Implementation of CW Temporal Profile and Plane Wave Transverse Profile #373

rob-shalloo · 2025-03-12T16:32:54Z

This PR aims to address the problem raised #371. Ideally one would like to be able to perform fast ~1D or ~2D manipulations of the temporal/spectral profile or the spatial profile respectively. One could approach this by adding more functionality to the respective profiles, but this would bring with it significant overhead. After discussion with @MaxThevenet we thought a more efficient solution would be to keep the laser object as is (together with all the excellent functionality that entails) and instead generate profiles which mimic the effect of a plane wave (in the event that one wants to manipulate just the transverse profile) or a continuous wave laser (in the event that one is interested in manipulating just the transverse profile).

In this PR, you will find implementations of both a Plane Wave Transverse Profile and a Continuous Wave Longitudinal Profile, together with some modifications to the grid object to ensure that smooth operation.

Sample usage for Plane Wave

from lasy.laser import Laser
from lasy.profiles.longitudinal import GaussianLongitudinalProfile
from lasy.profiles.transverse import PlaneWaveProfile
from lasy.profiles import CombinedLongitudinalTransverseProfile


# Physical Parameters
tau = 30e-15
wavelength = 800e-9
t_peak = 0.0
pol = (1, 0)
peak_power = 1e12  # W

long_prof = GaussianLongitudinalProfile(wavelength, tau, t_peak)
tran_prof = PlaneWaveProfile()
profile = CombinedLongitudinalTransverseProfile(
        wavelength, pol, long_prof, tran_prof, peak_power=peak_power
    )

# Computational Grid
dim = "xyt"
lo = (None, None, -5 * tau)
hi = (None, None, 5 * tau)
npoints = (1, 1, 1000)

laser = Laser(dim, lo, hi, npoints, profile)

Here you can see the lo and hi variables have None in the x an y elements. I believe this would be more intuitive for a user than selecting a random set of low and high points. Grid has been adapted to automatically set the range when None is provided. Of course the user could set the lo and hi variables equal to the area of the beam to naturally calculate the correct peak field. However, as can be seen in this example, one can simple overwrite this if needed.

Sample usage for Continuous Wave

from lasy.laser import Laser
from lasy.profiles.transverse import SuperGaussianTransverseProfile
from lasy.profiles.longitudinal import ContinuousWaveProfile
from lasy.profiles import CombinedLongitudinalTransverseProfile


# Physical Parameters
peak_intensity = 1.0e10  # W/m^2
spot_size = 10e-3
wavelength = 800e-9
pol = (1, 0)

long_prof = ContinuousWaveProfile(wavelength)
tran_prof = GaussianTransverseProfile(spot_size)
profile = CombinedLongitudinalTransverseProfile(
        wavelength, pol, long_prof, tran_prof, peak_intensity=peak_intensity
    )

# Computational Grid
dim = "xyt"
lo = (-5 * spot_size, -5 * spot_size, None)
hi = (5 * spot_size, 5 * spot_size, None)
npoints = (1000, 1000, 1)

laser = Laser(dim, lo, hi, npoints, profile)

for more information, see https://pre-commit.ci

rob-shalloo · 2025-03-12T16:51:00Z

Should add peak_power, peak_intensity to the LaserProfile as alternative options to energy. In the case that one of these special profiles are selected (cw or plane wave), then one option should be required.

We should change any meaningless number, eg dx in grid to be np.nan in the case of a plane wave. We should carefully think about how to handle lo and hi here.

for more information, see https://pre-commit.ci

lasy/profiles/combined_profile.py

lasy/profiles/longitudinal/continuous_wave_profile.py

lasy/profiles/transverse/plane_wave_profile.py

lasy/profiles/combined_profile.py

for more information, see https://pre-commit.ci

…o cwplanewave

for more information, see https://pre-commit.ci

…o cwplanewave

for more information, see https://pre-commit.ci

rob-shalloo · 2025-03-12T22:11:49Z

Had another think about the use of np.nan in place of upper and lower bounds of the grid. I suspect this will break quite a lot of things (eg the laser.show functionality). For this reason, I think its not too bad to have the laser for CW profile defined on a range of -1,1 seconds and for the Plane Wave pulses to be defined on a grid of -1,1 m for xyt and 0,1 m for rt. The dx in this case is still meaningful, in that it is correct for the grid.

