial
 from .tdma import TDMA
+from .utils import unwrap
 
 
 @njit_serial(fastmath=True)
@@ -91,7 +92,7 @@ def evolve_envelope(
 
         # Getting the phase of the envelope on axis.
         if use_phase:
-            phases = np.angle(a[:, 0])
+            phases = unwrap(np.angle(a[:, 0]))
 
         # Loop over z.
         for j in range(nz - 1, -1, -1):
@@ -101,16 +102,6 @@ def evolve_envelope(
                 d_theta1 = phases[j + 1] - phases[j]
                 d_theta2 = phases[j + 2] - phases[j + 1]
 
-                # Prevent phase jumps bigger than 1.5*pi.
-                if d_theta1 < -1.5 * np.pi:
-                    d_theta1 += 2 * np.pi
-                if d_theta2 < -1.5 * np.pi:
-                    d_theta2 += 2 * np.pi
-                if d_theta1 > 1.5 * np.pi:
-                    d_theta1 -= 2 * np.pi
-                if d_theta2 > 1.5 * np.pi:
-                    d_theta2 -= 2 * np.pi
-
             # Calculate D factor [Eq. (6)].
             D_jkn = (1.5 * d_theta1 - 0.5 * d_theta2) * inv_dz
 

wake_t/physics_models/laser/envelope_solver_non_centered.py

@@ -11,6 +11,7 @@
 
 from wake_t.utilities.numba import njit_serial
 from .tdma import TDMA
+from .utils import unwrap
 
 
 @njit_serial(fastmath=True)
@@ -91,7 +92,7 @@ def evolve_envelope_non_centered(
 
         # Getting the phase of the envelope on axis.
         if use_phase:
-            phases = np.angle(a[:, 0])
+            phases = unwrap(np.angle(a[:, 0]))
 
         # Loop over z.
         for j in range(nz - 1, -1, -1):
@@ -101,16 +102,6 @@ def evolve_envelope_non_centered(
                 d_theta1 = phases[j + 1] - phases[j]
                 d_theta2 = phases[j + 2] - phases[j + 1]
 
-                # Prevent phase jumps bigger than 1.5*pi.
-                if d_theta1 < -1.5 * np.pi:
-                    d_theta1 += 2 * np.pi
-                if d_theta2 < -1.5 * np.pi:
-                    d_theta2 += 2 * np.pi
-                if d_theta1 > 1.5 * np.pi:
-                    d_theta1 -= 2 * np.pi
-                if d_theta2 > 1.5 * np.pi:
-                    d_theta2 -= 2 * np.pi
-
             # Calculate D factor [Eq. (6)].
             D_jkn = (1.5 * d_theta1 - 0.5 * d_theta2) * inv_dz
 

wake_t/physics_models/laser/utils.py

@@ -0,0 +1,59 @@
+"""Utilities for the laser envelope solver."""
+import numpy as np
+from wake_t.utilities.numba import njit_serial
+
+
+@njit_serial
+def unwrap(p, discont=None, axis=-1, period=6.283185307179586):
+    """Numba version of numpy.unwrap.
+
+    The implementation is taken from
+    https://github.com/numba/numba/blob/main/numba/np/arraymath.py,
+    which currently is not yet included in the latest Numba release.
+    """
+    if axis != -1:
+        msg = 'Value for argument "axis" is not supported'
+        raise ValueError(msg)
+    # Flatten to a 2D array, keeping axis -1
+    p_init = np.asarray(p).astype(np.float64)
+    init_shape = p_init.shape
+    last_axis = init_shape[-1]
+    p_new = p_init.reshape((p_init.size // last_axis, last_axis))
+    # Manipulate discont and period
+    if discont is None:
+        discont = period / 2
+    interval_high = period / 2
+    boundary_ambiguous = True
+    interval_low = -interval_high
+
+    slice1 = (slice(1, None, None),)
+
+    # Work on each row separately
+    for i in range(p_init.size // last_axis):
+        row = p_new[i]
+        dd = np.diff(row)
+        ddmod = np.mod(dd - interval_low, period) + interval_low
+        if boundary_ambiguous:
+            ddmod = np.where(
+                (ddmod == interval_low) & (dd > 0),
+                interval_high,
+                ddmod
+            )
+        ph_correct = ddmod - dd
+
+        ph_correct = np.where(
+            np.array([abs(x) for x in dd]) < discont,
+            0,
+            ph_correct
+        )
+        ph_ravel = np.where(
+            np.array([abs(x) for x in dd]) < discont,
+            0,
+            ph_correct
+        )
+        ph_correct = np.reshape(ph_ravel, ph_correct.shape)
+        up = np.copy(row)
+        up[slice1] = row[slice1] + ph_correct.cumsum()
+        p_new[i] = up
+
+    return p_new.reshape(init_shape)