for more information, see https://pre-commit.ci

MaxThevenet

Looks great, thanks a lot for this PR! See minor comments below. In addition, as a follow-up of our offline discussion yesterday, I would raise a few points:

Currently, the amplitude of the laser pulse is typically specified with the energy. For symmetry, could we consider the following?
- full profile: energy (J);
- CW wave: power (J/s), as the longitudinal dimension is infinite; -> already the case in this PR.
- plane wave: fluence (J/m^2) as the transverse dimensions are infinite. -> currently intensity is used.
Likewise, for symmetry and consistency, if we have the 3 input parameters, it would be great to have the corresponding diagnostics, i.e., get_laser_energy (already there), get_peak_power (you added it), get_peak_fluence.
Should the time domain be 1-second long [-0.5, 0.5] in CW case, such that, numerically, energy [J] == power [J/s]? Likewise for transverse box size of [-0.5, 0.5] meter in plane wave case such that, numerically, energy[J] == fluency [J/m^2]?
I can envision that, before running the full expensive 3D beamline people check quick CW, then quick plane wave, then full thing. Do you think this is a reasonable workflow? If so, I recommend we make it simpler such that the user only needs to change either the longitudinal profile or the transverse profile. For this to work, we may need to overwrite some values (e.g. enforce 1 cell in transverse direction for plane wave, similar for lo and hi, etc.). What do you think?

docs/source/tutorials/1d_temporal_laser.ipynb

MaxThevenet · 2025-03-13T06:18:19Z

docs/source/tutorials/1d_temporal_laser.ipynb

+   "outputs": [],
+   "source": [
+    "tau0_meas = get_duration(laser.grid, laser.dim)\n",
+    "energy0_meas = compute_laser_energy(laser.dim, laser.grid)\n",


This may be slightly misleading, as the energy contained in a plane wave is infinite. Should we consider a compute_laser_fluence there?

MaxThevenet · 2025-03-13T06:21:01Z

docs/source/tutorials/1d_temporal_laser.ipynb

+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }


I think there's an empty cell at the end that can be removed.

MaxThevenet · 2025-03-13T06:27:48Z

lasy/profiles/longitudinal/continuous_wave_profile.py

+            Contains the value of the longitudinal envelope at the
+            specified points. This array has the same shape as the array t.
+        """
+        return np.ones_like(t + 0 * 1j)


Suggested change

return np.ones_like(t + 0 * 1j)

return np.ones_like(t + 0j)

MaxThevenet · 2025-03-13T06:30:48Z

lasy/utils/grid.py

+            lo[-1] = -1.0
+            hi[-1] = 1.0


Having a box length of 1 in SI (here 1 second) could come in handy to connect energy and power (i.e., the value of the energy in the box in J = the power in J/s).

Suggested change

lo[-1] = -1.0

hi[-1] = 1.0

lo[-1] = -0.5

hi[-1] = 0.5

Same can be done below for plane wave in transverse position.

MaxThevenet · 2025-03-13T06:32:37Z

tests/test_special_laser.py

+This test file verifys the correct implementation of these special cases of the laser object and
+additionally checks the implementation of the peak_intensity and peak_power normalizations
+as well as some measurement functionality from laser utils


Suggested change

This test file verifys the correct implementation of these special cases of the laser object and

additionally checks the implementation of the peak_intensity and peak_power normalizations

as well as some measurement functionality from laser utils

Check the implementation of these profiles and peak_intensity and peak_power normalizations, as well as some measurement functionality from laser utils.

…o cwplanewave

for more information, see https://pre-commit.ci

…o cwplanewave

for more information, see https://pre-commit.ci

rob-shalloo · 2025-03-14T08:00:59Z

All sounds good, these changes should be in place now

MaxThevenet

Thanks for this PR! See minor suggestions below. Just one more change: could you update the logic such that the user can also specify peak_power and peak_fluence in all cases to combined_profile and that the profile is normalized accordingly?

MaxThevenet · 2025-03-19T18:30:37Z

docs/source/tutorials/1d_temporal_laser.ipynb

+    "plt.ylim(\n",
+    "    0,\n",
+    ")\n",


Suggested change

"plt.ylim(\n",

" 0,\n",

")\n",

"plt.ylim(0, None)\n",

MaxThevenet · 2025-03-19T18:31:58Z

docs/source/tutorials/1d_temporal_laser.ipynb

+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# Defining and Manipulating a Laser Pulse in Time Only\n",


Suggested change

"# Defining and Manipulating a Laser Pulse in Time Only\n",

"# Laser Pulse in Time Only\n",

MaxThevenet · 2025-03-19T18:32:43Z

docs/source/tutorials/2d_spatial_laser.ipynb

+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "# Defining and Manipulating a Laser Pulse in Space Only\n",


Suggested change

"# Defining and Manipulating a Laser Pulse in Space Only\n",

"# Laser Pulse in Space Only\n",

MaxThevenet · 2025-03-19T18:35:48Z

docs/source/tutorials/1d_temporal_laser.ipynb

+    "print(\"Final Pulse Duration   = %.2f fs\" % (tau1_meas * 1e15))\n",
+    "print(\"\\n\")\n",


Suggested change

"print(\"Final Pulse Duration = %.2f fs\" % (tau1_meas * 1e15))\n",

"print(\"\\n\")\n",

"print(\"Final Pulse Duration = %.2f fs\" % (tau1_meas * 1e15) + "\n")\n",

MaxThevenet · 2025-03-19T18:36:18Z

docs/source/tutorials/2d_spatial_laser.ipynb

+    "print(\"Expected Final Spot Size = %.2f mm\" % (w1_meas * 1e3))\n",
+    "print(\"\\n\")\n",


Suggested change

"print(\"Expected Final Spot Size = %.2f mm\" % (w1_meas * 1e3))\n",

"print(\"\\n\")\n",

"print(\"Expected Final Spot Size = %.2f mm\" % (w1_meas * 1e3) + "\n")\n",

MaxThevenet · 2025-03-19T20:27:43Z

lasy/utils/laser_utils.py

+        field = grid.get_temporal_field()
+        power = get_laser_power(dim, grid)
+        input_peak_power = power.max()
+        if input_peak_power == 0.0:
+            print("Field is zero everywhere, normalization will be skipped")
+        else:
+            field *= np.sqrt(peak_power / input_peak_power)
+            grid.set_temporal_field(field)


Suggested change

field = grid.get_temporal_field()

power = get_laser_power(dim, grid)

input_peak_power = power.max()

if input_peak_power == 0.0:

print("Field is zero everywhere, normalization will be skipped")

else:

field *= np.sqrt(peak_power / input_peak_power)

grid.set_temporal_field(field)

power = get_laser_power(dim, grid)

input_peak_power = power.max()

if input_peak_power == 0.0:

print("Field is zero everywhere, normalization will be skipped")

else:

field = grid.get_temporal_field()

grid.set_temporal_field(field * np.sqrt(peak_power / input_peak_power))

MaxThevenet · 2025-03-19T20:29:21Z

tests/test_special_laser.py

+This test file verifys the correct implementation of these special cases of the laser object and
+additionally checks the implementation of the peak_fluence and peak_power normalizations
+as well as some measurement functionality from laser utils


Suggested change

This test file verifys the correct implementation of these special cases of the laser object and

additionally checks the implementation of the peak_fluence and peak_power normalizations

as well as some measurement functionality from laser utils

This test file verifies the correct implementation of these cases of the laser object and

checks the implementation of the peak_fluence and peak_power normalizations

as well as some measurement functionality from laser utils

MaxThevenet · 2025-03-19T20:35:00Z

lasy/profiles/combined_profile.py

+        The total energy of the laser pulse. The amplitude of the laser
+        field (:math:`E_0` in the above formula) is automatically
+        calculated so that the pulse has the prescribed energy.
+        Only used in case where laser is neither a plane wave or a continuous wave laser


This is still the general case so I would not make a specific case here.

Suggested change

Only used in case where laser is neither a plane wave or a continuous wave laser

MaxThevenet · 2025-03-19T20:36:08Z

lasy/profiles/combined_profile.py

+        The peak fluence of the laser pulse. The amplitude of the laser
+        field (:math:`E_0` in the above formula) is automatically
+        calculated so that the pulse has the prescribed peak fluence.
+        Only used in case where laser is a continuous wave laser


I think that is a good option in general and should be usable.

Suggested change

Only used in case where laser is a continuous wave laser

Required for a CW (monochromatic) profile.

MaxThevenet · 2025-03-19T20:36:34Z

lasy/profiles/combined_profile.py

+        The peak power of the laser pulse. The amplitude of the laser
+        field (:math:`E_0` in the above formula) is automatically
+        calculated so that the pulse has the prescribed peak power.
+        Only used in case where laser is a plane wave laser


Suggested change

Only used in case where laser is a plane wave laser

Required for a plane wave profile.

rob-shalloo and others added 2 commits March 12, 2025 17:21

cw and plane wave implementations

193a7d9

[pre-commit.ci] auto fixes from pre-commit.com hooks

03c7a6a

for more information, see https://pre-commit.ci

rob-shalloo and others added 2 commits March 12, 2025 18:39

adding more normalizations and knowledge of laser type

322e98d

[pre-commit.ci] auto fixes from pre-commit.com hooks

ba99d71

for more information, see https://pre-commit.ci

github-advanced-security bot found potential problems Mar 12, 2025

View reviewed changes

adhere to PEP8

c433104

github-advanced-security bot found potential problems Mar 12, 2025

View reviewed changes

lasy/profiles/combined_profile.py Fixed Show fixed Hide fixed

lasy/profiles/combined_profile.py Fixed Show fixed Hide fixed

rob-shalloo and others added 22 commits March 12, 2025 20:37

fix CI tests

a970dfc

[pre-commit.ci] auto fixes from pre-commit.com hooks

c58708c

for more information, see https://pre-commit.ci

fix tutorials

eea5a80

Merge branch 'cwplanewave' of https://github.com/rob-shalloo/lasy int…

ce0022b

…o cwplanewave

[pre-commit.ci] auto fixes from pre-commit.com hooks

55ed2c6

for more information, see https://pre-commit.ci

codeQL fix

b5f9daf

[pre-commit.ci] auto fixes from pre-commit.com hooks

1d94263

for more information, see https://pre-commit.ci

codeQL fix

6c7e758

Merge branch 'cwplanewave' of https://github.com/rob-shalloo/lasy int…

704d0dd

…o cwplanewave

[pre-commit.ci] auto fixes from pre-commit.com hooks

0ae62a5

for more information, see https://pre-commit.ci

fix

8bc582c

fix

3556cea

added CI tests

2129dc5

[pre-commit.ci] auto fixes from pre-commit.com hooks

5d963d8

for more information, see https://pre-commit.ci

fix docs

d75c1c2

add tutorial for plane wave

7e5a83d

[pre-commit.ci] auto fixes from pre-commit.com hooks

2844d76

for more information, see https://pre-commit.ci

add tutorial for continuous wave

bf09754

fix merge

d157b90

[pre-commit.ci] auto fixes from pre-commit.com hooks

492e6f6

for more information, see https://pre-commit.ci

update

b0d97ad

bring cw and plane wave booleans into grid

58be96b

[pre-commit.ci] auto fixes from pre-commit.com hooks

d095ca0

for more information, see https://pre-commit.ci

rob-shalloo changed the title ~~[WIP] Implementation of CW Temporal Profile and Plane Wave Transverse Profile~~ Implementation of CW Temporal Profile and Plane Wave Transverse Profile Mar 12, 2025

rob-shalloo and others added 4 commits March 13, 2025 08:42

update to peak_fluence for cw beams

e3f38be

[pre-commit.ci] auto fixes from pre-commit.com hooks

1c10fa4

for more information, see https://pre-commit.ci

more sensible axes limit choices

28549b0

[pre-commit.ci] auto fixes from pre-commit.com hooks

ab8c9e0

for more information, see https://pre-commit.ci

MaxThevenet reviewed Mar 13, 2025

View reviewed changes

rob-shalloo and others added 12 commits March 13, 2025 08:59

cleanup

4b86e3e

Merge branch 'cwplanewave' of https://github.com/rob-shalloo/lasy int…

788af4a

…o cwplanewave

[pre-commit.ci] auto fixes from pre-commit.com hooks

66f3c92

for more information, see https://pre-commit.ci

fix npoints for cw and plane wave

e83c3bb

Merge branch 'cwplanewave' of https://github.com/rob-shalloo/lasy int…

b9ac61a

…o cwplanewave

update docs

584fe9d

[pre-commit.ci] auto fixes from pre-commit.com hooks

b033bf7

for more information, see https://pre-commit.ci

fix

bdd4a09

typos

7b69258

[pre-commit.ci] auto fixes from pre-commit.com hooks

2fa3d98

for more information, see https://pre-commit.ci

formatting

873422c

[pre-commit.ci] auto fixes from pre-commit.com hooks

d29737e

for more information, see https://pre-commit.ci

MaxThevenet reviewed Mar 19, 2025

View reviewed changes

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Implementation of CW Temporal Profile and Plane Wave Transverse Profile #373

Implementation of CW Temporal Profile and Plane Wave Transverse Profile #373

rob-shalloo commented Mar 12, 2025 •

edited

Loading

rob-shalloo commented Mar 12, 2025

rob-shalloo commented Mar 12, 2025

MaxThevenet left a comment

MaxThevenet Mar 13, 2025

MaxThevenet Mar 13, 2025

MaxThevenet Mar 13, 2025

MaxThevenet Mar 13, 2025

MaxThevenet Mar 13, 2025

rob-shalloo commented Mar 14, 2025

MaxThevenet left a comment

MaxThevenet Mar 19, 2025

MaxThevenet Mar 19, 2025

MaxThevenet Mar 19, 2025

MaxThevenet Mar 19, 2025

MaxThevenet Mar 19, 2025

MaxThevenet Mar 19, 2025

MaxThevenet Mar 19, 2025

MaxThevenet Mar 19, 2025

MaxThevenet Mar 19, 2025

MaxThevenet Mar 19, 2025

	"# Defining and Manipulating a Laser Pulse in Time Only\n",
	"# Laser Pulse in Time Only\n",

	"# Defining and Manipulating a Laser Pulse in Space Only\n",
	"# Laser Pulse in Space Only\n",

		"print(\"Final Pulse Duration = %.2f fs\" % (tau1_meas * 1e15))\n",
		"print(\"\\n\")\n",

		"print(\"Expected Final Spot Size = %.2f mm\" % (w1_meas * 1e3))\n",
		"print(\"\\n\")\n",

	Only used in case where laser is a continuous wave laser
	Required for a CW (monochromatic) profile.

	Only used in case where laser is a plane wave laser
	Required for a plane wave profile.

Implementation of CW Temporal Profile and Plane Wave Transverse Profile #373

Are you sure you want to change the base?

Implementation of CW Temporal Profile and Plane Wave Transverse Profile #373

Conversation

rob-shalloo commented Mar 12, 2025 • edited Loading

rob-shalloo commented Mar 12, 2025

rob-shalloo commented Mar 12, 2025

MaxThevenet left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

rob-shalloo commented Mar 14, 2025

MaxThevenet left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

rob-shalloo commented Mar 12, 2025 •

edited

Loading